ECCV-2024-Papers

官网链接：https://eccv.ecva.net/

主会 :bell:：9 月 29 日（周日）至 10 月 4 日

历年综述论文分类汇总戳这里↘️CV-Surveys施工中~~~~~~~~~~

2025 年论文分类汇总戳这里

↘️WACV-2025-Papers ↘️CVPR-2025-Papers

2024 年论文分类汇总戳这里

↘️WACV-2024-Papers ↘️CVPR-2024-Papers ↘️ECCV-2024-Papers

2022 年论文分类汇总戳这里

2022 年论文分类汇总戳这里

2021 年论文分类汇总戳这里

2020 年论文分类汇总戳这里

💥💥💥全部论文已分类完毕

:thumbsup:ECCV 2024奖项公布，哥大摘得最佳论文奖桂冠

🏆Best Paper Award(最佳论文奖)

• Minimalist Vision with Freeform Pixels
  :house:project

🏅Best Paper Honorable Mention(最佳论文荣誉提名奖)

• Rasterized Edge Gradients: Handling Discontinuities Differentiably
• Concept Arithmetics for Circumventing Concept Inhibition in Diffusion Models
  :house:project

目录

:cat:	:dog:	:tiger:	:wolf:
1.Other(其它)	2.3D Visual	3.Face(人脸)	4.Pose(姿态估计)
5.OCR	6.Object Tracking(目标跟踪)	7.Object Detection(目标检测)	8.Super-Resolution(超分辨率)
9.Image Progress(图像/视频处理)	10.Image Classification(图像分类)	11.Image Segmentation(图像分割)	12.Image Retrieval(图像检索)
13.Image/video Compression(图像/视频压缩)	14.Image Captioning(图像/视频字幕)	15.GAN/Image Synthesis(图像生成)	16.Medical Image Progress(医学影响处理)
17.Video	18.Automated Driving(自动驾驶)	19.UAV/Remote Sensing/Satellite Image(无人机/遥感/卫星图像)	20.Scene
21.Vision-Language(视觉语言)	22.Few/Zero-Shot Learning/DG/A(小/零样本/域泛化/域适应)	23.Machine Learning(机器学习)	24.Vision Transformer
25.MC/KD/Pruning(模型压缩/知识蒸馏/剪枝)	26.NAS	27.GNN/GCN	28.Novel Class Discovery(新类发现)
29.Semi/self-supervised learning(半/自监督)	30.Anomaly Detection(异常检测)	31.Point Clouds(点云)	32.Person Re-Identification(人员重识别)
33.Motion Generation(人体运动生成)	34.Visual Question Answering(视觉问答)	35.Action Detection(动作检测)	36.Gaze Estimation
37.Style Transfer(风格迁移)	38.Human-Object Interaction(人机交互)	39.Robots(机器人)	40.Object Pose Estimation(物体姿态估计)
41.Biomedical(生物特征识别)	42.Optical Flow Estimation(光流估计)	43.Sound	44.Dataset/Benchmark(数据集/基准)
45.Neural Radiance Fields	46.Rendering(渲染)	47.Animal	48.Computer Graphics(计算机图形学)
49.Light-Field(光场)	50.Sketches(草图)	51.Feature Matching	52.Visual Entity Recognition(视觉实体识别)
53.Keypoint Detection(关键点检测)	54.Deepfake Detection	55.Information Security(信息安全)	56.Dense Prediction(密集预测)
57.Visual Relationship Detection(视觉关系检测)	58.全家桶

58.全家桶

• X-InstructBLIP: A Framework for Aligning Image, 3D, Audio, Video to LLMs and its Emergent Cross-modal Reasoning
  :star:code

57.Visual Relationship Detection(视觉关系检测)

• Visual Relationship Transformation
• Scene-Graph ViT: End-to-End Open-Vocabulary Visual Relationship Detection

56.Dense Prediction(密集预测)

• Chameleon: A Data-Efficient Generalist for Dense Visual Prediction in the Wild(https://github.com/GitGyun/chameleon)密集视觉预测
• Unsupervised Dense Prediction using Differentiable Normalized Cuts
• Three Things We Need to Know About Transferring Stable Diffusion to Visual Dense Prediciton Tasks
• Removing Rows and Columns of Tokens in Vision Transformer enables Faster Dense Prediction without Retraining
  :star:code

55.Information Security(信息安全)

• 版权保护
  • Rethinking Data Bias: Dataset Copyright Protection via Embedding Class-wise Hidden Bias
    :star:code保护数据集版权
• 图像水印
  • Certifiably Robust Image Watermark
    :star:code
  • A Secure Image Watermarking Framework with Statistical Guarantees via Adversarial Attacks on Secret Key Networks图像水印
  • Not Just Change the Labels, Learn the Features: Watermarking Deep Neural Networks with Multi-View Data
    :star:code
  • A Watermark-Conditioned Diffusion Model for IP Protection
    :star:code
  • A Geometric Distortion Immunized Deep Watermarking Framework with Robustness Generalizability
  • LaWa: Using Latent Space for In-Generation Image Watermarking

54.Deepfake Detection

• Real Appearance Modeling for More General Deepfake Detection
• Contrasting Deepfakes Diffusion via Contrastive Learning and Global-Local Similarities
  :star:code
• Fake It till You Make It: Curricular Dynamic Forgery Augmentations towards General Deepfake Detection
• Common Sense Reasoning for Deep Fake Detection
  :star:code
• 图像伪造检测和定位
  • Noise-assisted Prompt Learning for Image Forgery Detection and Localization
  • AdaIFL: Adaptive Image Forgery Localization via a Dynamic and Importance-aware Transformer Network
    :star:code
• 文档图像篡改检测
  • Enhancing Tampered Text Detection through Frequency Feature Fusion and Decomposition
    :thumbsup:文档图像篡改检测 (DITD) 方法——特征融合与分解网络 (FFDN)
• 合成图像检测
  • Leveraging Representations from Intermediate Encoder-blocks for Synthetic Image Detection
    :star:code

53.Keypoint Detection(关键点检测)

• OpenKD: Opening Prompt Diversity for Zero- and Few-shot Keypoint Detection
  :star:code
• KeypointDETR: An End-to-End 3D Keypoint Detector
  :star:code

52.Visual Entity Recognition(视觉实体识别)

• Grounding Language Models for Visual Entity Recognition视觉实体识别

51.Feature Matching

• Raising the Ceiling: Conflict-Free Local Feature Matching with Dynamic View Switching
• 图像匹配
  • CriSp: Leveraging Tread Depth Maps for Enhanced Crime-Scene Shoeprint Matching
    :star:code

50.Sketches(草图)

• Do Generalised Classifiers really work on Human Drawn Sketches?

49.Light-Field(光场)

• Deep Polarization Cues for Single-shot Shape and Subsurface Scattering Estimation
• 相机重定位
  • Differentiable Product Quantization for Memory Efficient Camera Relocalization

48.Computer Graphics(计算机图形学)

• 高动态范围成像
  • SAFNet: Selective Alignment Fusion Network for Efficient HDR Imaging
    :star:code

47.Animal

• Animal Avatars: Reconstructing Animatable 3D Animals from Casual Videos
  :house:project
• Ponymation: Learning Articulated 3D Animal Motions from Unlabeled Online Videos
  :house:project3D动物运动
• Adaptive High-Frequency Transformer for Diverse Wildlife Re-Identification
  :star:code

46.Rendering(渲染)

• City-on-Web: Real-time Neural Rendering of Large-scale Scenes on the Web
  :star:code
  :house:project
• A Probability-guided Sampler for Neural Implicit Surface Rendering
  :house:project渲染
• TextDiffuser-2: Unleashing the Power of Language Models for Text Rendering
  :house:project
• AnyLens: A Generative Diffusion Model with Any Rendering Lens(https://anylens-diffusion.github.io/)
• CityGaussian: Real-time High-quality Large-Scale Scene Rendering with Gaussians
  :star:code
  :house:project
• METACAP: Meta-learning Priors from Multi-View Imagery for Sparse-view Human Performance Capture and Rendering
  :house:project
• GAURA: Generalizable Approach for Unified Restoration and Rendering of Arbitrary Views
• MaRINeR: Enhancing Novel Views by Matching Rendered Images with Nearby References
  :star:code
• Learning Unsigned Distance Functions from Multi-view Images with Volume Rendering Priors
  :star:code
• CaesarNeRF: Calibrated Semantic Representation for Few-Shot Generalizable Neural Rendering
  :house:project
• IntrinsicAnything: Learning Diffusion Priors for Inverse Rendering Under Unknown Illumination
  :star:code渲染
• Photorealistic Object Insertion with Diffusion-Guided Inverse Rendering
  :house:project
• VersatileGaussian: Real-time Neural Rendering for Versatile Tasks using Gaussian Splatting神经渲染
• UniVoxel: Fast Inverse Rendering by Unified Voxelization of Scene Representation
  :star:code
• Photorealistic Object Insertion with Diffusion-Guided Inverse Rendering
  :house:project
• GeoGaussian: Geometry-aware Gaussian Splatting for Scene Rendering
  :star:code场景渲染
• GMT: Enhancing Generalizable Neural Rendering via Geometry-Driven Multi-Reference Texture Transfer
  :star:code
• Boost Your NeRF: A Model-Agnostic Mixture of Experts Framework for High Quality and Efficient Rendering

45.Neural Radiance Fields

• Invertible Neural Warp for NeRF
  :star:code
• VF-NeRF: Viewshed Fields for Rigid NeRF Registration
• NeRF-XL: NeRF at Any Scale with Multi-GPU
  :house:project
• Regularizing Dynamic Radiance Fields with Kinematic Fields
• KFD-NeRF: Rethinking Dynamic NeRF with Kalman Filter
  :star:code
• Dynamic Neural Radiance Field From Defocused Monocular Video
• Flash Cache: Reducing Bias in Radiance Cache Based Inverse Rendering
  :house:project
• Protecting NeRFs' Copyright via Plug-And-Play Watermarking Base Model
  :house:project
• GeometrySticker: Enabling Ownership Claim of Recolorized Neural Radiance Fields
  :star:code
  :house:project
• Efficient NeRF Optimization - Not All Samples Remain Equally Hard
• MeshFeat: Multi-Resolution Features for Neural Fields on Meshes
  :house:project
• DecentNeRFs: Decentralized Neural Radiance Fields from Crowdsourced Images
  :house:project
• TrackNeRF: Bundle Adjusting NeRF from Sparse and Noisy Views via Feature Tracks
  :star:code
• BeNeRF: Neural Radiance Fields from a Single Blurry Image and Event Stream
  :star:code
• TriNeRFLet: A Wavelet Based Multiscale Triplane NeRF Representation
  :house:project
• RS-NeRF: Neural Radiance Fields from Rolling Shutter Images
  :star:code
• Motion-Oriented Compositional Neural Radiance Fields for Monocular Dynamic Human Modeling
  :star:code
  :house:project
• RaFE: Generative Radiance Fields Restoration
  :house:project
• Few-shot NeRF by Adaptive Rendering Loss Regularization
  :star:code
• Depth-guided NeRF Training via Earth Mover’s Distance
• DatasetNeRF: Efficient 3D-aware Data Factory with Generative Radiance Fields
  :star:code
• Flowed Time of Flight Radiance Fields
• Volumetric Rendering with Baked Quadrature Fields
• BeNeRF:Neural Radiance Fields from a Single Blurry Image and Event Stream
  :star:code
• Taming Latent Diffusion Model for Neural Radiance Field Inpainting
  :house:project
• Mesh2NeRF: Direct Mesh Supervision for Neural Radiance Field Representation and Generation
  :house:project
  🤗huggingface
• SlotLifter: Slot-guided Feature Lifting for Learning Object-Centric Radiance Fields
  :house:project
• FisherRF: Active View Selection and Mapping with Radiance Fields using Fisher Information
  :star:code
• DMiT: Deformable Mipmapped Tri-Plane Representation for Dynamic ScenesNeRF
• Single-Mask Inpainting for Voxel-based Neural Radiance Fields
• Content-Aware Radiance Fields: Aligning Model Complexity with Scene Intricacy Through Learned Bitwidth Quantization
  :star:code
• Gaussian Frosting: Editable Complex Radiance Fields with Real-Time Rendering
  :house:project
• Physically Plausible Color Correction for Neural Radiance Fields
• Leveraging Thermal Modality to Enhance Reconstruction in Low-Light ConditionsNeRF
• PointNeRF++: A multi-scale, point-based Neural Radiance Field
  :house:project
• Omni-Recon: Harnessing Image-based Rendering for General-Purpose Neural Radiance Fields
• High-Fidelity and Transferable NeRF Editing by Frequency Decomposition
  :house:project
• TriNeRFLet: A Wavelet Based Triplane NeRF Representation
  :house:project
• Diffusion-Generated Pseudo-Observations for High-Quality Sparse-View Reconstruction
  :house:project
• G2fR: Frequency Regularization in Grid-based Feature Encoding Neural Radiance Fields
• NeRF-MAE: Masked AutoEncoders for Self-Supervised 3D Representation Learning for Neural Radiance Fields
  :house:project
• 新视图合成
  • Fast View Synthesis of Casual Videos
    :house:project
  • PolyOculus: Simultaneous Multi-view Image-based Novel View Synthesis
    :house:project
  • RING-NeRF : Rethinking Inductive Biases for Versatile and Efficient Neural Fields
  • Structured-NeRF: Hierarchical Scene Graph with Neural Representation
  • URS-NeRF: Unordered Rolling Shutter Bundle Adjustment for Neural Radiance Fields
  • A Compact Dynamic 3D Gaussian Representation for Real-Time Dynamic View Synthesis
    :star:code
    :house:project
  • High-Resolution and Few-shot View Synthesis from Asymmetric Dual-lens Inputs
    :star:code
  • Distractor-Free Novel View Synthesis via Exploiting Memorization Effect in Optimization
    :star:code
  • NVS-Adapter: Plug-and-Play Novel View Synthesis from a Single Image
    :star:code
  • FSGS: Real-Time Few-shot View Synthesis using Gaussian Splatting
    :star:code
  • Fast View Synthesis of Casual Videos with Soup-of-Planes
    :house:project
  • CoherentGS: Sparse Novel View Synthesis with Coherent 3D Gaussians
    :house:project
  • MegaScenes: Scene-Level View Synthesis at Scale
    :star:code
  • Radiative Gaussian Splatting for Efficient X-ray Novel View Synthesis
    :star:code视图合成
  • NGP-RT: Fusing Multi-Level Hash Features with Lightweight Attention for Real-Time Novel View Synthesis
  • Efficient Depth-Guided Urban View Synthesis
    :star:code
  • Generative Camera Dolly: Extreme Monocular Dynamic Novel View Synthesis
    :star:code
  • Generalizable Human Gaussians for Sparse View Synthesis
    :house:project
  • Thermal3D-GS: Physics-induced 3D Gaussians for Thermal Infrared Novel-view Synthesis
    :star:code

44.Dataset/Benchmark(数据集/基准)

• FYI: Flip Your Images for Dataset Distillation
• Neural Spectral Decomposition for Dataset Distillation
  :star:code
• Teddy: Efficient Large-Scale Dataset Distillation via Taylor-Approximated Matching
  :star:code
• Distill Gold from Massive Ores: Bi-level Data Pruning towards Efficient Dataset Distillation
  :star:code
• COM Kitchens: An Unedited Overhead-view Video Dataset as a Vision-Language Benchmark
• 基准
  • MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models
    :star:code
  • DailyDVS-200: A Comprehensive Benchmark Dataset for Event-Based Action Recognition
    :star:code
  • Urban Waterlogging Detection: A Challenging Benchmark and Large-Small Model Co-Adapter
    :star:code
  • MSD: A Benchmark Dataset for Floor Plan Generation of Building Complexes
  • BlinkVision: A Benchmark for Optical Flow, Scene Flow and Point Tracking Estimation using RGB Frames and Eventsbr>:house:project
  • SIMBA: Split Inference - Mechanisms, Benchmarks and Attacks
    :star:code
  • A Multimodal Benchmark Dataset and Model for Crop Disease Diagnosis
    :star:code
  • BAFFLE: A Baseline of Backpropagation-Free Federated Learning
    :star:code
  • Omni6DPose: A Benchmark and Model for Universal 6D Object Pose Estimation and Tracking
  • Affective Visual Dialog: A Large-Scale Benchmark for Emotional Reasoning Based on Visually Grounded Conversations
    :house:project
  • UniIR: Training and Benchmarking Universal Multimodal Information Retrievers
    :house:project
  • HyTAS: A Hyperspectral Image Transformer Architecture Search Benchmark and Analysis
  • OphNet: A Large-Scale Video Benchmark for Ophthalmic Surgical Workflow Understanding
    :house:project
  • PredBench: Benchmarking Spatio-Temporal Prediction across Diverse Disciplines
    :star:code
  • Cross-Platform Video Person ReID: A New Benchmark Dataset and Adaptation Approach
    :star:code
  • R^2-Bench: Benchmarking the Robustness of Referring Perception Models under Perturbations
    :star:code
  • m&m’s: A Benchmark to Evaluate Tool-Use for multi-step multi-modal Tasks
    :star:code
    🤗huggingface
  • PathMMU: A Massive Multimodal Expert-Level Benchmark for Understanding and Reasoning in Pathology
    🤗huggingface
  • LayeredFlow: A Real-World Benchmark for Non-Lambertian Multi-Layer Optical Flow
    :house:project
  • HIMO: A New Benchmark for Full-Body Human Interacting with Multiple Objects
    :star:code
  • When Pedestrian Detection Meets Multi-Modal Learning: Generalist Model and Benchmark Dataset
    :star:code
• 数据集
  • VITATECS: A Diagnostic Dataset for Temporal Concept Understanding of Video-Language Models
    :star:code
  • HaloQuest: A Visual Hallucination Dataset for Advancing Multimodal Reasoning
    :star:code
  • OmniACT: A Dataset and Benchmark for Enabling Multimodal Generalist Autonomous Agents for Desktop and Web
  • COM Kitchens: An Unedited Overhead-view Procedural Videos Dataset a Vision-Language Benchmark
    :sunflower:dataset
  • Seeing Faces in Things: A Model and Dataset for Pareidolia
    :sunflower:dataset
  • Towards Dual Transparent Liquid Level Estimation in Biomedical Lab: Dataset, Methods and Practice
    :sunflower:dataset
  • GarmentCodeData: A Dataset of 3D Made-to-Measure Garments With Sewing Patterns
    :house:project
  • SemTrack: A Large-scale Dataset for Semantic Tracking in the Wild
    :sunflower:dataset
  • WiMANS: A Benchmark Dataset for WiFi-based Multi-user Activity Sensing
    :star:code
  • BugNIST - a Large Volumetric Dataset for Detection under Domain Shift
  • Defect Spectrum: A Granular Look of Large-scale Defect Datasets with Rich Semantics
    :star:code
    :house:project大规模缺陷数据集
  • Raindrop Clarity: A Dual-Focused Dataset for Day and Night Raindrop Removal
    :star:code
  • PartImageNet++ Dataset: Scaling up Part-based Models for Robust Recognition
    :star:code
  • WTS: A Pedestrian-Centric Traffic Video Dataset for Fine-grained Spatial-Temporal Understanding
    :star:code
  • MMVR: Millimeter-wave Multi-View Radar Dataset and Benchmark for Indoor Perception
  • SkyScenes: A Synthetic Dataset for Aerial Scene Understanding
    :house:project
  • Caltech Aerial RGB-Thermal Dataset in the Wild
    :star:code
  • V2X-Real: a Largs-Scale Dataset for Vehicle-to-Everything Cooperative Perception
  • H-V2X: A Large Scale Highway Dataset for BEV Perception
  • PetFace: A Large-Scale Dataset and Benchmark for Animal Identification
    :star:code
  • Long-range Turbulence Mitigation: A Large-scale Dataset and A Coarse-to-fine Framework
  • OmniNOCS: A unified NOCS dataset and model for 3D lifting of 2D objects
    :star:code
  • SignAvatars: A Large-scale 3D Sign Language Holistic Motion Dataset and Benchmark
    :star:code
    :house:project
  • Insect Identification in the Wild: The AMI Dataset
    :star:code野外昆虫识别：AMI 数据集
  • RoScenes: A Large-scale Multi-view 3D Dataset for Roadside Perception
    :sunflower:dataset
• 数据增强
  • SUMix: Mixup with Semantic and Uncertain Information
    :star:code
  • Data Augmentation via Latent Diffusion for Saliency Prediction
  • FreeAugment: Data Augmentation Search Across All Degrees of Freedom
    :star:code
  • Enhancing Recipe Retrieval with Foundation Models: A Data Augmentation Perspective
    :star:code

43.Sound

• Audio-Synchronized Visual Animation
  :star:code
  :house:project
• Listen to Look into the Future: Audio-Visual Egocentric Gaze Anticipation
  :house:project
• Label-anticipated Event Disentanglement for Audio-Visual Video Parsing
• Masked Generative Video-to-Audio Transformers with Enhanced Synchronicity
  :star:code
• Spherical World-Locking for Audio-Visual Localization in Egocentric Videos
• Self-Supervised Audio-Visual Soundscape Stylization
  :house:project
• CAT: Enhancing Multimodal Large Language Model to Answer Questions in Dynamic Audio-Visual Scenarios
  :star:code视听场景
• Perceptual Evaluation of Audio-Visual Synchrony Grounded in Viewers’ Opinion Scores
• Siamese Vision Transformers are Scalable Audio-visual Learners
  :star:code视听学习器
• Action2Sound: Ambient-Aware Generation of Action Sounds from Egocentric Videos
  :house:project生成环境感知的动作声音
• Audio-visual Generalized Zero-shot Learning the Easy Way
• 视听分割
  • Ref-AVS: Refer and Segment Objects in Audio-Visual Scenes
    :star:code
  • Stepping Stones: A Progressive Training Strategy for Audio-Visual Semantic Segmentation
    :star:code
    :star:code
  • CPM: Class-conditional Prompting Machine for Audio-visual Segmentation视听分割

42.Optical Flow Estimation(光流估计)

• SEA-RAFT: Simple, Efficient, Accurate RAFT for Optical Flow
  :star:code

41.Biomedical(生物特征识别)

• Open-Set Biometrics: Beyond Good Closed-Set Models
  :star:code

40.Object Pose Estimation(物体姿态估计)

• SCAPE: A Simple and Strong Category-Agnostic Pose Estimator
  :star:code
• SRPose: Two-view Relative Pose Estimation with Sparse Keypoints
  :house:project
• FAFA: Frequency-Aware Flow-Aided Self-Supervision for Underwater Object Pose Estimation
  :star:code
• A Graph-Based Approach for Category-Agnostic Pose Estimation
  :house:project
• GS-Pose: Category-Level Object Pose Estimation via Geometric and Semantic Correspondence
• OP-Align: Object-level and Part-level Alignment for Self-supervised Category-level Articulated Object Pose Estimation
  :star:code
• FoundPose: Unseen Object Pose Estimation with Foundation Features
  :house:project
• LaPose: Laplacian Mixture Shape Modeling for RGB-Based Category-Level Object Pose Estimation
  :star:code
• U-COPE: Taking a Further Step to Universal 9D Category-level Object Pose Estimation
• PACE: Pose Annotations in Cluttered Environments
  :star:code
• 6-DoF
  • An Economic Framework for 6-DoF Grasp Detection
    :star:code
  • Pseudo-keypoint RKHS Learning for Self-supervised 6DoF Pose Estimation
  • Language-Driven 6-DoF Grasp Detection Using Negative Prompt Guidance
    :star:code
  • Omni6D: Large-Vocabulary 3D Object Dataset for Category-Level 6D Object Pose Estimation
    :star:code
  • 6DGS: 6D Pose Estimation from a Single Image and a 3D Gaussian Splatting Model
    :star:code
  • FreeZe: Training-free zero-shot 6D pose estimation with geometric and vision foundation models
    :house:project
• 相机姿态估计
  • ADen: Adaptive Density Representations for Sparse-view Camera Pose Estimation
  • Correspondences of the Third Kind: Camera Pose Estimation from Object Reflection
• 计数
  • AFreeCA: Annotation-Free Counting for All计数
  • Zero-shot Object Counting with Good Exemplars
  • ABC Easy as 123: A Blind Counter for Exemplar-Free Multi-Class Class-agnostic Counting
    :star:code
    :house:project计数
  • Class-Agnostic Object Counting with Text-to-Image Diffusion Model
  • Shifted Autoencoders for Point Annotation Restoration in Object Counting

39.Robots(机器人)

• See and Think: Embodied Agent in Virtual Environment
  :house:project
• SceneGraphLoc: Cross-Modal Coarse Visual Localization on 3D Scene Graphs
• V-IRL: Grounding Virtual Intelligence in Real Life
  :star:code
• 机器人
  • Robo-ABC: Affordance Generalization Beyond Categories via Semantic Correspondence for Robot Manipulation
    :house:project
  • Learning Cross-hand Policies of High-DOF Reaching and Grasping机器人
  • DISCO: Embodied Navigation and Interaction via Differentiable Scene Semantics and Dual-level Control
    :star:code
  • Real-time Holistic Robot Pose Estimation with Unknown States
    :star:code
  • ManiGaussian: Dynamic Gaussian Splatting for Multi-task Robotic Manipulation
    :star:code
    :house:project
  • Adapt2Reward: Adapting Video-Language Models to Generalizable Robotic Rewards via Failure Prompts
  • GraspXL: Generating Grasping Motions for Diverse Objects at Scale
    :star:code
    :house:project
  • UGG: Unified Generative Grasping
    :house:project机器人
  • Decomposed Vector-Quantized Variational Autoencoder for Human Grasp Generation
    :star:code
  • Track2Act: Predicting Point Tracks from Internet Videos enables Generalizable Robot Manipulation
    :house:project机器人
• 导航
  • NavGPT-2: Unleashing Navigational Reasoning Capability for Large Vision-Language Models
    :star:code
  • Prioritized Semantic Learning for Zero-shot Instance Navigation
    :star:code导航
• VPR
  • Close, But Not There: Boosting Geographic Distance Sensitivity in Visual Place Recognition
    :star:code
  • Navigation Instruction Generation with BEV Perception and Large Language Models
    :star:code
  • Revisit Anything: Visual Place Recognition via Image Segment Retrieval
    :star:code
  • VLAD-BuFF: Burst-aware Fast Feature Aggregation for Visual Place Recognition
    :star:code
  • MeshVPR: Citywide Visual Place Recognition Using 3D Meshes
    :star:code
• SLAM
  • Deep Patch Visual SLAM
    :star:code
  • RGBD GS-ICP SLAM
    :star:code
  • I2-SLAM: Inverting Imaging Process for Robust Photorealistic Dense SLAM
  • Hyperion - A fast, versatile symbolic Gaussian Belief Propagation framework for Continuous-Time SLAM
    :star:code
  • SGS-SLAM: Semantic Gaussian Splatting For Neural Dense SLAM
  • LRSLAM: Low-rank Representation of Signed Distance Fields in Dense Visual SLAM System
  • I$^2$-SLAM: Inverting Imaging Process for Robust Photorealistic Dense SLAM
  • Learn to Memorize and to Forget: A Continual Learning Perspective of Dynamic SLAM
  • Self-Supervised Underwater Caustics Removal and Descattering via Deep Monocular SLAM
  • CG-SLAM: Efficient Dense RGB-D SLAM in a Consistent Uncertainty-aware 3D Gaussian Field
    :star:code
• Try-On
  • Time-Efficient and Identity-Consistent Virtual Try-On Using A Variant of Altered Diffusion Models
  • Improving Virtual Try-On with Garment-focused Diffusion Models
    :star:code
  • Wear-Any-Way: Manipulable Virtual Try-on via Sparse Correspondence Alignment
    :star:code
    :house:project
  • Improving Diffusion Models for Authentic Virtual Try-on in the Wild
    :star:code
  • D4-VTON: Dynamic Semantics Disentangling for Differential Diffusion based Virtual Try-On
    :star:code
  • WildVidFit: Video Virtual Try-On in the Wild via Image-Based Controlled Diffusion Models
    :star:code
• 交叉地理定位
  • GAReT: Cross-view Video Geolocalization with Adapters and Auto-Regressive Transformers
    :star:code
  • Cross-view image geo-localization with Panorama-BEV Co-Retrieval Network
    :star:code
  • ConGeo: Robust Cross-view Geo-localization across Ground View Variations
    :star:code
    :house:project交叉视角地理定位
  • Benchmarking the Robustness of Cross-view Geo-localization Models
  • CityGuessr: City-Level Video Geo-Localization on a Global Scale
• 地理定位
  • Statewide Visual Geolocalization in the Wild
    :star:code
• Avatars(虚拟人)
  • CanonicalFusion: Generating Drivable 3D Human Avatars from Multiple Images
    :star:code
  • RodinHD: High-Fidelity 3D Avatar Generation with Diffusion Models
    :star:code
  • MeshAvatar: Learning High-quality Triangular Human Avatars from Multi-view Videos
    :star:code
  • PhysAvatar: Learning the Physics of Dressed 3D Avatars from Visual Observations
    :house:project
  • iHuman: Instant Animatable Digital Humans From Monocular Videos
  • PAV: Personalized Head Avatar from Unstructured Video Collection
    :house:project
  • Disentangled Clothed Avatar Generation from Text Descriptions
    :house:project服装头像生成
  • MagicMirror: Fast and High-Quality Avatar Generation with Constrained Search Space
    :house:project
  • 3DFG-PIFu: 3D Feature Grids for Human Digitization from Sparse Views
  • FAMOUS: High-Fidelity Monocular 3D Human Digitization Using View Synthesis
    :star:code3D 人体数字化
  • Instant 3D Human Avatar Generation using Image Diffusion Models
    :house:project
  • Let the Avatar Talk using Texts without Paired Training Data
• VR
  • EgoBody3M: Egocentric Body Tracking on a VR Headset using a Diverse Dataset

38.Human-Object Interaction(人机交互)

• Controllable Human-Object Interaction Synthesis
  :house:project
• F-HOI: Toward Fine-grained Semantic-Aligned 3D Human-Object Interactions
• Interaction-centric Spatio-Temporal Context Reasoning for Multi-Person Video HOI Recognition
  :star:code
• Look Hear: Gaze Prediction for Speech-directed Human Attention
  :star:code
• Boosting Gaze Object Prediction via Pixel-level Supervision from Vision Foundation Model
  :star:code
• Revisit Human-Scene Interaction via Space Occupancy
  :house:project人机交互
• Exploring Conditional Multi-Modal Prompts for Zero-shot HOI Detection
  :star:code
• AFF-ttention! Affordances and Attention models for Short-Term Object Interaction Anticipation
• 手-物
  • NL2Contact: Natural Language Guided 3D Hand-Object Contact Modeling with Diffusion Model
  • Dense Hand-Object(HO) GraspNet with Full Grasping Taxonomy and Dynamics
    :star:code
  • Are Synthetic Data Useful for Egocentric Hand-Object Interaction Detection?
    :star:code
  • Coarse-to-Fine Implicit Representation Learning for 3D Hand-Object Reconstruction from a Single RGB-D Image

37.Style Transfer(风格迁移)

• Towards compact reversible image representations for neural style transfer
• 运动迁移
  • Towards High-Quality 3D Motion Transfer with Realistic Apparel Animation
    :star:code

36.Gaze Estimation

• De-confounded Gaze Estimation
• 3DGazeNet: Generalizing Gaze Estimation with Weak Supervision from Synthetic Views
  :star:code
• LG-Gaze: Learning Geometry-aware Continuous Prompts for Language-Guided Gaze Estimation
• Gaze Target Detection Based on Head-Local-Global Coordination

35.Action Detection(动作检测)

• LEGO: Learning EGOcentric Action Frame Generation via Visual Instruction Tuning
  :star:code
  :house:project
• ActionSwitch: Class-agnostic Detection of Simultaneous Actions in Streaming Videos
• Spatio-Temporal Proximity-Aware Dual-Path Model for Panoramic Activity Recognition
• Motion Keyframe Interpolation for Any Human Skeleton using Point Cloud-based Human Motion Data Homogenisation运动关键帧插值
• 基于骨架的动作识别
  • SA-DVAE: Improving Zero-Shot Skeleton-Based Action Recognition by Disentangled Variational Autoencoders
    :star:code
  • Towards Physical World Backdoor Attacks against Skeleton Action Recognition
    :house:project
  • S-JEPA: A Joint Embedding Predictive Architecture for Skeletal Action Recognition
    :house:project
  • Idempotent Unsupervised Representation Learning for Skeleton-Based Action Recognition
    :star:code
  • CrossGLG: LLM Guides One-shot Skeleton-based 3D Action Recognition in a Cross-level Manner
• 小样本动作识别
  • Trajectory-aligned Space-time Tokens for Few-shot Action Recognition
    :house:project
  • Efficient Few-Shot Action Recognition via Multi-Level Post-Reasoning
    :star:code
• 时序动作检测
  • DyFADet: Dynamic Feature Aggregation for Temporal Action Detection
    :star:code
  • UniMD: Towards Unifying Moment Retrieval and Temporal Action Detection
    :star:code
• 时序动作定位
  • HAT: History-Augmented Anchor Transformer for Online Temporal Action Localization
    :star:code
  • Towards Adaptive Pseudo-label Learning for Semi-Supervised Temporal Action Localization
  • Online Temporal Action Localization with Memory-Augmented Transformer
    :house:project
  • Stepwise Multi-grained Boundary Detector for Point-supervised Temporal Action Localization
• 时序动作分割
  • Long-Tail Temporal Action Segmentation with Group-wise Temporal Logit Adjustment
    :star:code
  • Two-Stage Active Learning for Efficient Temporal Action Segmentation
  • Language-Assisted Skeleton Action Understanding for Skeleton-Based Temporal Action Segmentation
    :star:code
  • Synchronization is All You Need: Exocentric-to-Egocentric Transfer for Temporal Action Segmentation with Unlabeled Synchronized Video Pairs
    :star:code
• 动作质量评估
  • Semi-Supervised Teacher-Reference-Student Architecture for Action Quality Assessment
    :star:code
  • RICA^2: Rubric-Informed, Calibrated Assessment of Actions
    :house:project
  • Vision-Language Action Knowledge Learning for Semantic-Aware Action Quality Assessment动作质量评估
  • MAGR: Manifold-Aligned Graph Regularization for Continual Action Quality Assessment
    :star:code
• 动作预测
  • Semantically Guided Representation Learning For Action Anticipation
    :star:code
  • PALM: Predicting Actions through Language Models预测动作
• 动作识别
  • Referring Atomic Video Action Recognition
    :star:code
  • DEAR: Depth-Enhanced Action Recognition
  • Bayesian Evidential Deep Learning for Online Action Detection
  • C2C: Component-to-Composition Learning for Zero-Shot Compositional Action Recognition
    :star:code
  • Masked Video and Body-worn IMU Autoencoder for Egocentric Action Recognition
  • Classification Matters: Improving Video Action Detection with Class-Specific Attention
  • FinePseudo: Improving Pseudo-Labelling through Temporal-Alignablity for Semi-Supervised Fine-Grained Action Recognition
    :house:project
  • Context-Aware Action Recognition: Introducing a Comprehensive Dataset for Behavior Contrast
  • Multimodal Cross-Domain Few-Shot Learning for Egocentric Action Recognition
    :house:project
  • On the Utility of 3D Hand Poses for Action Recognition
    :house:project
  • POET: Prompt Offset Tuning for Continual Human Action Adaptation
    :star:code
  • Occluded Gait Recognition with Mixture of Experts: An Action Detection Perspective
    :star:code
  • Leveraging temporal contextualization for video action recognition
    :star:code
  • Optimizing Factorized Encoder Models: Time and Memory Reduction for Scalable and Efficient Action Recognition
  • SkateFormer: Skeletal-Temporal Transformer for Human Action Recognition
    :house:project
• 动作理解
  • EgoExo-Fitness: Towards Egocentric and Exocentric Full-Body Action Understanding
    :star:code
• 群体动作识别
  • Towards More Practical Group Activity Detection: A New Benchmark and Model
    :house:project
  • Flow-Assisted Motion Learning Network for Weakly-Supervised Group Activity Recognition
  • Skeleton-based Group Activity Recognition via Spatial-Temporal Panoramic Graph
    :star:code
• 癫痫发作检测
  • VSViG: Real-time Video-based Seizure Detection via Skeleton-based Spatiotemporal ViG

34.Visual Question Answering(视觉问答)

• DriveLM: Driving with Graph Visual Question Answering
  :star:code
• Diffusion-Refined VQA Annotations for Semi-Supervised Gaze Following
• WSI-VQA: Interpreting Whole Slide Images by Generative Visual Question Answering
  :star:code
• GRACE: Graph-Based Contextual Debiasing for Fair Visual Question Answering
• Q&A Prompts: Discovering Rich Visual Clues through Mining Question-Answer Prompts for VQA requiring Diverse World Knowledge
  :star:code
• Compositional Substitutivity of Visual Reasoning for Visual Question Answering
  :star:code
• Fully Authentic Visual Question Answering Dataset from Online Communities
  :house:project
• An Explainable Vision Question Answer Model via Diffusion Chain-of-Thought
• 音视频问答
  • Learning Trimodal Relation for AVQA with Missing Modality
• 视频问答
  • Video Question Answering with Procedural Programs
    :house:project
  • ViLA: Efficient Video-Language Alignment for Video Question Answering
    :star:code
  • TimeCraft: Navigate Weakly-Supervised Temporal Grounded Video Question Answering via Bi-directional ReasoningVQA
  • AutoEval-Video: An Automatic Benchmark for Assessing Large Vision Language Models in Open-Ended Video Question Answering
    :star:code
• 视听问答
  • Learning Trimodal Relation for Audio-Visual Question Answering with Missing Modality

33.Motion Generation(人体运动生成)

• Event-Based Motion Magnification
  :star:code
• Learning-based Axial Video Motion Magnification
  :house:project
• SMooDi: Stylized Motion Diffusion Model
  :star:code
• Length-Aware Motion Synthesis via Latent Diffusion
  :star:code
• HUMOS: Human Motion Model Conditioned on Body Shape
  :star:code
• HumanRefiner: Benchmarking Abnormal Human Generation and Refining with Coarse-to-fine Pose-Reversible Guidance
  :star:code
• Generating Physically Realistic and Directable Human Motions from Multi-Modal Inputs
  :house:project
• Generating Human Interaction Motions in Scenes with Text Control
  :house:project运动生成
• Motion Mamba: Efficient and Long Sequence Motion Generation
  :star:code
  :house:project
• Large Motion Model for Unified Multi-Modal Motion Generation
  :house:project
• EMDM: Efficient Motion Diffusion Model for Fast and High-Quality Motion Generation
  :star:code
  :house:project
• Bridging the Gap Between Human Motion and Action Semantics via Kinematics Phrases
  :house:project人体运动
• TRAM: Global Trajectory and Motion of 3D Humans from in-the-wild Videos
  :house:project人体运动
• Nymeria: A Massive Collection of Egocentric Multi-modal Human Motion in the Wild人体运动
• FreeMotion: MoCap-Free Human Motion Synthesis with Multimodal Large Language Models
• MotionLCM: Real-time Controllable Motion Generation via Latent Consistency Model
  :star:code
• Realistic Human Motion Generation with Cross-Diffusion Models
  :house:project人体运动
• CoMo: Controllable Motion Generation through Language Guided Pose Code Editing
  :house:project生成可控运动
• TLControl: Trajectory and Language Control for Human Motion Synthesis
  :house:project人体运动合成
• Retrieval Robust to Object Motion Blur
  :star:[code]((https://github.com/Rong-Zou/Retrieval-Robust-to-Object-Motion-Blur)
• 三维人体运动合成
  • ReMoS: 3D Motion-Conditioned Reaction Synthesis for Two-Person Interactions
    :house:project
• 文本-动作合成
  • FreeMotion: A Unified Framework for Number-free Text-to-Motion Synthesis
  • Local Action-Guided Motion Diffusion Model for Text-to-Motion Generation
    :star:code
  • Plan, Posture and Go: Towards Open-vocabulary Text-to-Motion Generation
    :house:project
  • ParCo: Part-Coordinating Text-to-Motion Synthesis
    :star:code
• 人体运动预测
  • Human Motion Forecasting in Dynamic Domain Shifts: A Homeostatic Continual Test-time Adaptation Framework人体运动预测
  • Scene-aware Human Motion Forecasting via Mutual Distance Prediction
• 人体运动估计
  • MANIKIN: Biomechanically Accurate Neural Inverse Kinematics for Human Motion Estimation
    :house:project
• 运动估计
  • Motion-prior Contrast Maximization for Dense Continuous-Time Motion Estimation
    :star:code
  • COIN: Control-Inpainting Diffusion Prior for Human and Camera Motion Estimation
• 舞蹈生成
  • Beat-It: Beat-Synchronized Multi-Condition 3D Dance Generation
    :house:project
• 行为生成
  • DIM: Dyadic Interaction Modeling for Social Behavior Generation
    :star:code
• 运动迁移
  • Temporal Residual Jacobians for Rig-free Motion Transfer
    :house:project
    🤗huggingface
• 运动预测
  • Enhanced Motion Forecasting with Visual Relation Reasoning

32.Person Re-Identification(人员重识别)

• Human-in-the-Loop Visual Re-ID for Population Size Estimation
  :star:code
• 行人重识别
  • Keypoint Promptable Re-Identification
    :star:code
  • Privacy-Preserving Adaptive Re-Identification without Image Transfer
  • Rethinking Normalization Layers for Domain Generalizable Person Re-identification
    :star:code
  • Domain Shifting: A Generalized Solution for Heterogeneous Cross-Modality Person Re-Identification
  • VI-ReID
    • Multi-Memory Matching for Unsupervised Visible-Infrared Person Re-Identification
      :thumbsup:无监督可见光-红外行人重识别（USL-VI-ReID）
    • WRIM-Net: Wide-Ranging Information Mining Network for Visible-Infrared Person Re-Identification
• 人物搜索
  • PLOT: Text-based Person Search with Part Slot Attention for Corresponding Part Discovery基于文本的人物搜索
• 步态识别
  • Camera-LiDAR Cross-modality Gait Recognition
  • Free Lunch for Gait Recognition: A Novel Relation Descriptor
  • Causality-inspired Discriminative Feature Learning in Triple Domains for Gait Recognition
  • Cut out the Middleman: Revisiting Pose-based Gait Recognition
    :star:code
• 计数
  • CountFormer: Multi-View Crowd Counting Transformer
  • Robust Zero-Shot Crowd Counting and Localization with Adaptive Resolution SAM
  • Multi-modal Crowd Counting via a Broker Modality
    :star:code
  • Improving Point-based Crowd Counting and Localization Based on Auxiliary Point Guidance
    :star:code

31.Point Clouds(点云)

• SEED: A Simple and Effective 3D DETR in Point Clouds
  :star:code
• PointLLM: Empowering Large Language Models to Understand Point Clouds
  :star:code
  :house:project
• TransCAD: A Hierarchical Transformer for CAD Sequence Inference from Point Clouds
• Learning to Adapt SAM for Segmenting Cross-domain Point Clouds
• Meerkat: Audio-Visual Large Language Model for Grounding in Space and Time
• milliFlow: Scene Flow Estimation on mmWave Radar Point Cloud for Human Motion Sensing
  :star:code
• Fast Point Cloud Geometry Compression with Context-based Residual Coding and INR-based Refinement
• Learning Local Pattern Modularization for Point Cloud Reconstruction from Unseen Classes
  :star:code
• T-MAE: Temporal Masked Autoencoders for Point Cloud Representation Learning
  :star:code
• Progressive Classifier and Feature Extractor Adaptation for Unsupervised Domain Adaptation on Point Clouds
  :star:code
• PFGS: High Fidelity Point Cloud Rendering via Feature Splatting
  :star:code
• Masked Motion Prediction with Semantic Contrast for Point Cloud Sequence Learning
  :star:code
• To Supervise or Not to Supervise: Understanding and Addressing the Key Challenges of Point Cloud Transfer Learning
• Relightable 3D Gaussians: Realistic Point Cloud Relighting with BRDF Decomposition and Ray Tracing
  :star:code
• FastPCI: Motion-Structure Guided Fast Point Cloud Frame Interpolation
  :star:code
• 点云生成
  • RangeLDM: Fast Realistic LiDAR Point Cloud Generation
    :star:code
  • Text2LiDAR: Text-guided LiDAR Point Clouds Generation via Equirectangular Transformer
    :star:code
  • Fast Training of Diffusion Transformer with Extreme Masking for 3D Point Clouds Generation
    :house:project
  • FrePolad: Frequency-Rectified Point Latent Diffusion for Point Cloud Generation
    :house:project
• 点云完成
  • Explicitly Guided Information Interaction Network for Cross-modal Point Cloud Completion
    :star:code
  • T-CorresNet: Template Guided 3D Point Cloud Completion with Correspondence Pooling Query Generation Strategy
    :star:code
  • AEDNet: Adaptive Embedding and Multiview-Aware Disentanglement for Point Cloud Completion
  • EINet: Point Cloud Completion via Extrapolation and Interpolation
    :star:code
  • Syn-to-Real Domain Adaptation for Point Cloud Completion via Part-based Approach
    :star:code
  • ProtoComp: Diverse Point Cloud Completion with Controllable Prototype
    :star:code
• 点云重建
  • DiffPMAE: Diffusion Masked Autoencoders for Point Cloud Reconstruction
    :star:code
• 点云理解
  • DG-PIC: Domain Generalized Point-In-Context Learning for Point Cloud Understanding
• 点云配准
  • ML-SemReg: Boosting Point Cloud Registration with Multi-level Semantic Consistency
    :star:code
  • PointRegGPT: Boosting 3D Point Cloud Registration using Generative Point-Cloud Pairs for Training
    :star:code
  • SemReg: Semantics Constrained Point Cloud Registration
    :star:code
  • Correspondence-Free SE(3) Point Cloud Registration in RKHS via Unsupervised Equivariant Learning
    :house:project
  • UMERegRobust – Universal Manifold Embedding Compatible Features for Robust Point Cloud Registration
    :star:code
  • PARE-Net: Position-Aware Rotation-Equivariant Networks for Robust Point Cloud Registration
    :star:code
  • UMERegRobust -- Universal Manifold Embedding Compatible Features for Robust Point Cloud Registration
    :star:code
  • Equi-GSPR: Equivariant SE(3) Graph Network Model for Sparse Point Cloud Registration点云配准
• 点云分割
  • Dual-level Adaptive Self-Labeling for Novel Class Discovery in Point Cloud Segmentation
  • HGL: Hierarchical Geometry Learning for Test-time Adaptation in 3D Point Cloud Segmentation
    :star:code
  • SegPoint: Segment Any Point Cloud via Large Language Model
    :star:code
  • Localization and Expansion: A Decoupled Framework for Point Cloud Few-shot Semantic Segmentation
  • Pseudo-Embedding for Generalized Few-Shot Point Cloud Segmentation
    :star:code
  • Subspace Prototype Guidance for Mitigating Class Imbalance in Point Cloud Semantic Segmentation
    :star:code
• 点云理解
  • GPSFormer: A Global Perception and Local Structure Fitting-based Transformer for Point Cloud Understanding
    :star:code
• 3D点云
  • Implicit Filtering for Learning Neural Signed Distance Functions from 3D Point Clouds
    :star:code
  • CloudFixer: Test-Time Adaptation for 3D Point Clouds via Diffusion-Guided Geometric Transformation
    :star:code
  • FLAT: Flux-aware Imperceptible Adversarial Attacks on 3D Point Clouds
  • RISurConv: Rotation Invariant Surface Attention-Augmented Convolutions for 3D Point Cloud Classification and Segmentation
  • P2P-Bridge: Diffusion Bridges for 3D Point Cloud Denoising
    :star:code
  • Heterogeneous Graph Learning for Scene Graph Prediction in 3D Point Clouds
  • Hiding Imperceptible Noise in Curvature-Aware Patches for 3D Point Cloud Attack3D 点云攻击
  • Continuous SO(3) Equivariant Convolution for 3D Point Cloud Analysis
    :star:code
  • Frugal 3D Point Cloud Model Training via Progressive Near Point Filtering and Fused Aggregation

30.Anomaly Detection(异常检测)

• Continuous Memory Representation for Anomaly Detection
  :star:code
• Dissolving Is Amplifying: Towards Fine-Grained Anomaly Detection
  :star:code
• Few-Shot Anomaly-Driven Generation for Anomaly Classification and Segmentation
  :star:code
• GeneralAD: Anomaly Detection Across Domains by Attending to Distorted Features
  :star:code
• Learning Diffusion Models for Multi-View Anomaly Detection
• Hierarchical Gaussian Mixture Normalizing Flow Modeling for Unified Anomaly Detection
  :star:code
• TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection
  :star:code
• Unsupervised, Online and On-The-Fly Anomaly Detection For Non-Stationary Image Distributions
  :star:code
• MoEAD: A Parameter-efficient Model for Multi-class Anomaly Detection
  :star:code
• 缺陷检测
  • An Incremental Unified Framework for Small Defect Inspection
    :star:code
• 故障检测
  • DECIDER: Leveraging Foundation Model Priors for Improved Model Failure Detection and Explanation
    :star:code
• 3D异常检测
  • R3D-AD: Reconstruction via Diffusion for 3D Anomaly Detection
• 工业异常检测
  • Self-supervised Feature Adaptation for 3D Industrial Anomaly Detection
  • A Unified Anomaly Synthesis Strategy with Gradient Ascent for Industrial Anomaly Detection and Localization
    :star:code
  • GLAD: Towards Better Reconstruction with Global and Local Adaptive Diffusion Models for Unsupervised Anomaly Detection
    :star:code
  • AD3: Introducing a score for Anomaly Detection Dataset Difficulty assessment using VIADUCT dataset
  • Learning to Detect Multi-class Anomalies with Just One Normal Image Prompt
• 零样本异常检测
  • AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detection
    :star:code
• 多类异常检测
  • Learning Unified Reference Representation for Unsupervised Multi-class Anomaly Detection
• OOD
  • Gradient-Regularized Out-of-Distribution Detection
  • SAFT: Towards Out-of-Distribution Generalization in Fine-Tuning
  • PixOOD: Pixel-Level Out-of-Distribution Detection
    :star:code
  • An Information Theoretical View for Out-Of-Distribution Detection
  • Learning Non-Linear Invariants for Unsupervised Out-of-Distribution Detection
  • LAPT: Label-driven Automated Prompt Tuning for OOD Detection with Vision-Language Models
    :star:code
  • ProSub: Probabilistic Open-Set Semi-Supervised Learning with Subspace-Based Out-of-Distribution Detection
    :star:code
  • Diffusion for Out-of-Distribution Detection on Road Scenes and Beyond
    :star:code
  • Can Your Generative Model Detect Out-of-Distribution Covariate Shift?
  • Gradient-based Out-of-Distribution Detection
  • Vision-Language Dual-Pattern Matching for Out-of-Distribution Detection
  • TAG: Text Prompt Augmentation for Zero-Shot Out-of-Distribution Detection
    :star:code
• 异常值检测
  • Rethinking Unsupervised Outlier Detection via Multiple Thresholding无监督异常值检测
• 零样本异常分割
  • VCP-CLIP: A visual context prompting model for zero-shot anomaly segmentation
    :star:code

29.Semi/self-supervised learning(半/自监督)

• SweepNet: Unsupervised Learning Shape Abstraction via Neural Sweepers
  :house:project
• Region-aware Distribution Contrast: A Novel Approach to Multi-Task Partially Supervised Learning
  :star:code
• 自监督
  • CroMo-Mixup: Augmenting Cross-Model Representations for Continual Self-Supervised Learning
    :star:code
  • HPFF: Hierarchical Locally Supervised Learning with Patch Feature Fusion
    :star:code
  • SCPNet: Unsupervised Cross-modal Homography Estimation via Intra-modal Self-supervised Learning
    :star:code
  • Efficient Unsupervised Visual Representation Learning with Explicit Cluster Balancing
  • OmniSat: Self-Supervised Modality Fusion for Earth Observation
    :star:code
    :house:project
    :sunflower:dataset
  • FroSSL: Frobenius Norm Minimization for Efficient Multiview Self-Supervised Learning
  • Self-supervised visual learning from interactions with objects
  • Exploiting Supervised Poison Vulnerability to Strengthen Self-Supervised Defense
  • GenView: Enhancing View Quality with Pretrained Generative Model for Self-Supervised Learning
    :star:code
  • On Pretraining Data Diversity for Self-Supervised Learning
    :star:code
  • Decoupling Common and Unique Representations for Multimodal Self-supervised Learning
    :star:code
  • POA: Pre-training Once for Models of All Sizes
    :star:code
  • ViC-MAE: Self-Supervised Representation Learning from Images and Video with Contrastive Masked Autoencoders自监督表示学习
  • Pose-Aware Self-Supervised Learning with Viewpoint Trajectory Regularization
    :house:project自监督学习
  • SSL-Cleanse: Trojan Detection and Mitigation in Self-Supervised Learning
    :star:code
• 半监督
  • Image-Feature Weak-to-Strong Consistency: An Enhanced Paradigm for Semi-Supervised Learning
  • Improving 3D Semi-supervised Learning by Effectively Utilizing All Unlabelled Data
    :star:code
  • SCOMatch: Alleviating Overtrusting in Open-set Semi-supervised Learning
    :star:code
  • ExMatch: Self-guided Exploitation for Semi-Supervised Learning with Scarce Labeled Samples
  • Rebalancing Using Estimated Class Distribution for Imbalanced Semi-Supervised Learning under Class Distribution Mismatch半监督学习
  • Towards Latent Masked Image Modeling for Self-Supervised Visual Representation Learning
    :star:code
  • Flexible Distribution Alignment: Towards Long-tailed Semi-supervised Learning with Proper Calibration
    :star:code

28.Novel Class Discovery(新类发现)

• Self-Cooperation Knowledge Distillation for Novel Class Discovery

27.GNN/GCN

• GKGNet: Group K-Nearest Neighbor based Graph Convolutional Network for Multi-Label Image Recognition
  :star:codeGNN
• Graph Neural Network Causal Explanation via Neural Causal Models
  :star:code
• On the Topology Awareness and Generalization Performance of Graph Neural Networks
• Causal Subgraphs and Information Bottlenecks: Redefining OOD Robustness in Graph Neural Networks

26.NAS

• Auto-GAS: Automated Proxy Discovery for Training-free Generative Architecture Search
  :star:code
• Auto-DAS: Automated Proxy Discovery for Training-free Distillation-aware Architecture Search
  :star:code蒸馏感
• SuperFedNAS: Cost-Efficient Federated Neural Architecture Search for On-Device Inference
• Dependency-aware Differentiable Neural Architecture Search

25.MC/KD/Pruning(模型压缩/知识蒸馏/剪枝)

• DεpS: Delayed ε-Shrinking for Faster Once-For-All Training
• 模型压缩
  • Clean & Compact: Efficient Data-Free Backdoor Defense with Model Compactness
• 剪枝
  • Non-transferable Pruning
  • Straightforward Layer-wise Pruning for More Efficient Visual Adaptation
  • Isomorphic Pruning for Vision Models
    :star:code
  • LPViT: Low-Power Semi-structured Pruning for Vision Transformers
  • PaPr: Training-Free One-Step Patch Pruning with Lightweight ConvNets for Faster Inference
    :star:code剪
  • Enhanced Sparsification via Stimulative Training
    :star:code
  • SNP: Structured Neuron-level Pruning to Preserve Attention Scores
    :star:code
• 量化
  • GenQ: Quantization in Low Data Regimes with Generative Synthetic Data
    :star:code
  • MetaAug: Meta-Data Augmentation for Post-Training Quantization
  • Toward INT4 Fixed-Point Training via Exploring Quantization Error for Gradients
  • CLAMP-ViT: Contrastive Data-Free Learning for Adaptive Post-Training Quantization of ViTs
    :star:code
  • AdaLog: Post-Training Quantization for Vision Transformers with Adaptive Logarithm Quantizer
    :star:code
  • POCA: Post-training Quantization with Temporal Alignment for Codec Avatars
    :house:project量化
• KD
  • Simple Unsupervised Knowledge Distillation With Space Similarity知识蒸馏
  • Direct Distillation between Different DomainsKD
  • Harmonizing knowledge Transfer in Neural Network with Unified Distillation
  • Good Teachers Explain: Explanation-Enhanced Knowledge Distillation
  • The Role of Masking for Efficient Supervised Knowledge Distillation of Vision Transformers
  • Improving Knowledge Distillation via Regularizing Feature Direction and Norm
  • Adversarially Robust Distillation by Reducing the Student-Teacher Variance Gap蒸馏
  • Improving Zero-shot Generalization of Learned Prompts via Unsupervised Knowledge Distillation
    :star:code
  • UNIKD: UNcertainty-Filtered Incremental Knowledge Distillation for Neural Implicit Representation
    :star:code
  • BKDSNN: Enhancing the Performance of Learning-based Spiking Neural Networks Training with Blurred Knowledge Distillation
  • Nickel and Diming Your GAN: A Dual-Method Approach to Enhancing GAN Efficiency via Knowledge Distillation
  • How to Train the Teacher Model for Effective Knowledge Distillation
  • Markov Knowledge Distillation: Make Nasty Teachers trained by Self-undermining Knowledge Distillation Fully Distillable

24.Vision Transformer

• Spline-based Transformers
• Denoising Vision Transformers
• FairViT: Fair Vision Transformer via Adaptive Masking
• Rotary Position Embedding for Vision Transformer
  :star:code
• Bidirectional Progressive Transformer for Interaction Intention Anticipation
• Robustness Tokens: Towards Adversarial Robustness of Transformers
• SpecFormer: Guarding Vision Transformer Robustness via Maximum Singular Value Penalization
  :star:code
• PDiscoFormer: Relaxing Part Discovery Constraints with Vision Transformers
• OAT: Object-Level Attention Transformer for Gaze Scanpath Prediction
  :star:code
• AugDETR: Improving Multi-scale Learning for Detection TransformerTransformer
• AttnZero: Efficient Attention Discovery for Vision Transformers
  :star:code
• SpatialFormer: Towards Generalizable Vision Transformers with Explicit Spatial Understanding
  :star:code
• Efficient Vision Transformers with Partial Attention
• SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers
  :star:code
• Stitched ViTs are Flexible Vision Backbones
  :star:code
• Token Compensator: Altering Inference Cost of Vision Transformer without Re-Tuning
• Uncertainty-Driven Spectral Compressive Imaging with Spatial-Frequency Transformer
  :star:code
• GiT: Towards Generalist Vision Transformer through Universal Language Interface
  :star:code
• An Optimal Control View of LoRA and Binary Controller Design for Vision Transformers
• Fairness-aware Vision Transformer via Debiased Self-Attention
  :star:code
• ScatterFormer: Efficient Voxel Transformer with Scattered Linear Attention
  :star:code
• LiFT: A Surprisingly Simple Lightweight Feature Transform for Dense ViT Descriptors
  :house:project
• Parameter-Efficient and Memory-Efficient Tuning for Vision Transformer: A Disentangled Approach
  :house:project
• LayoutDETR: Detection Transformer Is a Good Multimodal Layout Designer
  :star:code
• Efficient Diffusion Transformer with Step-wise Dynamic Attention Mediators
  :star:code
• BAM-DETR: Boundary-Aligned Moment Detection Transformer for Temporal Sentence Grounding in Videos
  :star:code
• An Efficient and Effective Transformer Decoder-Based Framework for Multi-Task Visual Grounding
  :star:code

23.Machine Learning(机器学习)

• Learning to Unlearn for Robust Machine Unlearning
• Is Retain Set All You Need in Machine Unlearning? Restoring Performance of Unlearned Models with Out-Of-Distribution Images
  :star:code机器学习
• 机器遗忘
  • MultiDelete for Multimodal Machine Unlearning
    :star:code
  • Challenging Forgets: Unveiling the Worst-Case Forget Sets in Machine Unlearning
    :star:code
• 对抗
  • Improving Adversarial Transferability via Model Alignment
    :star:code
  • Event Trojan: Asynchronous Event-based Backdoor Attacks
    :star:code
  • Data Poisoning Quantization Backdoor Attack
  • Flatness-aware Sequential Learning Generates Resilient Backdoors
  • WBP: Training-time Backdoor Attacks through Hardware-based Weight Bit Poisoning
    :star:code
  • Cocktail Universal Adversarial Attack on Deep Neural Networks
  • TrojVLM: Backdoor Attack Against Vision Language Models
  • CatchBackdoor: Backdoor Detection via Critical Trojan Neural Path Fuzzing
  • Prompt-Driven Contrastive Learning for Transferable Adversarial Attacks
    :star:code
  • Self-Supervised Representation Learning for Adversarial Attack Detection
  • Prediction Exposes Your Face: Black-box Model Inversion via Prediction Alignment
  • CLIP-Guided Networks for Transferable Targeted Attacks
  • CLIP-Guided Generative Networks for Transferable Targeted Adversarial Attacks
  • Exploring Vulnerabilities in Spiking Neural Networks: Direct Adversarial Attacks on Raw Event Data
  • UNIT: Backdoor Mitigation via Automated Neural Distribution Tightening
    :star:code
  • Inter-Class Topology Alignment for Efficient Black-Box Substitute Attacks黑盒
  • Any Target Can be Offense: Adversarial Example Generation via Generalized Latent Infection
    :star:code
  • AdvDiff: Generating Unrestricted Adversarial Examples using Diffusion Models
    :star:code
  • Enhancing Tracking Robustness with Auxiliary Adversarial Defense Networks
  • DIFFender: Diffusion-Based Adversarial Defense against Patch Attacks
    :star:code
• 持续学习
  • CLEO: Continual Learning of Evolving Ontologies
  • One-stage Prompt-based Continual Learning
  • Exemplar-free Continual Representation Learning via Learnable Drift Compensation
    :star:code
  • Diffusion-Driven Data Replay: A Novel Approach to Combat Forgetting in Federated Class Continual Learning
    :star:code
  • Semantic Residual Prompts for Continual Learning
    :star:code
  • Pick-a-back: Selective Device-to-Device Knowledge Transfer in Federated Continual Learning
    :star:code
  • RCS-Prompt: Learning Prompt to Rearrange Class Space for Prompt-based Continual Learning
    :star:code
  • PromptFusion: Decoupling Stability and Plasticity for Continual Learning
    :star:code
  • Information Bottleneck Based Data Correction in Continual Learning
  • Revisiting Supervision for Continual Representation Learning
    :star:code持续
  • Anytime Continual Learning for Open Vocabulary Classification
    :star:code
  • MagMax: Leveraging Model Merging for Seamless Continual Learning
  • Beyond Prompt Learning: Continual Adapter for Efficient Rehearsal-Free Continual Learning
• 迁移学习
  • SHERL: Synthesizing High Accuracy and Efficient Memory for Resource-Limited Transfer Learning
    :star:code
• 主动学习
  • Dataset Quantization with Active Learning based Adaptive Sampling
  • Generalized Coverage for More Robust Low-Budget Active Learning
  • Bad Students Make Great Teachers: Active Learning Accelerates Large-Scale Visual Understanding主动学习
  • Bidirectional Uncertainty-Based Active Learning for Open-Set Annotation主动学习
  • Exploring Active Learning in Meta-Learning: Enhancing Context Set Labeling
• 强化学习
  • Reinforcement Learning Meets Visual Odometry
  • Large-scale Reinforcement Learning for Diffusion Models
  • Reinforcement Learning via Auxillary Task Distillation
  • Reinforcement Learning Friendly Vision-Language Model for Minecraft
    :star:code
  • Multimodal Label Relevance Ranking via Reinforcement Learning
    :star:code
  • Enhancing Diffusion Models with Text-Encoder Reinforcement Learning
    :star:code
  • Diffusion Models as Optimizers for Efficient Planning in Offline RL
    :star:code
  • Unified Local-Cloud Decision-Making via Reinforcement Learning
    :house:project强化学习
• 联邦学习
  • Towards Multi-modal Transformers in Federated Learning
    :star:code
  • FedHide: Federated Learning by Hiding in the Neighbors
  • FedHARM: Harmonizing Model Architectural Diversity in Federated Learning
    :star:code
  • FedTSA: A Cluster-based Two-Stage Aggregation Method for Model-heterogeneous Federated Learning
  • Unlocking the Potential of Federated Learning: The Symphony of Dataset Distillation via Deep Generative Latents
    :star:code
  • PFedEdit: Personalized Federated Learning via Automated Model Editing
    :star:code
  • Overcome Modal Bias in Multi-modal Federated Learning via Balanced Modality Selection
  • Fisher Calibration for Backdoor-Robust Heterogeneous Federated Learning
    :star:code
  • Federated Learning with Local Openset Noisy Labels
    :star:code
  • SkyMask: Attack-agnostic Robust Federated Learning with Fine-grained Learnable Masks
    :star:code
• 对比学习
  • FlowCon: Out-of-Distribution Detection using Flow-based Contrastive Learning
    :star:code
  • Improving Medical Multi-modal Contrastive Learning with Expert Annotations
  • Contrastive Learning with Synthetic Positives对比学习
  • Understanding and Mitigating Human-Labelling Errors in Supervised Contrastive Learning
  • Adaptive Multi-head Contrastive Learning
  • CLAP: Isolating Content from Style through Contrastive Learning with Augmented Prompts
    :star:code对比学习
  • Learning the Unlearned: Mitigating Feature Suppression in Contrastive Learning
    :star:code
• 类增量
  • Rethinking Few-shot Class-incremental Learning: Learning from Yourself
    :star:code
  • Few-shot Class Incremental Learning with Attention-Aware Self-Adaptive Prompt
  • Class-Incremental Learning with CLIP: Adaptive Representation Adjustment and Parameter Fusion
    :star:code
  • Versatile Incremental Learning: Towards Class and Domain-Agnostic Incremental Learning
    :star:code
  • Confidence Self-Calibration for Multi-Label Class-Incremental Learning
    :star:[code](https://github.com/ Kaile-Du/CSC)
  • Canonical Shape Projection is All You Need for 3D Few-shot Class Incremental Learning
    :star:code
  • Personalized Federated Domain-Incremental Learning based on Adaptive Knowledge Matching
  • PILoRA: Prototype Guided Incremental LoRA for Federated Class-Incremental Learning
    :star:code
  • CLOSER: Towards Better Representation Learning for Few-Shot Class-Incremental Learning
    :star:code
  • Non-Exemplar Domain Incremental Learning via Cross-Domain Concept Integration
    :star:code
  • On the Approximation Risk of Few-Shot Class-Incremental Learning
    :star:code
  • iNeMo: Incremental Neural Mesh Models for Robust Class-Incremental Learning
    :star:code
  • DiffClass: Diffusion-Based Class Incremental Learning
• 上下文学习
  • Visual Prompt Selection for In-Context Learning Segmentation
• 多任务学习
  • Learning Representation for Multitask Learning through Self-Supervised Auxiliary Learning
  • 多标签学习
    • IGNORE: Information Gap-based False Negative Loss Rejection for Single Positive Multi-Label Learning
    • Dual-Decoupling Learning and Metric-Adaptive Thresholding for Semi-Supervised Multi-Label Learning
• 多实例学习
  • cDP-MIL: Robust Multiple Instance Learning via Cascaded Dirichlet Process
    :star:code
• 多模态学习
  • Diagnosing and Re-learning for Balanced Multimodal Learning
    :star:code
  • Missing Modality Prediction for Unpaired Multimodal Learning via Joint Embedding of Unimodal Models
  • Robust Multimodal Learning via Representation Decoupling
  • Object-Oriented Anchoring and Modal Alignment in Multimodal Learning

22.Few/Zero-Shot Learning/DG/A(小/零样本/域泛化/域适应)

• Source-Free Domain-Invariant Performance Prediction
• The Devil is in the Few Shots: Iterative Visual Knowledge Completion for Few-shot Learning
  :star:code
• DG
  • Towards Multimodal Open-Set Domain Generalization and Adaptation through Self-supervision
    :star:code
  • Feature Diversification and Adaptation for Federated Domain Generalization
  • Soft Prompt Generation for Domain Generalization
    :star:code
  • Integrating Markov Blanket Discovery into Causal Representation Learning for Domain Generalization
  • Rethinking LiDAR Domain Generalization: Single Source as Multiple Density Domains
    :star:code
  • Improving Zero-Shot Generalization for CLIP with Variational Adapter
  • Representation Enhancement-Stabilization: Reducing Bias-Variance of Domain Generalization
    :star:code
  • Local and Global Flatness for Federated Domain Generalization
    :star:code
  • Learn to Preserve and Diversify: Parameter-Efficient Group with Orthogonal Regularization for Domain Generalization
  • Disentangling Masked Autoencoders for Unsupervised Domain Generalization
    :star:code
• DA
  • Training-Free Model Merging for Multi-target Domain Adaptation
    :star:code
  • MC-PanDA: Mask Confidence for Panoptic Domain Adaptation
    :star:code
  • Is user feedback always informative? Retrieval Latent Defending for Semi-Supervised Domain Adaptation without Source Data
    :house:project
  • De-Confusing Pseudo-Labels in Source-Free Domain Adaptation
  • Open-set Domain Adaptation via Joint Error based Multi-class Positive and Unlabeled Learning
  • Robust Nearest Neighbors for Source-Free Domain Adaptation under Class Distribution Shift
    :star:code
  • HVCLIP: High-dimensional Vector in CLIP for Unsupervised Domain Adaptation
  • Hierarchical Unsupervised Relation Distillation for Source Free Domain Adaptation
  • Learn from the Learnt: Source-Free Active Domain Adaptation via Contrastive Sampling and Visual Persistence
    :star:code
  • Improving Unsupervised Domain Adaptation: A Pseudo-Candidate Set Approach
  • UDA-Bench: Revisiting Common Assumptions in Unsupervised Domain Adaptation Using a Standardized Framework
    :star:code
  • Forget More to Learn More: Domain-specific Feature Unlearning for Semi-supervised and Unsupervised Domain Adaptation
  • Train Till You Drop: Towards Stable and Robust Source-free Unsupervised 3D Domain Adaptation
  • Get Your Embedding Space in Order: Domain-Adaptive Regression for Forest Monitoring
    :house:project
  • CoDA: Instructive Chain-of-Domain Adaptation with Severity-Aware Visual Prompt Tuning
    :star:code
  • COD: Learning Conditional Invariant Representation for Domain Adaptation Regression
  • Plug and Play: A Representation Enhanced Domain Adapter for Collaborative Perception
    :star:code
  • DA-BEV: Unsupervised Domain Adaptation for Bird's Eye View Perception
• 零样本
  • Efficient and Versatile Robust Fine-Tuning of Zero-shot Models
  • Prompting Language-Informed Distribution for Compositional Zero-Shot Learning
    :star:code
  • MRSP: Learn Multi-Representations of Single Primitive for Compositional Zero-Shot Learning

21.Vision-Language(视觉语言)

• Sapiens: Foundation for Human Vision Models
• Conceptual Codebook Learning for Vision-Language Models
• DEAL: Disentangle and Localize Concept-level Explanations for VLMs
• FlexAttention for Efficient High-Resolution Vision-Language Models
  :house:project
• QUAR-VLA: Vision-Language-Action Model for Quadruped Robots
• Vary: Scaling up the Vision Vocabulary for Large Vision-Language Models
• REVISION: Rendering Tools Enable Spatial Fidelity in Vision-Language Models
  :house:project
• Octopus: Embodied Vision-Language Programmer from Environmental Feedback
  :house:project
• GazeXplain: Learning to Predict Natural Language Explanations of Visual Scanpaths
• Learning Chain of Counterfactual Thought for Bias-Robust Vision-Language Reasoning
  :star:code
• Boosting Transferability in Vision-Language Attacks via Diversification along the Intersection Region of Adversarial Trajectory
  :star:code
• Cascade Prompt Learning for Vision-Language Model Adaptation
  :star:code
• The Hard Positive Truth about Vision-Language Compositionality
• Improving 2D Feature Representations by 3D-Aware Fine-Tuning
  :star:code
• Unveiling Typographic Deceptions: Insights of the Typographic Vulnerability in Large Vision-Language Models
  :star:code
• Turbo: Informativity-Driven Acceleration Plug-In for Vision-Language Large Models
• ClearCLIP: Decomposing CLIP Representations for Dense Vision-Language Inference
  :star:code
• FALIP: Visual Prompt as Foveal Attention Boosts CLIP Zero-Shot Performance
  :house:project
• Deciphering the Role of Representation Disentanglement: Investigating Compositional Generalization in CLIP Models
  :star:code
• GalLoP: Learning Global and Local Prompts for Vision-Language Models
• Quantized Prompt for Efficient Generalization of Vision-Language Models
  :star:code
• AddressCLIP: Empowering Vision-Language Models for City-wide Image Address Localization
  :star:code
  :Thumbsup:AddressCLIP：一张图实现街道级定位，端到端图像地理定位大模型
• SQ-LLaVA: Self-Questioning for Large Vision-Language Assistant
• Contrastive Region Guidance: Improving Grounding in Vision-Language Models without Training
  :house:project
• Cascade Prompt Learning for Visual-Language Model Adaptation
  :star:code
• Select and Distill: Selective Dual-Teacher Knowledge Transfer for Continual Learning on Vision-Language Models
  :house:project
• Unveiling Typographic Deceptions: Insights of the Typographic Vulnerability in Large Vision-Language Model
• Introducing Routing Functions to Vision-Language Parameter-Efficient Fine-Tuning with Low-Rank Bottlenecks
• Take A Step Back: Rethinking the Two Stages in Visual Reasoning
  :star:code
• HYDRA: A Hyper Agent for Dynamic Compositional Visual Reasoning
  :star:code视觉推理
• Mind the Interference: Retaining Pre-trained Knowledge in Parameter Efficient Continual Learning of Vision-Language Models
  :star:code
• An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models
  :star:code
• Improving Vision and Language Concepts Understanding with Multimodal Counterfactual Samples
  :star:code
• Reflective Instruction Tuning: Mitigating Hallucinations in Large Vision-Language Models
  :star:code
• SDPT: Synchronous Dual Prompt Tuning for Fusion-based Visual-Language Pre-trained Models
  :star:code
• Robust Calibration of Large Vision-Language Adapters
  :star:code
• BEAF: Observing BEfore-AFter Changes to Evaluate Hallucination in Vision-language Models
  :star:code
• CLIP-DPO: Vision-Language Models as a Source of Preference for Fixing Hallucinations in LVLMs
• MyVLM: Personalizing VLMs for User-Specific Queries
  :star:code
• BRAVE: Broadening the visual encoding of vision-language models
  :house:project
• IVTP: Instruction-guided Visual Token Pruning for Large Vision-Language Models
• ShareGPT4V: Improving Large Multi-Modal Models with Better Captions
  :star:code
  :house:project
• The First to Know: How Token Distributions Reveal Hidden Knowledge in Large Vision-Language Models?
  :star:code
• Safe-CLIP: Removing NSFW Concepts from Vision-and-Language Models
  :star:code
• Adapt without Forgetting: Distill Proximity from Dual Teachers in Vision-Language Models
  :star:code
• Exploiting Semantic Reconstruction to Mitigate Hallucinations in Vision-Language Models
• uCAP: An Unsupervised Prompting Method for Vision-Language Models
• Training A Small Emotional Vision Language Model for Visual Art Comprehension
  :star:code
• Understanding Multi-compositional learning in Vision and Language models via Category Theory
  :star:code
• Adversarial Prompt Tuning for Vision-Language Models
  :star:code
• Language-Image Pre-training with Long Captions
  :star:code
• CoReS: Orchestrating the Dance of Reasoning and Segmentation
  :star:code
  :house:project
• Attention Prompting on Image for Large Vision-Language Models
  :star:code
• SILC: Improving Vision Language Pretraining with Self-Distillation
• SCLIP: Rethinking Self-Attention for Dense Vision-Language Inference
  :star:code
• AdaGlimpse: Active Visual Exploration with Arbitrary Glimpse Position and Scale
  :star:code
• Video-Language
  • Meta-optimized Angular Margin Contrastive Framework for Video-Language Representation Learning
  • PiTe: Pixel-Temporal Alignment for Large Video-Language Model
• VLN
  • PRET: Planning with Directed Fidelity Trajectory for Vision and Language Navigation
    :star:code
  • LLM as Copilot for Coarse-grained Vision-and-Language Navigation
  • Frontier-enhanced Topological Memory with Improved Exploration Awareness for Embodied Visual Navigation
    :star:code
• LLM
  • BLINK: Multimodal Large Language Models Can See but Not Perceive
    :house:project
  • Ferret-UI: Grounded Mobile UI Understanding with Multimodal LLMs
  • X-InstructBLIP: A Framework for aligning X-Modal instruction-aware representations to LLMs and Emergent Cross-modal Reasoning
    :star:code
    :house:project
  • X-Former: Unifying Contrastive and Reconstruction Learning for MLLMs
  • Instruction Tuning-free Visual Token Complement for Multimodal LLMs
  • Merlin: Empowering Multimodal LLMs with Foresight Minds
    :house:project
  • Groma: Localized Visual Tokenization for Grounding Multimodal Large Language Models
    :house:project
  • MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?
    :star:code
  • UniCode: Learning a Unified Codebook for Multimodal Large Language Models
  • When Do We Not Need Larger Vision Models?
    :star:code
  • ControlLLM: Augment Language Models with Tools by Searching on Graphs
    :star:code
  • Towards Open-Ended Visual Recognition with Large Language Models
    :star:code
  • SPHINX: A Mixer of Weights, Visual Embeddings and Image Scales for Multi-modal Large Language Models
    :star:code
  • ST-LLM: Large Language Models Are Effective Temporal Learners
    :star:code
  • Dissecting Dissonance: Benchmarking Large Multimodal Models Against Self-Contradictory Instructions
    :house:project
  • How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMs
    :star:code
  • BenchLMM: Benchmarking Cross-style Visual Capability of Large Multimodal Models
    :star:code
  • MoAI: Mixture of All Intelligence for Large Language and Vision Models
    :star:code
    🤗huggingface
  • Paying More Attention to Images: A Training-Free Method for Alleviating Hallucination in LVLMs
    :house:project
  • LLaVA-UHD: an LMM Perceiving any Aspect Ratio and High-Resolution Images
    :star:code
  • Griffon: Spelling out All Object Locations at Any Granularity with Large Language Models
    :star:code
  • LLMGA: Multimodal Large Language Model based Generation Assistant
    :star:code
    :house:project
  • Images are Achilles' Heel of Alignment: Exploiting Visual Vulnerabilities for Jailbreaking Multimodal Large Language Models
    :star:code
  • LLaMA-VID: An Image is Worth 2 Tokens in Large Language Models
    :star:code
  • LLM as Dataset Analyst: Subpopulation Structure Discovery with Large Language Model
    :house:project
  • Strengthening Multimodal Large Language Model with Bootstrapped Preference Optimization
  • MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training
  • LLaVA-Grounding: Grounded Visual Chat with Large Multimodal Models
    :star:code
  • ShapeLLM: Universal 3D Object Understanding for Embodied Interaction
    :star:code
  • Making Large Language Models Better Planners with Reasoning-Decision Alignment
  • Self-Adapting Large Visual-Language Models to Edge Devices across Visual Modalities
  • Propose, Assess, Search: Harnessing LLMs for Goal-Oriented Planning in Instructional Videos
    :house:project
  • AdaShield: Safeguarding Multimodal Large Language Models from Structure-based Attack via Adaptive Shield Prompting
    :star:code
  • DetToolChain: A New Prompting Paradigm to Unleash Detection Ability of MLLM
    :star:code
  • GENIXER: Empowering Multimodal Large Language Models as a Powerful Data Generator
    :star:code
  • Elysium: Exploring Object-level Perception in Videos through Semantic Integration Using MLLMs
    :star:code
• 视觉定位
  • The Nerfect Match: Exploring NeRF Features for Visual Localization
  • MAD-DR: Map Compression for Visual Localization with Matchness Aware Descriptor Dimension Reduction
• Visual Grounding
  • LoA-Trans: Enhancing Visual Grounding by Location-Aware Transformers
  • Visual Grounding for Object-Level Generalization in Reinforcement Learning
    :star:code
  • ViGoR: Improving Visual Grounding of Large Vision Language Models with Fine-Grained Reward Modeling
• 视觉意图理解
  • Synergy of Sight and Semantics: Visual Intention Understanding with CLIP
    :star:code
• 引用表达理解
  • APL: Anchor-based Prompt Learning for One-stage Weakly Supervised Referring Expression Comprehension
    :star:code
• 视觉语言理解
  • Uni3DL: A Unified Model for 3D Vision-Language Understanding
    :house:project
  • Unifying 3D Vision-Language Understanding via Promptable Queries
    :house:project
  • GRiT: A Generative Region-to-text Transformer for Object Understanding
    :star:code

20.Scene

• LatentEditor: Text Driven Local Editing of 3D Scenes
  :house:project
• RoomTex: Texturing Compositional Indoor Scenes via Iterative Inpainting
  :star:code室内场景
• Forecasting Future Videos from Novel Views via Disentangled 3D Scene Representation
  :star:code
• Compact 3D Scene Representation via Self-Organizing Gaussian Grids
  :star:code
• CARFF: Conditional Auto-encoded Radiance Field for 3D Scene Forecasting
  :house:project
• 场景合成
  • Pyramid Diffusion for Fine 3D Large Scene Generation
    :star:code
    :house:project
    :Thumbsup:西南交大&利兹大学等联合提出金字塔离散扩散模型（PDD），实现了3D户外场景生成的粗到细的策略
  • External Knowledge Enhanced 3D Scene Generation from Sketch3D 场景生成
  • SceneTeller: Language-to-3D Scene Generation
    :star:code
  • Forest2Seq: Revitalizing Order Prior for Sequential Indoor Scene Synthesis
  • Gaussian Grouping: Segment and Edit Anything in 3D Scenes
    :star:code
  • EchoScene: Indoor Scene Generation via Information Echo over Scene Graph Diffusion
    :star:code
  • AnyHome: Open-Vocabulary Large-Scale Indoor Scene Generation with First-Person View Exploration室内场景生成
  • BeyondScene: Higher-Resolution Human-Centric Scene Generation With Pretrained Diffusion
    :house:project
  • The Fabrication of Reality and Fantasy: Scene Generation with LLM-Assisted Prompt Interpretation
    :star:code
  • Language-Driven Physics-Based Scene Synthesis and Editing via Feature Splatting
    :house:project场景合成和编辑
  • WoVoGen: World Volume-aware Diffusion for Controllable Multi-camera Driving Scene Generation
    :star:code驾驶场景生成
• 场景理解
  • N2F2: Hierarchical Scene Understanding with Nested Neural Feature Fields
  • Dense Multimodal Alignment for Open-Vocabulary 3D Scene Understanding
  • SceneVerse: Scaling 3D Vision-Language Learning for Grounded Scene Understanding
    :house:project
  • Shape2Scene: 3D Scene Representation Learning Through Pre-training on Shape Data
    :star:code
  • nuCraft: Crafting High Resolution 3D Semantic Occupancy for Unified 3D Scene Understanding
  • R3DS: Reality-linked 3D Scenes for Panoramic Scene Understanding
    :house:project
  • Open Vocabulary 3D Scene Understanding via Geometry Guided Self-Distillation
    :star:code
  • Agent3D-Zero: An Agent for Zero-shot 3D Understanding
    :house:project
  • MTMamba: Enhancing Multi-Task Dense Scene Understanding by Mamba-Based Decoders
    :star:code密集场景理解
• 语义场景完
  • Hierarchical Temporal Context Learning for Camera-based Semantic Scene Completion
    :star:code
  • Easing 3D Pattern Reasoning with Side-view Features for Semantic Scene Completion
  • Zero-Shot Multi-Object Scene Completion
    :house:project
• 场景图生成
  • OpenPSG: Open-set Panoptic Scene Graph Generation via Large Multimodal Models
    :star:code
  • Semantic Diversity-aware Prototype-based Learning for Unbiased Scene Graph Generation
    :star:code
  • Fine-Grained Scene Graph Generation via Sample-Level Bias Prediction
    :star:code
  • Expanding Scene Graph Boundaries: Fully Open-vocabulary Scene Graph Generation via Visual-Concept Alignment and Retention
    :star:code
    :thumbsup:突破场景图生成的边界：OvSGTR 实现全开放词汇场景图生成
  • A Fair Ranking and New Model for Panoptic Scene Graph Generation
    :house:project
  • Multi-Granularity Sparse Relationship Matrix Prediction Network for End-to-End Scene Graph Generation
    :star:code
  • Towards Scene Graph Anticipation
    :star:code

19.UAV/Remote Sensing/Satellite Image(无人机/遥感/卫星图像)

• Masked Angle-Aware Autoencoder for Remote Sensing Images
  :star:code
• Radiance Field Learners As UAV First-Person Viewers
• Geospecific View Generation - Geometry-Context Aware High-resolution Ground View Inference from Satellite Views
  :house:project卫星视图
• Probabilistic Image-Driven Traffic Modeling via Remote Sensing
• UAV First-Person Viewers Are Radiance Field Learners
  :house:project
• MutDet: Mutually Optimizing Pre-training for Remote Sensing Object Detection
  :star:code
• Towards Natural Language-Guided Drones: GeoText-1652 Benchmark with Spatial Relation Matching
• Free-Viewpoint Video of Outdoor Sports Using a Drone
• Learning Representations of Satellite Images From Metadata Supervision(https://github.com/preligens-lab/satmip)卫星图像
• Multi-scale Cross Distillation for Object Detection in Aerial Images
• LHRS-Bot: Empowering Remote Sensing with VGI-Enhanced Large Multimodal Language Model
  :star:code
• PDT Uav Target Detection Dataset for Pests and Diseases Tree
  :star:code
• Contrastive ground-level image and remote sensing pre-training improves representation learning for natural world imagery
  🤗huggingface遥感
• Toward Open Vocabulary Aerial Object Detection with CLIP-Activated Student-Teacher Learning
  :star:code

18.Automated Driving(自动驾驶)

• Online Vectorized HD Map Construction using Geometry
  :star:code
• MUSES: The Multi-Sensor Semantic Perception Dataset for Driving under Uncertainty
  :star:code
• HENet: Hybrid Encoding for End-to-end Multi-task 3D Perception from Multi-view Cameras
  :star:code
• Continuity Preserving Online CenterLine Graph Learning
• Accelerating Online Mapping and Behavior Prediction via Direct BEV Feature Attention
  :star:code
• RepVF: A Unified Vector Fields Representation for Multi-task 3D Perception
  :star:code
• MapDistill: Boosting Efficient Camera-based HD Map Construction via Camera-LiDAR Fusion Model Distillation
• Generative End-to-End Autonomous Driving
  :star:code
• CARB-Net: Camera-Assisted Radar-Based Network for Vulnerable Road User Detection
  :star:code驾驶
• FipTR: A Simple yet Effective Transformer Framework for Future Instance Prediction in Autonomous Driving
  :star:code
• Mask2Map: Vectorized HD Map Construction Using Bird's Eye View Segmentation Masks
  :star:code驾驶
• Gated Temporal Diffusion for Stochastic Long-Term Dense Anticipation
  :star:code
• CarFormer: Self-Driving with Learned Object-Centric Representations
  :star:code
• Image-to-Lidar Relational Distillation for Autonomous Driving Data
• Modelling Competitive Behaviors in Autonomous Driving Under Generative World Model
  :star:code
• LingoQA: Video Question Answering for Autonomous Driving
  :star:code
• PreSight: Enhancing Autonomous Vehicle Perception with City-Scale NeRF Priors
  :star:code
• VQA-Diff: Exploiting VQA and Diffusion for Zero-Shot Image-to-3D Vehicle Asset Generation in Autonomous Driving
• TCLC-GS: Tightly Coupled LiDAR-Camera Gaussian Splatting for Autonomous Driving
• Learning to Drive via Asymmetric Self-Play
  :house:project
• Embodied Understanding of Driving Scenarios
  :star:code
• Early Anticipation of Driving Maneuvers
  :house:project
• RealGen: Retrieval Augmented Generation for Controllable Traffic Scenarios
  :house:project
• Event-Aided Time-To-Collision Estimation for Autonomous Driving
  :house:project
• Dolphins: Multimodal Language Model for Driving
  :house:project
• PPAD: Iterative Interactions of Prediction and Planning for End-to-end Autonomous Driving
  :star:code
• Asynchronous Large Language Model Enhanced Planner for Autonomous Driving
  :star:code
• Neural Volumetric World Models for Autonomous Driving
• SeFlow: A Self-Supervised Scene Flow Method in Autonomous Driving
  :star:code
• Random Walk on Pixel Manifolds for Anomaly Segmentation of Complex Driving Scenes
  :star:code自动驾驶
• SLEDGE: Synthesizing Driving Environments with Generative Models and Rule-Based Traffic
• I Can't Believe It's Not Scene Flow!
  :star:code场景流
• Safe-Sim: Safety-Critical Closed-Loop Traffic Simulation with Diffusion-Controllable Adversaries
  :house:project交通
• UniM2AE: Multi-modal Masked Autoencoders with Unified 3D Representation for 3D Perception in Autonomous Driving
  :star:code
• DriveDreamer: Towards Real-world-driven World Models for Autonomous Driving
  :house:project
• Think2Drive: Efficient Reinforcement Learning by Thinking with Latent World Model for Autonomous Driving (in CARLA-v2)
  :house:project
• Reason2Drive: Towards Interpretable and Chain-based Reasoning for Autonomous Driving
  :star:code
• Improving Agent Behaviors with RL Fine-tuning for Autonomous Driving
• Lane Graph as Path: Continuity-preserving Path-wise Modeling for Online Lane Graph Construction
  :star:code
• Beyond the Data Imbalance: Employing the Heterogeneous Datasets for Vehicle Maneuver Prediction
  :star:code
• 轨迹预测
  • Learning Semantic Latent Directions for Accurate and Controllable Human Motion Prediction
    :star:code
  • NeRMo: Learning Implicit Neural Representations for 3D Human Motion Prediction
  • CoMusion: Towards Consistent Stochastic Human Motion Prediction via Motion Diffusion人体运动预测
  • Risk-Aware Self-Consistent Imitation Learning for Trajectory Planning in Autonomous Driving
  • Progressive Pretext Task Learning for Human Trajectory Prediction
    :star:code
  • DySeT: a Dynamic Masked Self-distillation Approach for Robust Trajectory Prediction
  • VisionTrap: Vision-Augmented Trajectory Prediction Guided by Textual Descriptions
    :star:code
  • Optimizing Diffusion Models for Joint Trajectory Prediction and Controllable Generation
    :star:code
  • Adaptive Human Trajectory Prediction via Latent Corridors
    :house:project
  • NeuroNCAP: Photorealistic Closed-loop Safety Testing for Autonomous Driving
    :star:code
  • MART: MultiscAle Relational Transformer Networks for Multi-agent Trajectory Prediction
    :star:code
  • Local Occupancy-Enhanced Object Grasping with Multiple Triplanar Projection
  • 车辆轨迹预测
    • UniTraj: A Unified Framework for Scalable Vehicle Trajectory Prediction
      :star:code
• 占据预测
  • VEON: Vocabulary-Enhanced Occupancy Prediction
  • OccWorld: Learning a 3D Occupancy World Model for Autonomous Driving
    :star:code
  • OccGen: Generative Multi-modal 3D Occupancy Prediction for Autonomous Driving
    :house:project
  • Fully Sparse 3D Occupancy Prediction
    :star:code
  • Monocular Occupancy Prediction for Scalable Indoor Scenes
    :star:code
  • ViewFormer: Exploring Spatiotemporal Modeling for Multi-View 3D Occupancy Perception via View-Guided Transformers
    :star:code
  • CVT-Occ: Cost Volume Temporal Fusion for 3D Occupancy Prediction
    :star:code
  • GaussianFormer: Scene as Gaussians for Vision-Based 3D Semantic Occupancy Prediction
    :star:code3D 语义占用预测
• 车道线检测
  • OMR: Occlusion-Aware Memory-Based Refinement for Video Lane Detection
    :star:code
• 车辆监控
  • TrafficNight : An Aerial Multimodal Benchmark For Nighttime Vehicle Surveillance

17.Video

• Stable Video Portraits
  :house:project
• Text-Guided Video Masked Autoencoder
• Multi-Modal Video Dialog State Tracking in the Wild
• Training-free Video Temporal Grounding using Large-scale Pre-trained Models
  :star:code
• Weakly-Supervised Spatio-Temporal Video Grounding with Variational Cross-Modal Alignment
• E3M: Zero-Shot Spatio-Temporal Video Grounding with Expectation-Maximization Multimodal Modulation
  :star:code
• Rethinking Weakly-supervised Video Temporal Grounding From a Game Perspective
• Fast Encoding and Decoding for Implicit Video Representation
  :star:code
  :house:project
• DEVIAS: Learning Disentangled Video Representations of Action and Scene
  :star:code
• VideoStudio: Generating Consistent-Content and Multi-Scene Videos
  :house:project
• VAD
  • Cross-Domain Learning for Video Anomaly Detection with Limited Supervision
  • Learning Anomalies with Normality Prior for Unsupervised Video Anomaly Detection
    :star:code
  • Follow the Rules: Reasoning for Video Anomaly Detection with Large Language Models
    :star:code
  • FedVAD: Enhancing Federated Video Anomaly Detection with GPT-Driven Semantic Distillation
    :star:code
  • Interleaving One-Class and Weakly-Supervised Models with Adaptive Thresholding for Unsupervised Video Anomaly Detection
    :star:code视频异常检测
• 视频摘要
  • Contextual Correspondence Matters: Bidirectional Graph Matching for Video Summarization
• 视频理解
  • VideoMamba: Spatio-Temporal Selective State Space Model
    :star:code
  • VideoMamba: State Space Model for Efficient Video Understanding
    :star:code
  • Goldfish: Vision-Language Understanding of Arbitrarily Long Videos
    :star:code
  • Learning Video Context as Interleaved Multimodal Sequences
    :star:code
  • FunQA: Towards Surprising Video Comprehension
    :house:project
  • Vamos: Versatile Action Models for Video Understanding
    :star:code
    :house:project
  • Towards Neuro-Symbolic Video Understanding
    :star:code
  • LongVLM: Efficient Long Video Understanding via Large Language Models
    :star:code
  • VideoAgent: A Memory-augmented Multimodal Agent for Video Understanding
    :star:code
    :house:project
  • Ex2Eg-MAE: A Framework for Adaptation of Exocentric Video Masked Autoencoders for Egocentric Social Role Understanding
  • VideoAgent: Long-form Video Understanding with Large Language Model as Agent
    🤗huggingface
  • InternVideo2: Scaling Foundation Models for Multimodal Video Understanding
    :star:code
  • Text-Conditioned Resampler For Long Form Video Understanding
• 视频分类
  • Open Vocabulary Multi-Label Video Classification
• 视频解析
  • CoLeaF: A Contrastive-Collaborative Learning Framework for Weakly Supervised Audio-Visual Video Parsing
    :star:code
• 视频帧插值
  • IAM-VFI : Interpolate Any Motion for Video Frame Interpolation with motion complexity map
  • Clearer Frames, Anytime: Resolving Velocity Ambiguity in Video Frame Interpolation
    :house:project
  • TimeLens-XL: Real-time Event-based Video Frame Interpolation with Large Motion
    :house:project
• 视频类增量
  • STSP: Spatial-Temporal Subspace Projection for Video Class-incremental Learning
• 视频抄袭片段定位
  • Self-Supervised Video Copy Localization with Regional Token Representation

16.Medical Image Progress(医学影响处理)

• Adaptive Compressed Sensing with Diffusion-Based Posterior Sampling
  :star:code
• Co-synthesis of Histopathology Nuclei Image-Label Pairs using a Context-Conditioned Joint Diffusion Model
• Identity-Consistent Diffusion Network for Grading Knee Osteoarthritis Progression in Radiographic Imaging
• Multistain Pretraining for Slide Representation Learning in Pathology
  :star:code
• Energy-induced Explicit quantification for Multi-modality MRI fusion
  :star:code
• Brain-ID: Learning Contrast-agnostic Anatomical Representations for Brain Imaging
  :star:code
• CardiacNet: Learning to Reconstruct Abnormalities for Cardiac Disease Assessment from Echocardiogram Videos
  :star:code心脏病评估
• Knowledge-enhanced Visual-Language Pretraining for Computational Pathology
  :star:code
• Effective Lymph Nodes Detection in CT Scans Using Location Debiased Query Selection and Contrastive Query Representation in TransformerCT
• Bridging the Pathology Domain Gap: Efficiently Adapting CLIP for Pathology Image Analysis with Limited Labeled Data病理学图像分析
• Unified Medical Image Pre-training in Language-Guided Common Semantic Space
• Rethinking Deep Unrolled Model for Accelerated MRI Reconstruction
  :star:code
• Style-Extracting Diffusion Models for Semi-Supervised Histopathology Segmentation
  :star:code半监督组织病理学分割
• 组织病理学图像分类
  • Test-Time Stain Adaptation with Diffusion Models for Histopathology Image Classification
• 切片图像分类
  • DGR-MIL: Exploring Diverse Global Representation in Multiple Instance Learning for Whole Slide Image Classification
    :star:code
  • Pathology-knowledge Enhanced Multi-instance Prompt Learning for Few-shot Whole Slide Image Classification
  • Snuffy: Efficient Whole Slide Image Classifier
  • Norma: A Noise Robust Memory-Augmented Framework for Whole Slide Image Classification
    :star:code
  • Attention-Challenging Multiple Instance Learning for Whole Slide Image Classification
    :star:code
• 医学图像分割
  • FairDomain: Achieving Fairness in Cross-Domain Medical Image Segmentation and Classification
    :house:project
    :star:code
  • PMT: Progressive Mean Teacher via Exploring Temporal Consistency for Semi-Supervised Medical Image Segmentation
    :star:code
  • The Devil is in the Statistics: Mitigating and Exploiting Statistics Difference for Generalizable Semi-supervised Medical Image Segmentation
    :star:code
  • Domesticating SAM for Breast Ultrasound Image Segmentation via Spatial-frequency Fusion and Uncertainty Correction
    :star:code
  • Gradient-Aware for Class-Imbalanced Semi-supervised Medical Image Segmentation
    :star:code
  • AnatoMask: Enhancing Medical Image Segmentation with Reconstruction-guided Self-masking
    :star:code
  • Alternate Diverse Teaching for Semi-supervised Medical Image Segmentation
    :star:code
  • I-MedSAM: Implicit Medical Image Segmentation with Segment Anything
    :star:code
  • VP-SAM: Taming Segment Anything Model for Video Polyp Segmentation via Disentanglement and Spatio-temporal Side Network
    :star:code息肉分割
• 医学图像配准
  • Unsupervised Multi-modal Medical Image Registration via Invertible Translation
    :star:code
  • NePhi: Neural Deformation Fields for Approximately Diffeomorphic Medical Image Registration
    :star:code
  • Adaptive Correspondence Scoring for Unsupervised Medical Image Registration
    :house:project
• 医学报告生成
  • Contrastive Learning with Counterfactual Explanations for Radiology Report Generation
  • HERGen: Elevating Radiology Report Generation with Longitudinal Data
  • MedRAT: Unpaired Medical Report Generation via Auxiliary Tasks
    :star:code
• X 光片
  • ChEX: Interactive Localization and Region Description in Chest X-rays
    :star:code胸部 X 光片
  • Plug-and-Play Learned Proximal Trajectory for 3D Sparse-View X-Ray Computed Tomography
    :star:code
  • DiffuX2CT: Diffusion Learning to Reconstruct CT Images from Biplanar X-Rays
• 医学机器人
  • Shape-guided Configuration-aware Learning for Endoscopic-image-based Pose Estimation of Flexible Robotic Instruments
    :house:project
• 生物医学图像
  • ScribblePrompt: Fast and Flexible Interactive Segmentation for Any Biomedical Image
    :star:code
• CT
  • GenerateCT: Text-Conditional Generation of 3D Chest CT Volumes
    :star:code

15.GAN/Image Synthesis(图像生成)

• Diffusion Models as Data Mining Tools
  :star:code
• ProCreate, Don't Reproduce! Propulsive Energy Diffusion for Creative Generation
  :house:project
• Learning to Generate Conditional Tri-plane for 3D-aware Expression Controllable Portrait Animation
  :house:project
• HiEI: A Universal Framework for Generating High-quality Emerging Images from Natural Images
• UDiffText: A Unified Framework for High-quality Text Synthesis in Arbitrary Images via Character-aware Diffusion Models
  :star:code
• Score Distillation Sampling with Learned Manifold Corrective
• CTRLorALTer: Conditional LoRAdapter for Efficient 0-Shot Control & Altering of T2I Models
  :house:project
• EMO: Emote Portrait Alive - Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions
• Inf-DiT: Upsampling any-resolution image with memory-efficient diffusion transformer
  :star:code
• The Lottery Ticket Hypothesis in Denoising: Towards Semantic-Driven Initialization
  :house:project
• MONTAGE: Monitoring Training for Attribution of Generative Diffusion Models
• TP2O: Creative Text Pair-to-Object Generation using Balance Swap-Sampling
• Free-ATM: Harnessing Free Attention Masks for Representation Learning on Diffusion-Generated Images
• Idea2Img: Iterative Self-Refinement with GPT-4V for Automatic Image Design and Generation
  :house:project
• OMG: Occlusion-friendly Personalized Multi-concept Generation in Diffusion Models
  :star:code
• DreamDiffusion: High-Quality EEG-to-Image Generation with Temporal Masked Signal Modeling and CLIP Alignment
• V-Trans4Style: Visual Transition Recommendation for Video Production Style Adaptation视频制作风格适配的视觉转场推荐
• GAN
  • CLR-GAN: Improving GANs Stability and Quality via Consistent Latent Representation and Reconstruction
    :star:code
  • A Closer Look at GAN Priors: Exploiting Intermediate Features for Enhanced Model Inversion Attacks
    :star:code
  • Distilling Diffusion Models into Conditional GANs
  • Exploring Guided Sampling of Conditional GANs
    :star:code
  • Learning 3D-aware GANs from Unposed Images with Template Feature Field
    :house:project
• 扩散
  • Measuring Style Similarity in Diffusion Models
    :star:code
  • Do text-free diffusion models learn discriminative visual representations
  • Iterative Ensemble Training with Anti-Gradient Control for Mitigating Memorization in Diffusion Models
    :star:code
  • ShoeModel: Learning to Wear on the User-specified Shoes via Diffusion Model
  • HiDiffusion: Unlocking Higher-Resolution Creativity and Efficiency in Pretrained Diffusion Models
    :star:code
  • Chains of Diffusion Models
  • To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images ... For Now
    :star:code
  • FontStudio: Shape-Adaptive Diffusion Model for Coherent and Consistent Font Effect Generation
    :house:project
  • Beta-Tuned Timestep Diffusion Model
  • SMooDi: Stylized Motion Diffusion Model
    :house:project
  • Make a Cheap Scaling: A Self-Cascade Diffusion Model for Higher-Resolution Adaptation
    :star:code
  • Surf-D: Generating High-Quality Surfaces of Arbitrary Topologies Using Diffusion Models
    :star:code
    :house:project
  • Implicit Concept Removal of Diffusion Models
    :house:project
  • ZigMa: A DiT-style Zigzag Mamba Diffusion Model
    :star:code
  • ColorPeel: Color Prompt Learning with Diffusion Models via Color and Shape Disentanglement
    :star:code
    :house:project
  • Timestep-Aware Correction for Quantized Diffusion Models
  • Shapefusion: 3D localized human diffusion models
    :house:project
  • MVDD: Multi-View Depth Diffusion Models
  • SwiftBrush v2: Make Your One-step Diffusion Model Better Than Its Teacher
    :star:code
    :house:project
    :tv:video
  • Layout-Corrector: Alleviating Layout Sticking Phenomenon in Discrete Diffusion Model
    :star:code
  • Compensation Sampling for Improved Convergence in Diffusion Models
    :star:code
  • ConceptExpress: Harnessing Diffusion Models for Single-image Unsupervised Concept Extraction
    :star:code
    :star:code
  • Self-Guided Generation of Minority Samples Using Diffusion Models
    :star:code
• Zero-Shot Adaptation for Approximate Posterior Sampling of Diffusion Models in Inverse Problems
  :star:code
• LogoSticker: Inserting Logos into Diffusion Models for Customized Generation
  :star:code
• 纹理合成
  • VCD-Texture: Variance Alignment based 3D-2D Co-Denoising for Text-Guided Texturing
  • TexGen: Text-Guided 3D Texture Generation with Multi-view Sampling and Resampling
    :star:code
• 图像合成
  • Editable Image Elements for Controllable Synthesis
    :house:project
  • Assessing Sample Quality via the Latent Space of Generative Models
    :star:code
  • SCP-Diff: Photo-Realistic Semantic Image Synthesis with Spatial-Categorical Joint Prior
    :house:project
  • Zero-shot Text-guided Infinite Image Synthesis with LLM guidance
  • $\infty$-Brush: Controllable Large Image Synthesis with Diffusion Models in Infinite Dimensions
    :star:code
    :star:code
  • EpipolarGAN: Omnidirectional Image Synthesis with Explicit Camera Control
  • LayerDiff: Exploring Text-guided Multi-layered Composable Image Synthesis via Layer-Collaborative Diffusion Model
  • Layered Rendering Diffusion Model for Controllable Zero-Shot Image Synthesis
  • Label-free Neural Semantic Image Synthesis
  • Improving image synthesis with diffusion-negative sampling
  • SCP-Diff: Spatial-Categorical Joint Prior for Diffusion Based Semantic Image Synthesis
    :house:project
  • 2S-ODIS: Two-Stage Omni-Directional Image Synthesis by Geometric Distortion Correction
    :star:code
  • Rejection Sampling IMLE: Designing Priors for Better Few-Shot Image Synthesis
  • FouriScale: A Frequency Perspective on Training-Free High-Resolution Image Synthesis
    :star:code
• 图像生成
  • MoMA: Multimodal LLM Adapter for Fast Personalized Image Generation
    :star:code
    :house:project
  • Few-Shot Image Generation by Conditional Relaxing Diffusion Inversion
  • Context Diffusion: In-Context Aware Image Generation
    :house:project
  • Few-shot Defect Image Generation based on Consistency Modeling
    :star:code
  • Linearly Controllable GAN: Unsupervised Feature Categorization and Decomposition for Image Generation and Manipulation
  • PanoFree: Tuning-Free Holistic Multi-view Image Generation with Cross-view Self-Guidance
  • AdaNAT: Exploring Adaptive Policy for Token-Based Image Generation
    :star:code
  • AccDiffusion: An Accurate Method for Higher-Resolution Image Generation
    :house:project
    :thumbsup:成功地进行无重复高分辨率的图像生成
  • Towards Reliable Advertising Image Generation Using Human Feedback
    :star:code
  • StoryImager: A Unified and Efficient Framework for Coherent Story Visualization and Completion
    :star:code
  • Model-agnostic Origin Attribution of Generated Images with Few-shot Examples
  • Improving Geo-diversity of Generated Images with Contextualized Vendi Score Guidance
    :star:code
  • Tuning-Free Image Customization with Image and Text Guidance
    :house:project
  • Collaborative Control for Geometry-Conditioned PBR Image Generation
    :house:project
  • DiffiT: Diffusion Vision Transformers for Image Generation
    :star:code
  • MultiGen: Zero-shot Image Generation from Multi-modal Prompts
  • Accelerating Image Generation with Sub-path Linear Approximation Model
    :star:code
• 视频生成
  • FreeInit: Bridging Initialization Gap in Video Diffusion Models
    :star:code
    :house:project
  • HARIVO: Harnessing Text-to-Image Models for Video Generation
    :house:project
  • SignGen: End-to-End Sign Language Video Generation with Latent Diffusion
    :star:code
  • DragAnything: Motion Control for Anything using Entity Representation
    :star:code
    :house:project
  • Physics-Based Interaction with 3D Objects via Video Generation
    :star:code
    :house:project
  • DrivingDiffusion: Layout-Guided Multi-View Driving Scenarios Video Generation with Latent Diffusion Model
    :house:project
  • Photorealistic Video Generation with Diffusion Models
    :house:project
  • DynamiCrafter: Animating Open-domain Images with Video Diffusion Priors
    :star:code
  • PoseCrafter: One-Shot Personalized Video Synthesis Following Flexible Pose Control
    :house:project
  • MoVideo: Motion-Aware Video Generation with Diffusion Models
    :house:project
  • IDOL: Unified Dual-Modal Latent Diffusion for Human-Centric Joint Video-Depth Generation
    :star:code
  • MagDiff: Multi-Alignment Diffusion for High-Fidelity Video Generation and Editing
  • 文本-视频质量评估
    • Subjective-Aligned Dateset and Metric for Text-to-Video Quality Assessment
      :star:code
• 视频编辑
  • DragVideo: Interactive Drag-style Video Editing
    :house:project
  • Video Editing via Factorized Diffusion Distillation
  • SAVE: Protagonist Diversification with Structure Agnostic Video Editing
    :house:project
  • DNI: Dilutional Noise Initialization for Diffusion Video Editing
  • DeCo: Decoupled Human-Centered Diffusion Video Editing with Motion Consistency
  • DreamMotion: Space-Time Self-Similar Score Distillation for Zero-Shot Video Editing
    :house:project
  • Videoshop: Localized Semantic Video Editing with Noise-Extrapolated Diffusion Inversion
    :star:code
  • Object-Centric Diffusion for Efficient Video Editing
    :house:project
• 图像编辑
  • ObjectAdd: Adding Objects into Image via a Training-Free Diffusion Modification Fashion
  • Guide-and-Rescale: Self-Guidance Mechanism for Effective Tuning-Free Real Image Editing
    🤗huggingface
  • COMPOSE: Comprehensive Portrait Shadow Editing
  • Fast Diffusion-Based Counterfactuals for Shortcut Removal and Generation编辑
  • FreeDiff: Progressive Frequency Truncation for Image Editing with Diffusion Models
    :star:code
  • ByteEdit: Boost, Comply and Accelerate Generative Image Editing
  • RegionDrag: Fast Region-Based Image Editing with Diffusion Models
    :star:code
  • 3DEgo: 3D Editing on the Go!
    :star:code
  • View-Consistent 3D Editing with Gaussian Splatting
    :house:project
  • Diffusion Models are Geometry Critics: Single Image 3D Editing Using Pre-Trained Diffusion Priors
    :house:project
  • Watch Your Steps: Local Image and Scene Editing by Text Instructions
    :house:project
  • Chat-Edit-3D: Interactive 3D Scene Editing via Text Prompts
    :house:project
  • Free-Editor: Zero-shot Text-driven 3D Scene Editing
    :house:project
  • InstructGIE: Towards Generalizable Image Editing
  • Lazy Diffusion Transformer for Interactive Image Editing
    :house:project
  • DATENeRF: Depth-Aware Text-based Editing of NeRFs
    :star:code
    :house:project
  • TurboEdit: Real-time text-based disentangled real image editing
  • DGE: Direct Gaussian 3D Editing by Consistent Multi-view Editing
    :star:code
  • StableDrag: Stable Dragging for Point-based Image Editing
  • ST-LDM: A Universal Framework for Text-Grounded Object Generation in Real Images图像编辑
  • SwapAnything: Enabling Arbitrary Object Swapping in Personalized Image Editing
    :house:project
  • Source Prompt Disentangled Inversion for Boosting Image Editability with Diffusion Models
    :star:code
  • Robust-Wide: Robust Watermarking against Instruction-driven Image Editing
    :star:code
  • FlexiEdit: Frequency-Aware Latent Refinement for Enhanced Non-Rigid Editing
  • Eta Inversion: Designing an Optimal Eta Function for Diffusion-based Real Image Editing
    :star:code
  • RadEdit: stress-testing biomedical vision models via diffusion image editing
  • Responsible Visual Editing
    :star:code
  • 3D-GOI: 3D GAN Omni-Inversion for Multifaceted and Multi-object Editing
    :house:project
  • Thinking Outside the BBox: Unconstrained Generative Object Compositing物体合成
  • EditShield: Protecting Unauthorized Image Editing by Instruction-guided Diffusion Models
• 图像-视频
  • Rethinking Image-to-Video Adaptation: An Object-centric Perspective
  • R2-Tuning: Efficient Image-to-Video Transfer Learning for Video Temporal Grounding
    :star:code
  • PhysGen: Rigid-Body Physics-Grounded Image-to-Video Generation
    :star:code
  • ZeroI2V: Zero-Cost Adaptation of Pre-Trained Transformers from Image to Video
• 文本-视频
  • E.T. the Exceptional Trajectories: Text-to-camera-trajectory generation with character awareness
    :house:project
  • WAVE: Warping DDIM Inversion Features for Zero-shot Text-to-Video Editing
    :house:project
  • MotionDirector: Motion Customization of Text-to-Video Diffusion Models
    :star:code
    :house:project
  • Factorizing Text-to-Video Generation by Explicit Image Conditioning
    🤗huggingface
  • SparseCtrl: Adding Sparse Controls to Text-to-Video Diffusion Models
    :house:project
  • MEVG: Multi-event Video Generation with Text-to-Video Models
    :house:project
  • Customize-A-Video: One-Shot Motion Customization of Text-to-Video Diffusion Models
    :house:project
  • xGen-VideoSyn-1: High-fidelity Text-to-Video Synthesis with Compressed Representations
• 文本-3D
  • Diverse Text-to-3D Synthesis with Augmented Text Embedding
    :star:code
    :house:project
  • LATTE3D: Large-scale Amortized Text-To-Enhanced3D Synthesis
    :house:project
  • DreamMesh: Jointly Manipulating and Texturing Triangle Meshes for Text-to-3D Generation
    :star:code
  • DreamReward: Aligning Human Preference in Text-to-3D Generation
    :house:project
  • DreamDissector: Learning Disentangled Text-to-3D Generation from 2D Diffusion Priors
    :star:code
  • DreamReward: Text-to-3D Generation with Human Preference
    :house:project
  • GVGEN: Text-to-3D Generation with Volumetric Representation
    :star:code
  • WordRobe: Text-Guided Generation of Textured 3D Garments
    :house:project
  • UniDream: Unifying Diffusion Priors for Relightable Text-to-3D Generation
    :star:code
    :house:project
  • ScaleDreamer: Scalable Text-to-3D Synthesis with Asynchronous Score Distillation
    :star:code
  • CRM: Single Image to 3D Textured Mesh with Convolutional Reconstruction Model
    :house:project
  • Repaint123: Fast and High-quality One Image to 3D Generation with Progressive Controllable Repainting
    :house:project
  • VividDreamer: Invariant Score Distillation for Hyper-Realistic Text-to-3D Generation
  • DreamScene: 3D Gaussian-based Text-to-3D Scene Generation via Formation Pattern Sampling
    :star:code
  • HiFi-123: Towards High-fidelity One Image to 3D Content Generation
    :house:project
  • JointDreamer: Ensuring Geometry Consistency and Text Congruence in Text-to-3D Generation via Joint Score Distillation
  • Connecting Consistency Distillation to Score Distillation for Text-to-3D Generation
    :star:code
  • TPA3D: Triplane Attention for Fast Text-to-3D Generation
    :house:project
  • DreamView: Injecting View-specific Text Guidance into Text-to-3D Generation
    :star:code
• 文本-图像
  • [Navigating Text-to-lmage Generative Bias acrossIndic Languages]
  • Bridging Different Language Models and Generative Vision Models for Text-to-Image Generation
    :star:code
    :house:project
  • MobileDiffusion: Instant Text-to-Image Generation on Mobile Devices
  • PreciseControl: Enhancing Text-To-Image Diffusion Models with Fine-Grained Attribute Control
    :house:project
  • PEA-Diffusion: Parameter-Efficient Adapter with Knowledge Distillation in non-English Text-to-Image Generation
    :star:code
  • Diffusion Soup: Model Merging for Text-to-Image Diffusion Models
  • IMMA: Immunizing text-to-image Models against Malicious Adaptation
    :star:code
  • Receler: Reliable Concept Erasing of Text-to-Image Diffusion Models via Lightweight Erasers
    :house:project
  • ControlNet-XS: Rethinking the Control of Text-to-Image Diffusion Models as Feedback-Control Systems
  • Skews in the Phenomenon Space Hinder Generalization in Text-to-Image Generation
    :star:code
  • Enhancing Semantic Fidelity in Text-to-Image Synthesis: Attention Regulation in Diffusion Models
    :star:code
  • Post-training Quantization with Progressive Calibration and Activation Relaxing for Text-to-Image Diffusion Models
  • Textual-Visual Logic Challenge: Understanding and Reasoning in Text-to-Image Generation
    :star:code
  • Navigating Text-to-Image Generative Bias across Indic Languages
    :house:project
  • Safeguard Text-to-Image Diffusion Models with Human Feedback Inversion
    :star:code
  • Diff-Tracker: Text-to-Image Diffusion Models are Unsupervised Trackers
  • DreamDrone: Text-to-Image Diffusion Models are Zero-shot Perpetual View Generators
    :house:project
  • Harnessing Text-to-Image Diffusion Models for Category-Agnostic Pose Estimation
  • MaxFusion: Plug&Play Multi-Modal Generation in Text-to-Image Diffusion Models
    :star:code
    :house:project
  • Deep Reward Supervisions for Tuning Text-to-Image Diffusion Models
  • R.A.C.E.: Robust Adversarial Concept Erasure for Secure Text-to-Image Diffusion Model
    :star:code
  • MixDQ: Memory-Efficient Few-Step Text-to-Image Diffusion Models with Metric-Decoupled Mixed Precision Quantization
    :star:code
    :house:project
  • ReCON: Training-Free Acceleration for Text-to-Image Synthesis with Retrieval of Concept Prompt Trajectories
    :house:project
  • LCM-Lookahead for Encoder-based Text-to-Image Personalization
    :star:code
    :house:project
  • Lego: Learning to Disentangle and Invert Personalized Concepts Beyond Object Appearance in Text-to-Image Diffusion Models
  • Reliable and Efficient Concept Erasure of Text-to-Image Diffusion Models
    :star:code
  • Exploring Phrase-Level Grounding with Text-to-Image Diffusion Model
    :star:code
    :thumbsup:DiffPNG实现了最佳的性能，证明了T2I扩散模型在短语级理解视觉内容的能力
  • T2IShield: Defending Against Backdoors on Text-to-Image Diffusion Models
    :star:code
  • Be Yourself: Bounded Attention for Multi-Subject Text-to-Image Generation
    :house:project
  • Latent Guard: a Safety Framework for Text-to-image Generation
    :star:code
  • Getting it Right: Improving Spatial Consistency in Text-to-Image Models
    :star:code
    :house:project
  • Unveiling and Mitigating Memorization in Text-to-image Diffusion Models through Cross Attention
    :star:code
  • Powerful and Flexible: Personalized Text-to-Image Generation via Reinforcement Learning
    :star:code
  • PanGu-Draw: Advancing Resource-Efficient Text-to-Image Synthesis with Time-Decoupled Training and Reusable Coop-Diffusion
    :house:project
  • Text-Anchored Score Composition: Tackling Condition Misalignment in Text-to-Image Diffusion Models
    :star:code
  • MasterWeaver: Taming Editability and Face Identity for Personalized Text-to-Image Generation
    :star:code
  • Parrot: Pareto-optimal Multi-Reward Reinforcement Learning Framework for Text-to-Image Generation
  • ProTIP: Probabilistic Robustness Verification on Text-to-Image Diffusion Models against Stochastic Perturbation
  • PixArt-Sigma: Weak-to-Strong Training of Diffusion Transformer for 4K Text-to-Image Generation
  • AnyControl: Create Your Artwork with Versatile Control on Text-to-Image Generation
    :star:code
    :house:project
  • CogView3: Finer and Faster Text-to-Image Generation via Relay Diffusion
    :star:code
  • Object-Conditioned Energy-Based Attention Map Alignment in Text-to-Image Diffusion Models
  • SpeedUpNet: A Plug-and-Play Adapter Network for Accelerating Text-to-Image Diffusion Models
    :house:project
  • TIBET: Identifying and Evaluating Biases in Text-to-Image Generative Models
    :house:project
  • An Empirical Study and Analysis of Text-to-Image Generation Using Large Language Model-Powered Textual Representation
  • Adversarial Robustification via Text-to-Image Diffusion Models
    :star:code
  • Stable Preference: Redefining training paradigm of human preference model for Text-to-Image Synthesis
  • Lost in Translation: Latent Concept Misalignment in Text-to-Image Diffusion Models
    :house:project
  • Text-to-Sticker: Style Tailoring Latent Diffusion Models for Human Expression
• 图像-文本
  • Evaluating Text-to-Visual Generation with Image-to-Text Generation
    :house:project
  • Eyes Closed, Safety On: Protecting Multimodal LLMs via Image-to-Text Transformation
    :house:project
• 文本-视频对齐
  • Learning to Localize Actions in Instructional Videos with LLM-Based Multi-Pathway Text-Video Alignment
• 图像-文本对齐
  • Removing Distributional Discrepancies in Captions Improves Image-Text Alignment
    :house:project
  • TTD: Text-Tag Self-Distillation Enhancing Image-Text Alignment in CLIP to Alleviate Single Tag Bias
    :star:code
• 图像-文本
  • Emergent Visual-Semantic Hierarchies in Image-Text Representations
    :star:code
    :house:project
  • Mismatch Quest: Visual and Textual Feedback for Image-Text Misalignment
    :star:code
    :house:project
• 3D(内容)生成
  • Make-Your-3D: Fast and Consistent Subject-Driven 3D Content Generation
    :house:project
  • LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation
    :house:project
  • LN3Diff: Scalable Latent Neural Fields Diffusion for Speedy 3D Generation
    :house:project
  • SV3D: Novel Multi-view Synthesis and 3D Generation from a Single Image using Latent Video Diffusion
    :house:project
  • VFusion3D: Learning Scalable 3D Generative Models from Video Diffusion Models
    :house:project
  • Compress3D: a Compressed Latent Space for 3D Generation from a Single Image
  • AnimatableDreamer: Text-Guided Non-rigid 3D Model Generation and Reconstruction with Canonical Score Distillation
    :house:project
  • Learn to Optimize Denoising Scores: A Unified and Improved Diffusion Prior for 3D Generation
    :house:project
• 视觉文本渲染
  • Glyph-ByT5: A Customized Text Encoder for Accurate Visual Text Rendering
    :house:project
• GIF 生成
  • Pix2Gif: Motion-Guided Diffusion for GIF Generation
    :house:project
• 布局生成
  • COHO: Context-Sensitive City-Scale Hierarchical Urban Layout Generation
    :star:code
  • Self-training Room Layout Estimation via Geometry-aware Ray-casting
  • LayoutFlow: Flow Matching for Layout Generation
    :house:project
• 布局-图像
  • Training-free Composite Scene Generation for Layout-to-Image Synthesis
    :star:code
• 图像-图像翻译
  • Every Pixel Has its Moments: Ultra-High-Resolution Unpaired Image-to-Image Translation via Dense Normalization
    :star:code
  • Diffusion-Based Image-to-Image Translation by Noise Correction via Prompt Interpolation
  • EBDM: Exemplar-guided Image Translation with Brownian-bridge Diffusion Models
• 图像翻译
  • Tackling Structural Hallucination in Image Translation with Local Diffusion
    :star:code
• Text-to-4D
  • TC4D: Trajectory-Conditioned Text-to-4D Generation
    :star:code
    :house:project
• Video-to-4D
  • SC4D: Sparse-Controlled Video-to-4D Generation and Motion Transfer
    :star:code
    :house:project
• 网页设计
  • WebRPG: Automatic Web Rendering Parameters Generation for Visual Presentation
    :star:code
• Text-to-Garment
  • GarmentAligner: Text-to-Garment Generation via Retrieval-augmented Multi-level Corrections
• 图像风格化
  • StyleTokenizer: Defining Image Style by a Single Instance for Controlling Diffusion Models
    :star:code
  • ZipLoRA: Any Subject in Any Style by Effectively Merging LoRAs
    :star:code
    :house:project风格
  • InstaStyle: Inversion Noise of a Stylized Image is Secretly a Style Adviser
    :star:code风格化
  • StyleCity: Large-Scale 3D Urban Scenes Stylization
    :house:project城市场景风格化
  • Scene-Conditional 3D Object Stylization and Composition
• 图像矢量化
  • Segmentation-guided Layer-wise Image Vectorization with Gradient Fills
• 视频拼接
  • Eliminating Warping Shakes for Unsupervised Online Video Stitching
    :star:code
• 文本到相机轨迹生成
  • E.T. the Exceptional Trajectory: Text-to-camera-trajectory generation with character awareness
    :house:project
• 文本到 3D 场景
  • DreamScene360: Unconstrained Text-to-3D Scene Generation with Panoramic Gaussian Splatting
    :house:project
• 身份保留的个性化
  • Infinite-ID: Identity-preserved Personalization via ID-semantics Decoupling Paradigm
    :house:project
  • ComFusion: Enhancing Personalized Generation by Instance-Scene Compositing and Fusion
• 主题驱动生成
  • HybridBooth: Hybrid Prompt Inversion for Efficient Subject-Driven Generation
• 风格内容分离
  • Implicit Style-Content Separation using B-LoRA
    :star:code
    :house:project
• 文本生成多运动
  • M2D2M: Multi-Motion Generation from Text with Discrete Diffusion Models
• 文本驱动的3D编辑
  • GaussCtrl: Multi-View Consistent Text-Driven 3D Gaussian Splatting Editing
    :star:code文本驱动的 3D 高斯泼溅编辑
• 图像插值
  • DreamMover: Leveraging the Prior of Diffusion Models for Image Interpolation with Large Motion
    :star:code
• 图像合成
  • FreeCompose: Generic Zero-Shot Image Composition with Diffusion Prior
    :star:code零样本图像合成
• 图像动画
  • LivePhoto: Real Image Animation with Text-guided Motion Control
    :star:code文本引导运动控制的真实图像动画
  • MOFA-Video: Controllable Image Animation via Generative Motion Field Adaptions in Frozen Image-to-Video Diffusion Model
    :star:code
    :house:project
  • ZoLA: Zero-Shot Creative Long Animation Generation with Short Video Model
    :house:project
  • Champ: Controllable and Consistent Human Image Animation with 3D Parametric Guidance
    :star:code人体图像动画
  • TCAN: Animating Human Images with Temporally Consistent Pose Guidance using Diffusion Models
    :house:project
• 集体照合成
  • AddMe: Zero-shot Group-photo Synthesis by Inserting People into Scenes集体照合成
• 图像裁剪
  • Mitigating Perspective Distortion-induced Shape Ambiguity in Image Crops
    :house:project

14.Image Captioning(图像/视频字幕)

• DECap: Towards Generalized Explicit Caption Editing via Diffusion Mechanism
• ControlCap: Controllable Region-level Captioning
  :star:code字幕
• MarineInst: A Foundation Model for Marine Image Analysis with Instance Visual Description
  :star:code视觉描述
• Distractors-Immune Representation Learning with Cross-modal Contrastive Regularization for Change Captioning
  :star:code
• CIC-BART-SSA: Controllable Image Captioning with Structured Semantic Augmentation
  :star:code
• Controllable Contextualized Image Captioning: Directing the Visual Narrative through User-Defined Highlights
  :star:code
• BRIDGE: Bridging Gaps in Image Captioning Evaluation with Stronger Visual Cues
  :star:code
• View Selection for 3D Captioning via Diffusion Ranking
  :star:code
• Unleashing Text-to-Image Diffusion Prior for Zero-Shot Image Captioning
  :house:project
• HiFi-Score: Fine-grained Image Description Evaluation with Hierarchical Parsing Graphs细粒度图像描述
• 视频字幕
  • HowToCaption: Prompting LLMs to Transform Video Annotations at Scale
    :star:code
• 密集字幕
  • TOD3Cap: Towards 3D Dense Captioning in Outdoor Scenes
    :star:code
  • Bi-directional Contextual Attention for 3D Dense Captioning

13.Image/video Compression(图像/视频压缩)

• SAH-SCI: Self-Supervised Adapter for Efficient Hyperspectral Snapshot Compressive Imaging
• Adaptive Selection of Sampling-Reconstruction in Fourier Compressed Sensing
• Image Compression for Machine and Human Vision with Spatial-Frequency Adaptation
  :star:code
• Bidirectional Stereo Image Compression with Cross-Dimensional Entropy Model
• Rate-Distortion-Cognition Controllable Versatile Neural Image Compression
• BaSIC: BayesNet Structure Learning for Computational Scalable Neural Image Compression
  :star:code
• Region-Adaptive Transform with Segmentation Prior for Image Compression
  :star:code
• EGIC: Enhanced Low-Bit-Rate Generative Image Compression Guided by Semantic Segmentation
• Lagrangian Hashing for Compressed Neural Field Representations
  :house:project
• Latent Diffusion Prior Enhanced Deep Unfolding for Snapshot Spectral Compressive Imaging
  :star:code快照光谱压缩
• Image Compression for Machine and Human Vision With Spatial-Frequency Adaptation
  :star:code
• Learned HDR Image Compression for Perceptually Optimal Storage and Display
  :star:code
• WeConvene: Learned Image Compression with Wavelet-Domain Convolution and Entropy Model
• Lossy Image Compression with Foundation Diffusion Models
• A Unified Image Compression Method for Human Perception and Multiple Vision Tasks
• 视频压缩
  • Hierarchical Separable Video Transformer for Snapshot Compressive Imaging
    :star:code
  • A Simple Low-bit Quantization Framework for Video Snapshot Compressive Imaging
    :star:code
  • Free-VSC: Free Semantics from Visual Foundation Models for Unsupervised Video Semantic Compression
  • Long-term Temporal Context Gathering for Neural Video Compression
  • Learned Rate Control for Frame-Level Adaptive Neural Video Compression via Dynamic Neural Network
• 视频解码
  • Enhancing Cross-Subject fMRI-to-Video Decoding with Global-Local Functional Alignment视频解码
• 快照光谱成像
  • Efficient Snapshot Spectral Imaging: Calibration-Free Parallel Structure with Aperture Diffraction Fusion
• 运动估计
  • Leveraging scale- and orientation-covariant features for planar motion estimation
    :star:code视频预测编码-运动估计

12.Image Retrieval(图像检索)

• RGNet: A Unified Clip Retrieval and Grounding Network for Long Videos
  :star:code
  :house:project
• AMES: Asymmetric and Memory-Efficient Similarity Estimation for Instance-level Retrieval
  :star:code
• IRGen: Generative Modeling for Image Retrieval
  :star:code
• FastCAD: Real-Time CAD Retrieval and Alignment from Scans and Videos
• Spherical Linear Interpolation and Text-Anchoring for Zero-shot Composed Image Retrieval
  :star:code
• FreestyleRet: Retrieving Images from Style-Diversified Queries
  :star:code
• 基于草图的图像检索
  • Freeview Sketching: View-Aware Fine-Grained Sketch-Based Image Retrieval
  • Elevating All Zero-Shot Sketch-Based Image Retrieval Through Multimodal Prompt Learning
    :star:code
• 视频-文本检索
  • EA-VTR: Event-Aware Video-Text Retrieval
  • Rethinking Video-Text Understanding: Retrieval from Counterfactually Augmented Data
    :star:code
  • KDProR: A Knowledge-Decoupling Probabilistic Framework for Video-Text Retrieval
• 图像-文本检索
  • Object-Aware Query Perturbation for Cross-Modal Image-Text Retrieval
• 视频检索
  • EgoCVR: An Egocentric Benchmark for Fine-Grained Composed Video Retrieval
    :star:code
  • Uncertainty-aware sign language video retrieval with probability distribution modeling
    :star:code
• 近邻搜索
  • Group Testing for Accurate and Efficient Range-Based Near Neighbor Search for Plagiarism Detection

11.Image Segmentation(图像分割)

• Occlusion-Aware Seamless Segmentation
  :star:code
• SegVG: Transferring Object Bounding Box to Segmentation for Visual Grounding
  :star:code
  :Thumbsup:视觉定位新SOTA！SegVG：将视觉定位的目标边界框转化为分割信号（已开源）
• Segment, Lift and Fit: Automatic 3D Shape Labeling from 2D Prompts
• Open-Vocabulary SAM: Segment and Recognize Twenty-thousand Classes Interactively
  :star:code
  :house:project
• Segment and Recognize Anything at Any Granularity
  :star:code
• Enriching Information and Preserving Semantic Consistency in Expanding Curvilinear Object Segmentation Datasets
  :star:code
• Semi-supervised Segmentation of Histopathology Images with Noise-Aware Topological Consistency
  :star:code
• CoPT: Unsupervised Domain Adaptive Segmentation using Domain-Agnostic Text Embeddings
  :star:code
• From Pixels to Objects: A Hierarchical Approach for Part and Object Segmentation Using Local and Global Aggregation
• CC-SAM: Enhancing SAM with Cross-feature Attention and Context for Ultrasound Image Segmentation
• Unsupervised Moving Object Segmentation with Atmospheric Turbulence
• Lite-SAM Is Actually What You Need for Segment Everything
• Textual Query-Driven Mask Transformer for Domain Generalized Segmentation
  :star:code
• Can Textual Semantics Mitigate Sounding Object Segmentation Preference?
  :star:code
• RAPiD-Seg: Range-Aware Pointwise Distance Distribution Networks for 3D LiDAR Segmentation
  :star:code
• SPIN: Hierarchical Segmentation with Subpart Granularity in Natural Images
  :star:code
• CC-SAM: SAM with Cross-feature Attention and Context for Ultrasound Image Segmentation
• FlashSplat: 2D to 3D Gaussian Splatting Segmentation Solved Optimally
  :star:code
• SegGen: Supercharging Segmentation Models with Text2Mask and Mask2Img Synthesis
  :house:project
• PQ-SAM: Post-training Quantization for Segment Anything Model
• Active Coarse-to-Fine Segmentation of Moveable Parts from Real Images
  :house:project
• LiteSAM is Actually what you Need for segment Everything
• SEGIC: Unleashing the Emergent Correspondence for In-Context Segmentation
  :star:code
• A Semantic Space is Worth 256 Language Descriptions: Make Stronger Segmentation Models with Descriptive Properties
  :star:code
• Placing Objects in Context via Inpainting for Out-of-distribution Segmentation
  :star:code
• Rethinking and Improving Visual Prompt Selection for In-Context Learning Segmentation Framework
  :star:code
• Better Call SAL: Towards Learning to Segment Anything in Lidar
  :star:code
• 抠图
  • DiffuMatting: Synthesizing Arbitrary Objects with Matting-level Annotation
    :star:code
    :house:project
  • Diffusion for Natural Image Matting
    :star:code
  • COIN-Matting: Confounder Intervention for Image Matting
• 3D分割
  • Bayesian Self-Training for Semi-Supervised 3D Segmentation
  • Segment3D: Learning Fine-Grained Class-Agnostic 3D Segmentation without Manual Labels
    :house:project
  • EgoLifter: Open-world 3D Segmentation for Egocentric Perception
    :house:project
    🤗huggingface
  • View-Consistent Hierarchical 3D Segmentation Using Ultrametric Feature Fields
    :star:code
• 视频分割
  • General and Task-Oriented Video Segmentation
    :star:code
  • DVIS-DAQ: Improving Video Segmentation via Dynamic Anchor Queries
    :star:code
    :house:project
• 实例分割
  • Unleashing the Power of Prompt-driven Nucleus Instance Segmentation
    :star:code
  • 3D实例分割
    • Part2Object: Hierarchical Unsupervised 3D Instance Segmentation
      :star:code
    • SAM-guided Graph Cut for 3D Instance Segmentation
      :star:code
      :house:project
    • Continual Learning and Unknown Object Discovery in 3D Scenes via Self-Distillation
      :star:code实例分割
    • OpenIns3D: Snap and Lookup for 3D Open-vocabulary Instance Segmentation
      :star:code
  • 无监督实例分割
    • ProMerge: Prompt and Merge for Unsupervised Instance Segmentation
  • 开发世界实例分割
    • SOS: Segment Object System for Open-World Instance Segmentation With Object Priors
      :star:code
• 全景分割
  • Open Panoramic Segmentation
    :star:code
  • A Simple Latent Diffusion Approach for Panoptic Segmentation and Mask Inpainting
    :star:code
  • Point-supervised Panoptic Segmentation via Estimating Pseudo Labels from Learnable Distance
  • Strike a Balance in Continual Panoptic Segmentation
    :star:code
  • 3D Open-Vocabulary Panoptic Segmentation with 2D-3D Vision-Language Distillation
• 语义分割
  • Rethinking Data Augmentation for Robust LiDAR Semantic Segmentation in Adverse Weather
    :star:code
  • Exploring Reliable Matching with Phase Enhancement for Night-time Semantic Segmentation
  • MTA-CLIP: Language-Guided Semantic Segmentation with Mask-Text Alignment
  • Sparse Refinement for Efficient High-Resolution Semantic Segmentation
    :house:project
  • Embedding-Free Transformer with Inference Spatial Reduction for Efficient Semantic Segmentation
    :star:code
  • Weakly Supervised Co-training with Swapping Assignments for Semantic Segmentation
    :star:code
  • Context-Guided Spatial Feature Reconstruction for Efficient Semantic Segmentation
    :star:code
  • On-the-fly Category Discovery for LiDAR Semantic Segmentation
    :star:code
  • On the Viability of Monocular Depth Pre-training for Semantic Segmentation
  • Efficient Active Domain Adaptation for Semantic Segmentation by Selecting Information-rich Superpixels
    :star:code
  • Distributed Semantic Segmentation with Efficient Joint Source and Task Decoding
  • FREST: Feature RESToration for Semantic Segmentation under Multiple Adverse Conditions
  • Make a Strong Teacher with Label Assistance: A Novel Knowledge Distillation Approach for Semantic Segmentation
    :star:code
  • MeshSegmenter: Zero-Shot Mesh Segmentation via Texture Synthesis
    :star:code
  • Towards Reliable Evaluation and Fast Training of Robust Semantic Segmentation Models
    :star:code
  • ItTakesTwo: Leveraging Peer Representations for Semi-supervised LiDAR Semantic Segmentation
    :star:code
  • Evaluating the Adversarial Robustness of Semantic Segmentation: Trying Harder Pays Off
    :star:code
  • Open-Vocabulary RGB-Thermal Semantic Segmentation
    :star:code
  • Centering the Value of Every Modality: Towards Efficient and Resilient Modality-agnostic Semantic Segmentation
  • Learning Modality-agnostic Representation for Semantic Segmentation from Any Modalities
  • SFPNet: Sparse Focal Point Network for Semantic Segmentation on General LiDAR Point Clouds
    :house:project
    :star:code
  • Reliability in Semantic Segmentation: Can We Use Synthetic Data?
    :star:code
  • Cs2K: Class-specific and Class-shared Knowledge Guidance for Incremental Semantic Segmentation
  • MeshSegmenter: Zero-Shot Mesh Semantic Segmentation via Texture Synthesis
    :star:code
  • 3D语义分割
    • Open-Vocabulary 3D Semantic Segmentation with Text-to-Image Diffusion Models
    • LASS3D: Language-Assisted Semi-Supervised 3D Semantic Segmentation with Progressive Unreliable Data Exploitation
  • 跨域语义分割
    • Cross-Domain Semantic Segmentation on Inconsistent Taxonomy using VLMs
      :star:code
    • Attention Decomposition for Cross-Domain Semantic Segmentation
  • 无监督语义分割
    • Progressive Proxy Anchor Propagation for Unsupervised Semantic Segmentation
      :star:code
  • 半监督语义分割
    • Beyond Pixels: Semi-Supervised Semantic Segmentation with a Multi-scale Patch-based Multi-Label Classifier
      :star:code
    • Weighting Pseudo-Labels via High-Activation Feature Index Similarity and Object Detection for Semi-Supervised Segmentation
      :star:code
    • SemiVL: Semi-Supervised Semantic Segmentation with Vision-Language Guidance
      :star:code
  • 弱监督语义分割
    • Knowledge Transfer with Simulated Inter-Image Erasing for Weakly Supervised Semantic Segmentation
      :star:code
      通过模拟图像间擦除实现知识转移，弱监督语义分割再也不怕过扩展问题，助力精准目标定位！
    • Tendency-driven Mutual Exclusivity for Weakly Supervised Incremental Semantic Segmentation
    • DIAL: Dense Image-text ALignment for Weakly Supervised Semantic Segmentation
    • Finding Meaning in Points: Weakly Supervised Semantic Segmentation for Event Cameras
      :star:code
    • 3D weakly supervised semantic segmentation with 2D vision-language guidance
    • Learning from the Web: Language Drives Weakly-Supervised Incremental Learning for Semantic Segmentation
    • Phase Concentration and Shortcut Suppression for Weakly Supervised Semantic Segmentation
      :star:code
    • DHR: Dual Features-Driven Hierarchical Rebalancing in Inter- and Intra-Class Regions for Weakly-Supervised Semantic Segmentation
      :star:code
    • 3D Weakly Supervised Semantic Segmentation with 2D Vision-Language Guidance
      :star:code
    • Diffusion-Guided Weakly Supervised Semantic Segmentation
      :star:code
  • 域适应语义分割
    • MICDrop: Masking Image and Depth Features via Complementary Dropout for Domain-Adaptive Semantic Segmentation
  • 域泛化语义分割
    • DGInStyle: Domain-Generalizable Semantic Segmentation with Image Diffusion Models and Stylized Semantic Control
      :house:project
  • 类增量语义分割
    • Background Adaptation with Residual Modeling for Exemplar-Free Class-Incremental Semantic Segmentation
      :star:code
    • Mitigating Background Shift in Class-Incremental Semantic Segmentation
      :star:code
    • Early Preparation Pays Off: New Classifier Pre-tuning for Class Incremental Semantic Segmentation
      :star:code
  • 零样本语义分割
    • AlignZeg: Mitigating Objective Misalignment for Zero-shot Semantic Segmentation
    • OTSeg: Multi-prompt Sinkhorn Attention for Zero-Shot Semantic Segmentation
      :house:project
  • 开放词汇语义分割
    • CLIP-DINOiser: Teaching CLIP a few DINO tricks for open-vocabulary semantic segmentation
      :star:code
      :house:project
    • Explore the Potential of CLIP for Training-Free Open Vocabulary Semantic Segmentation
      :star:code
    • In Defense of Lazy Visual Grounding for Open-Vocabulary Semantic Segmentation
    • ProxyCLIP: Proxy Attention Improves CLIP for Open-Vocabulary Segmentation
      :star:code
• 部分分割
  • 3x2: 3D Object Part Segmentation by 2D Semantic Correspondences
    :star:code
  • WPS-SAM: Towards Weakly-Supervised Part Segmentation with Foundation Models
  • PartSTAD: 2D-to-3D Part Segmentation Task Adaptation
    :star:code
  • PARIS3D: Reasoning-based 3D Part Segmentation Using Large Multimodal Model
    :star:code
• 运动分割
  • Un-EVIMO: Unsupervised Event-based Independent Motion Segmentation
    :house:project
  • Appearance-based Refinement for Object-Centric Motion Segmentation
    :house:project
• 烟雾分割
  • DSA: Discriminative Scatter Analysis for Early Smoke Segmentation
• 线段分割
  • LineFit: A Geometric Approach for Fitting Line Segments in Images
• 场景解析
  • OLAF: A Plug-and-Play Framework for Enhanced Multi-object Multi-part Scene Parsing场景解析
• 交互式分割
  • Click-Gaussian: Interactive Segmentation to Any 3D Gaussians
    :star:code
  • Click Prompt Learning with Optimal Transport for Interactive Segmentation
    :house:project
• 小样本分割
  • Eliminating Feature Ambiguity for Few-Shot Segmentation
    :star:code
  • CAT-SAM: Conditional Tuning for Few-Shot Adaptation of Segment Anything Model
    :house:project
  • AlignDiff: Aligning Diffusion Models for General Few-Shot Segmentation
• 伪装目标分割
  • Open-Vocabulary Camouflaged Object Segmentation
    :star:code
• 参考图像分割
  • Pseudo-RIS: Distinctive Pseudo-supervision Generation for Referring Image Segmentation
• 指代图像分割
  • ReMamber: Referring Image Segmentation with Mamba Twister
    :star:code
  • GTMS: A Gradient-driven Tree-guided Mask-free Referring Image Segmentation Method
    :star:code
  • Finding NeMo: Negative-mined Mosaic Augmentation for Referring Image Segmentation
    :house:project
• 场景文本分割
  • EAFormer: Scene Text Segmentation with Edge-Aware Transformers
    :house:project
• 开放词汇分割
  • Diffusion Models for Open-Vocabulary Segmentation
  • Collaborative Vision-Text Representation Optimizing for Open-Vocabulary Segmentation
    :star:code
• 指代表达式分割
  • SafaRi:Adaptive Sequence Transformer for Weakly Supervised Referring Expression Segmentation
  • SAM4MLLM: Enhance Multi-Modal Large Language Model for Referring Expression Segmentation
  • SAFARI: Adaptive Sequence Transformer for Weakly Supervised Referring Expression Segmentation
    :house:project
• VIS
  • VISAGE: Video Instance Segmentation with Appearance-Guided Enhancement
    :star:code
  • Unified Embedding Alignment for Open-Vocabulary Video Instance Segmentation
    :star:code
• VOS
  • ActionVOS: Actions as Prompts for Video Object Segmentation
    :star:code
  • VISA: Reasoning Video Object Segmentation via Large Language Model
    :star:code
  • PanoVOS: Bridging Non-panoramic and Panoramic Views with Transformer for Video Segmentation
    :star:code
  • Exploring Pre-trained Text-to-Video Diffusion Models for Referring Video Object Segmentation
    :star:code
  • Betrayed by Attention: A Simple yet Effective Approach for Self-supervised Video Object Segmentation
    :star:code
  • OneVOS: Unifying Video Object Segmentation with All-in-One Transformer Framework
  • Spatial-Temporal Multi-level Association for Video Object Segmentation

10.Image Classification(图像分类)

• Labeled Data Selection for Category Discovery
• Active Generation for Image Classification
  :star:code
• Meta-Prompting for Automating Zero-shot Visual Recognition with LLMs
  :house:project
• Dyn-Adapter: Towards Disentangled Representation for Efficient Visual Recognition
• Wavelet Convolutions for Large Receptive Fields
  :star:code
• Momentum Auxiliary Network for Supervised Local Learning
  :star:code
• An accurate detection is not all you need to combat label noise in web-noisy datasets
  :star:code
• Dual-stage Hyperspectral Image Classification Model with Spectral Supertoken
  :star:code
• DEPICT: Diffusion-Enabled Permutation Importance for Image Classification Tasks
• NOVUM: Neural Object Volumes for Robust Object Classification
  :star:code
• EntAugment: Entropy-Driven Adaptive Data Augmentation Framework for Image Classification
  :star:code
• Distribution-Aware Robust Learning from Long-Tailed Data with Noisy Labels
  :star:code
• Discovering Unwritten Visual Classifiers with Large Language Models
• 广义类别发现
  • SelEx: Self-Expertise in Fine-Grained Generalized Category Discovery
    :star:code
  • Textual Knowledge Matters: Cross-Modality Co-Teaching for Generalized Visual Class Discovery
    :star:code广义类别发现(Generalized Category Discovery,GCD)
  • Learning to Distinguish Samples for Generalized Category Discovery
    :star:code
  • PromptCCD: Learning Gaussian Mixture Prompt Pool for Continual Category Discovery
    :star:code
  • Online Continuous Generalized Category Discovery
    :star:code广义类别发现
  • Category Adaptation Meets Projected Distillation in Generalized Continual Category Discovery
    :star:code
• 多标签图像分类
  • Distributionally Robust Loss for Long-Tailed Multi-Label Image Classification
    :star:code
• 小样本分类
  • Benchmarking Spurious Bias in Few-Shot Image Classifiers
    :star:code
  • Learning to Obstruct Few-Shot Image Classification over Restricted Classes
    :star:code
  • Semantic-guided Robustness Tuning for Few-Shot Transfer Across Extreme Domain Shift小样本分类
• 零样本分类
  • Online Zero-Shot Classification with CLIP
    :star:code
  • IG Captioner: Information Gain Captioners are Strong Zero-shot Classifiers
• 多标签识别
  • Modeling Label Correlations with Latent Context for Multi-Label Recognition
• 长尾识别
  • LTRL: Boosting Long-tail Recognition via Reflective Learning
  • Echoes of the Past: Boosting Long-tail Recognition via Reflective Learning
    :star:code
• 细粒度
  • On Learning Discriminative Features from Synthesized Data for Self-Supervised Fine-Grained Visual Recognition
  • A Rotation-invariant Texture ViT for Fine-Grained Recognition of Esophageal Cancer Endoscopic Ultrasound Images
    :star:code
  • Adapting Fine-Grained Cross-View Localization to Areas without Fine Ground Truth

9.Image Progress(图像/视频处理)

• ReNoise: Real Image Inversion Through Iterative Noising
  :star:code
  :house:project
• UniProcessor: A Text-induced Unified Low-level Image Processor
• 恢复
  • MoE-DiffIR: Task-customized Diffusion Priors for Universal Compressed Image Restoration
    :star:code
  • Panel-Specific Degradation Representation for Raw Under-Display Camera Image Restoration
    :star:code
  • Unsupervised Variational Translator for Bridging Image Restoration and High-Level Vision Tasks
  • A Comparative Study of Image Restoration Networks for General Backbone Network Design
  • GRIDS: Grouped Multiple-Degradation Restoration with Image Degradation Similarity
  • Restoring Images in Adverse Weather Conditions via Histogram Transformer
  • InstructIR: High-Quality Image Restoration Following Human Instructions
    :star:code
  • Diffusion Prior-Based Amortized Variational Inference for Noisy Inverse Problems
    :star:code
  • Towards Real-World Adverse Weather Image Restoration: Enhancing Clearness and Semantics with Vision-Language Models
  • Teaching Tailored to Talent: Adverse Weather Restoration via Prompt Pool and Depth-Anything Constraint
  • Seeing the Unseen: A Frequency Prompt Guided Transformer for Image Restoration
    :star:code
  • MambaIR: A Simple Baseline for Image Restoration with State-Space Model
    :star:code
    :thumbsup:MambaIR: 基于Mamba的图像复原基准模型
  • AutoDIR: Automatic All-in-One Image Restoration with Latent Diffusion
    :star:code
  • SPIRE: Semantic Prompt-Driven Image Restoration
    :house:project
  • Efficient Cascaded Multiscale Adaptive Network for Image Restoration
  • Restore Anything with Masks: Leveraging Mask Image Modeling for Blind All-in-One Image Restoration
    :star:code
  • When Fast Fourier Transform Meets Transformer for Image Restoration
    :star:code
  • Osmosis: RGBD Diffusion Prior for Underwater Image Restoration
    :house:project
  • Contribution-based Low-Rank Adaptation with Pre-training Model for Real Image Restoration
    :house:project
  • DiffBIR: Toward Blind Image Restoration with Generative Diffusion Prior
    :star:code
  • MetaWeather: Few-Shot Weather-Degraded Image Restoration
    :star:code
  • Depth-Aware Blind Image Decomposition for Real-World Adverse Weather Recovery
• 修补
  • A Task is Worth One Word: Learning with Task Prompts for High-Quality Versatile Image Inpainting
    :star:code
    :house:project
  • Improving Text-guided Object Inpainting with Semantic Pre-inpainting
    :star:code
  • BrushNet: A Plug-and-Play Image Inpainting Model with Decomposed Dual-Branch Diffusion
    :house:project
• 去雨
  • Efficient Frequency-Domain Image Deraining with Contrastive Regularization
    :star:code
• 去噪
  • TTT-MIM: Test-Time Training with Masked Image Modeling for Denoising Distribution Shifts
    :star:code
  • Region-Aware Sequence-to-Sequence Learning for Hyperspectral Denoising
    :star:code
  • DualDn: Dual-domain Denoising via Differentiable ISP
    :star:code
  • Exploiting Dual-Correlation for Multi-frame Time-of-Flight Denoising
    :star:code
  • EDformer: Transformer-Based Event Denoising Across Varied Noise Levels
  • denoiSplit: a method for joint microscopy image splitting and unsupervised denoising去噪
  • Asymmetric Mask Scheme for Self-Supervised Real Image Denoising
    :star:code
  • Switch Diffusion Transformer: Synergizing Denoising Tasks with Sparse Mixture-of-Experts
    :house:project
  • Enhancing Plausibility Evaluation for Generated Designs with Denoising Autoencoder
    :star:code
• 去雾
  • Unleashing the Potential of the Semantic Latent Space in Diffusion Models for Image Dehazing
  • Prompt-Based Test-Time Real Image Dehazing: A Novel Pipeline
    :star:code
• 去模糊
  • Deblur e-NeRF: NeRF from Motion-Blurred Events under High-speed or Low-light Conditions
    :star:code
  • UniINR: Event-guided Unified Rolling Shutter Correction, Deblurring, and Interpolation
    :star:code
  • Blind image deblurring with noise-robust kernel estimation
    :star:code
  • Motion Aware Event Representation-driven Image Deblurring(https://github.com/ZhijingS/DA_event_deblur)
  • BAD-Gaussians: Bundle Adjusted Deblur Gaussian Splatting
    :star:code
• 去卷积
  • Blind Image Deconvolution by Generative-based Kernel Prior and Initializer via Latent Encoding
    :star:code
  • Kernel Diffusion: An Alternate Approach to Blind Deconvolution
• 去反射
  • L-DiffER: Single Image Reflection Removal with Language-based Diffusion Model消除单图像反射
  • Flash-Splat: 3D Reflection Removal with Flash Cues and Gaussian Splats
    :house:project消除 3D 反射
• 去伪影
  • OAPT: Offset-Aware Partition Transformer for Double JPEG Artifacts Removal
    :star:code
• 去摩尔纹
  • Image Demoireing in RAW and sRGB Domains
    :star:code
• 去马赛克
  • Wavelength-Embedding-guided Filter-Array Transformer for Spectral Demosaicing去马赛克
• 目标移除
  • GScream: Learning 3D Geometry and Feature Consistent Gaussian Splatting for Object Removal
    :house:project
  • ObjectDrop: Bootstrapping Counterfactuals for Photorealistic Object Removal and Insertion
    🤗huggingface
• 扩图
  • Be-Your-Outpainter: Mastering Video Outpainting through Input-Specific Adaptation
    :star:code
• 图像修饰
  • Taming Lookup Tables for Efficient Image Retouching
    :star:code
• 图像增强
  • LightenDiffusion: Unsupervised Low-Light Image Enhancement with Latent-Retinex Diffusion Models
    :star:code
  • LMT-GP: Combined Latent Mean-Teacher and Gaussian Process for Semi-supervised Low-light Image Enhancement
    :star:code
  • RAVE: Residual Vector Embedding for CLIP-Guided Backlit Image Enhancement
    :star:code
  • Image-adaptive 3D Lookup Tables for Real-time Image Enhancement with Bilateral Grids
    :star:code
  • NamedCurves: Learned Image Enhancement via Color Naming
  • Joint RGB-Spectral Decomposition Model Guided Image Enhancement in Mobile Photography
    :star:code
  • GLARE: Low Light Image Enhancement via Generative Latent Feature based Codebook Retrieval
    :star:code
    :thumbsup:GLARE 利用外部正常光照先验，实现逼真的低光照增强效果！
  • Fast Context-Based Low-Light Image Enhancement via Neural Implicit Representations
    :star:code
  • Unveiling Advanced Frequency Disentanglement Paradigm for Low-Light Image Enhancement
    :star:code
• 图像质量评估
  • DSMix: Distortion-Induced Sensitivity Map Based Pre-training for No-Reference Image Quality Assessment
    :star:code
  • A Comprehensive Study of Multimodal Large Language Models for Image Quality Assessment
  • Towards Open-ended Visual Quality Comparison
    :star:code
    :Thumbsup:Co-Instruct: 让通用多模态大模型学会比较视觉质量
  • PromptIQA: Boosting the Performance and Generalization for No-Reference Image Quality Assessment via Prompts无参考图像质量评估
  • Assessing UHD Image Quality from Aesthetics, Distortions, and Saliency
    :star:code
  • Depicting Beyond Scores: Advancing Image Quality Assessment through Multi-modal Language Models
    :house:project
  • DSMix: Distortion-Induced Saliency Map Based Pre-training for No-Reference Image Quality Assessment
    :star:code
• 图像美学质量评价
  • Scaling Up Personalized Image Aesthetic Assessment via Task Vector Customization
    :house:project图像美学评估
• 视频恢复
  • Quanta Video Restoration
    :star:code
• 视频着色
  • ColorMNet: A Memory-based Deep Spatial-Temporal Feature Propagation Network for Video Colorization
    :star:code
• 视频增强
  • Noise Calibration: Plug-and-play Content-Preserving Video Enhancement using Pre-trained Video Diffusion Models
    :star:code
    :star:code
  • Unrolled Decomposed Unpaired Learning for Controllable Low-Light Video Enhancement
    :star:code
• 视频去雨
  • Dual-Rain: Video Rain Removal using Assertive and Gentle Teachers
• 视频去噪
  • Temporal As a Plugin: Unsupervised Video Denoising with Pre-Trained Image Denoisers
    :star:code
• 视频去雪
  • Semi-Supervised Video Desnowing Network via Temporal Decoupling Experts and Distribution-Driven Contrastive Regularization
    :star:code去雪
• 视频去模糊
  • Domain-adaptive Video Deblurring via Test-time Blurring
    :star:code
  • CMTA: Cross-Modal Temporal Alignment for Event-guided Video Deblurring
    :star:code
  • Cross-Modal Temporal Alignment for Event-guided Video Deblurring
    :star:code
  • Towards Real-world Event-guided Low-light Video Enhancement and Deblurring
    :star:code
  • Rethinking Video Deblurring with Wavelet-Aware Dynamic Transformer and Diffusion Model
    :star:code
• 视频去闪烁
  • BlazeBVD: Make Scale-Time Equalization Great Again for Blind Video Deflickering
    :thumbsup:盲视频去闪烁通用方法BlazeBVD来了，美图&国科大联合提出
• 视频去马赛克
  • Self-Supervised Video Desmoking for Laparoscopic Surgery
    :star:code
• 视频质量增强
  • Online Video Quality Enhancement with Spatial-Temporal Look-up Tables
• 重照明
  • Personalized Video Relighting With an At-Home Light Stage
    :star:code
  • The Sky's the Limit: Relightable Outdoor Scenes via a Sky-pixel Constrained Illumination Prior and Outside-In Visibility
    :star:code重照明

8.Super-Resolution(超分辨率)

• Data Overfitting for On-Device Super-Resolution with Dynamic Algorithm and Compiler Co-Design
  :star:code
• SMFANet: A Lightweight Self-Modulation Feature Aggregation Network for Efficient Image Super-Resolution
  :star:code
• Towards Robust Full Low-bit Quantization of Super Resolution Networks
• BurstM: Deep Burst Multi-scale SR using Fourier Space with Optical Flow
  :star:code
• HiT-SR: Hierarchical Transformer for Efficient Image Super-Resolution
• Pairwise Distance Distillation for Unsupervised Real-World Image Super-Resolution
  :star:code
• UCIP: A Universal Framework for Compressed Image Super-Resolution using Dynamic Prompt
  :star:code
• Accelerating Image Super-Resolution Networks with Pixel-Level Classification
  :star:code
• Rethinking Image Super-Resolution from Training Data Perspectives
• Spatially-Variant Degradation Model for Dataset-free Super-resolution
  :star:code
• Learning Exhaustive Correlation for Spectral Super-Resolution: Where Spatial-Spectral Attention Meets Linear Dependence
• Contourlet Residual for Prompt Learning Enhanced Infrared Image Super-Resolution
  :star:code
• Confidence-Based Iterative Generation for Real-World Image Super-Resolution
  :star:code
• Adaptive Multi-modal Fusion of Spatially Variant Kernel Refinement with Diffusion Model for Blind Image Super-Resolution
• Pixel-Aware Stable Diffusion for Realistic Image Super-Resolution and Personalized Stylization
  :star:code
• XPSR: Cross-modal Priors for Diffusion-based Image Super-Resolution
  :star:code
• AdaDiffSR: Adaptive Region-aware Dynamic acceleration Diffusion Model for Real-World Image Super-Resolution
• Overcoming Distribution Mismatch in Quantizing Image Super-Resolution Networks
  :star:code
• Rethinking Image Super Resolution from Training Data Perspectives
  :star:code
• MTKD: Multi-Teacher Knowledge Distillation for Image Super-Resolution
  :star:code
• OmniSSR: Zero-shot Omnidirectional Image Super-Resolution using Stable Diffusion Model
• Learning Dual-Level Deformable Implicit Representation for Real-World Scale Arbitrary Super-Resolution
  :star:code
• You Only Need One Step: Fast Super-Resolution with Stable Diffusion via Scale Distillation
  :star:code
• A New Dataset and Framework for Real-World Blurred Images Super-Resolution
  :star:code
• 场景文本图像超分辨率
  • DCDM: Diffusion-Conditioned-Diffusion Model for Scene Text Image Super-Resolution
    :star:code
• VSR
  • Arbitrary-Scale Video Super-Resolution with Structural and Textural Priors
    :star:code
  • RealViformer: Investigating Attention for Real-World Video Super-Resolution
    :star:code
  • Enhancing Perceptual Quality in Video Super-Resolution through Temporally-Consistent Detail Synthesis using Diffusion Models
    :star:code
  • SuperGaussian: Repurposing Video Models for 3D Super Resolution
    :house:project
  • Event-Adapted Video Super-Resolution
  • Motion-Guided Latent Diffusion for Temporally Consistent Real-world Video Super-resolution
    :star:code

7.Object Detection(目标检测)

• Can OOD Object Detectors Learn from Foundation Models?
  :star:code
• Crowd-SAM:SAM as a smart annotator for object detection in crowded scenes
  :star:code
• Distilling Knowledge from Large-Scale Image Models for Object Detection
• DeTra: A Unified Model for Object Detection and Trajectory Forecasting
• Modality Translation for Object Detection Adaptation without forgetting prior knowledge
• OpenSight: A Simple Open-Vocabulary Framework for LiDAR-Based Object Detection
• LEROjD: Lidar Extended Radar-Only Object Detection
  :star:code
• Bucketed Ranking-based Losses for Efficient Training of Object Detectors
  :star:code
• Plain-Det: A Plain Multi-Dataset Object Detector
  :star:code
• On Calibration of Object Detectors: Pitfalls, Evaluation and Baselines
  :star:code
• Towards Open-World Object-based Anomaly Detection via Self-Supervised Outlier Synthesis
  :star:code
• Weak-to-Strong Compositional Learning from Generative Models for Language-based Object Detection
• PartGLEE: A Foundation Model for Recognizing and Parsing Any Objects
  :star:code
• Crowd-SAM: SAM as a Smart Annotator for Object Detection in Crowded Scenes
  :star:code
• Relation DETR: Exploring Explicit Position Relation Prior for Object Detection
  :star:code
• Bridge Past and Future: Overcoming Information Asymmetry in Incremental Object Detection
  :star:code
• Modality Translation for Object Detection Adaptation Without Forgetting Prior Knowledge
  :star:code
• T-Rex2: Towards Generic Object Detection via Text-Visual Prompt Synergy
  :star:code
• Fine-grained Dynamic Network for Generic Event Boundary Detection
  :star:code
• CSOT: Cross-Scan Object Transfer for Semi-Supervised LiDAR Object Detection
• Bayesian Detector Combination for Object Detection with Crowdsourced Annotations
  :star:code
• Simplifying Source-Free Domain Adaptation for Object Detection: Effective Self-Training Strategies and Performance Insights
  :star:code
• Out-of-Bounding-Box Triggers: A Stealthy Approach to Cheat Object Detectors
  :star:code
• A Simple Background Augmentation Method for Object Detection with Diffusion Model
• Look Around and Learn: Self-Training Object Detection by Exploration
  :star:code
• Co-Student: Collaborating Strong and Weak Students for Sparsely Annotated Object Detection
  :star:code
• Benchmarking Object Detectors with COCO: A New Path Forward
  :sunflower:dataset
• DAMSDet: Dynamic Adaptive Multispectral Detection Transformer with Competitive Query Selection and Adaptive Feature Fusion
  :star:code目标检测
• Integer-Valued Training and Spike-driven Inference Spiking Neural Network for High-performance and Energy-efficient Object Detection
  :star:code
• YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information
  :star:code
• Equivariant Spatio-Temporal Self-Supervision for LiDAR Object Detection
• Projecting Points to Axes: Oriented Object Detection via Point-Axis Representation
• GRA: Detecting Oriented Objects through Group-wise Rotating and Attention
• Embracing Events and Frames with Hierarchical Feature Refinement Network for Object Detection
  :star:code
• Dynamic Retraining-Updating Mean Teacher for Source-Free Object Detection
  :star:code
• Zero-Shot Detection of AI-Generated Images
  :star:code
  :house:project
• MOD-UV: Learning Mobile Object Detectors from Unlabeled Videos
  :star:code
• Category-level Object Detection, Pose Estimation and Reconstruction from Stereo Images
  :house:project
• Rethinking Features-Fused-Pyramid-Neck for Object Detection
  :star:code
• 3D目标检测
  • Make Your ViT-based Multi-view 3D Detectors Faster via Token Compression
    :star:code
  • Approaching Outside: Scaling Unsupervised 3D Object Detection from 2D Scene
  • SparseLIF: High-Performance Sparse LiDAR-Camera Fusion for 3D Object Detection
  • MonoWAD: Weather-Adaptive Diffusion Model for Robust Monocular 3D Object Detection
    :star:code
  • Transfer Learning from Simulated to Real Scenes for Monocular 3D Object Detection
    :star:code
  • Domain Generalization of 3D Object Detection by Density-Resampling
    :star:code
  • Learning High-resolution Vector Representation from Multi-Camera Images for 3D Object Detection
    :star:code
  • LiDAR-based All-weather 3D Object Detection via Prompting and Distilling 4D Radar
  • Towards Stable 3D Object Detection
  • RecurrentBEV: A Long-term Temporal Fusion Framework for Multi-view 3D Detection
    :star:code
  • LISO: Lidar-only Self-Supervised 3D Object Detection
    :star:code
  • Diffusion Model for Robust Multi-Sensor Fusion in 3D Object Detection and BEV Segmentation
    :star:code
  • SAMFusion: Sensor-Adaptive Multimodal Fusion for 3D Object Detection in Adverse Weather
  • Unlocking Textual and Visual Wisdom: Open-Vocabulary 3D Object Detection Enhanced by Comprehensive Guidance from Text and Image
  • OV-Uni3DETR: Towards Unified Open-Vocabulary 3D Object Detection via Cycle-Modality Propagation
    :star:code
  • Better Regression Makes Better Test-time Adaptive 3D Object Detection
  • Find n' Propagate: Open-Vocabulary 3D Object Detection in Urban Environments
    :star:code
  • Diff3DETR: Agent-based Diffusion Model for Semi-supervised 3D Object Detection
  • Weakly Supervised 3D Object Detection via Multi-Level Visual Guidance
    :star:code
  • SimPB: A Single Model for 2D and 3D Object Detection from Multiple Cameras
    :star:code
  • CMD: A Cross Mechanism Domain Adaptation Dataset for 3D Object Detection
    :star:code
    :thumbsup:DIG从密度、强度和几何三方面缓和传感器体制带来的点云数据差异，显著提升了域自适应算法的性能。
  • Ray Denoising: Depth-aware Hard Negative Sampling for Multi-view 3D Object Detection
    :star:code
  • OPEN: Object-wise Position Embedding for Multi-view 3D Object Detection
    :star:code
  • LabelDistill: Label-guided Cross-modal Knowledge Distillation for Camera-based 3D Object Detection
    :star:code
  • MonoTTA: Fully Test-Time Adaptation for Monocular 3D Object Detection
    :star:code
  • FSD-BEV: Foreground Self-Distillation for Multi-view 3D Object Detection
    :star:code
  • General Geometry-aware Weakly Supervised 3D Object Detection
    :star:code
  • Interactive 3D Object Detection with Prompts
  • Beyond Viewpoint: Robust 3D Object Recognition under Arbitrary Views through Joint Multi-Part Representation
  • Detecting As Labeling: Rethinking LiDAR-camera Fusion in 3D Object Detection
    :star:code
  • TCC-Det: Temporarily consistent cues for weakly-supervised 3D detection
    :star:code
  • GraphBEV: Towards Robust BEV Feature Alignment for Multi-Modal 3D Object Detection
    :star:code
• 小目标检测
  • IRSAM: Advancing Segment Anything Model for Infrared Small Target Detection
  • Visible and Clear: Finding Tiny Objects in Difference Map
  • DQ-DETR: DETR with Dynamic Query for Tiny Object Detection
    :star:code
  • 3D Small Object Detection with Dynamic Spatial Pruning
    :star:code
    :thumbsup:DSPDet3D：基于动态空间剪枝的高效率3D小目标检测
• 伪装目标检测
  • CamoTeacher: Dual-Rotation Consistency Learning for Semi-Supervised Camouflaged Object Detection
    :thumbsup:有效减少了像素级、实例级噪声问题
  • Learning Camouflaged Object Detection from Noisy Pseudo Label
    :star:code
  • Just a Hint: Point-Supervised Camouflaged Object Detection
  • SAM-COD: SAM-guided Unified Framework for Weakly-Supervised Camouflaged Object Detection
  • Frequency-Spatial Entanglement Learning for Camouflaged Object Detection
    :star:code
  • FocusDiffuser: Perceiving Local Disparities for Camouflaged Object Detection
    :star:code
• 长尾目标检测
  • Rectify the Regression Bias in Long-Tailed Object Detection
• 显著目标检测
  • CoLA: Conditional Dropout and Language-driven Robust Dual-modal Salient Object Detection
    :star:code
• 域适应目标检测
  • Enhancing Source-Free Domain Adaptive Object Detection with Low-confidence Pseudo Label Distillation
    :star:code
  • Revisiting Domain-Adaptive Object Detection in Adverse Weather by the Generation and Composition of High-Quality Pseudo-Labels
    :star:code
• 小样本目标检测
  • SMILe: Leveraging Submodular Mutual Information For Robust Few-Shot Object Detection
    :house:project
  • Adaptive Multi-task Learning for Few-shot Object Detection
    :star:code
  • Cross-Domain Few-Shot Object Detection via Enhanced Open-Set Object Detector
    :house:project
    :thumbsup:跨域小样本物体检测CD-FSOD新数据集、CD-ViTO新方法（数据代码均已开源）
• 共同显著目标检测
  • CONDA: Condensed Deep Association Learning for Co-Salient Object Detection
    :star:code
  • Self-supervised co-salient object detection via feature correspondences at multiple scales
    :star:code
• 开放词汇目标检测
  • Global-Local Collaborative Inference with LLM for Lidar-Based Open-Vocabulary Detection
    :star:code
  • LaMI-DETR: Open-Vocabulary Detection with Language Model Instruction
  • MarvelOVD: Marrying Object Recognition and Vision-Language Models for Robust Open-Vocabulary Object Detection
    :star:code
  • Region-centric Image-Language Pretraining for Open-Vocabulary Detection
    :star:code
  • CLIFF: Continual Latent Diffusion for Open-Vocabulary Object Detection
    :star:code
• 水印检测
  • Finding a needle in a haystack: A Black-Box Approach to Invisible Watermark Detection
• 阴影检测
  • RenDetNet: Weakly-supervised Shadow Detection with Shadow Caster Verification
    :star:code
  • Two-Stage Video Shadow Detection via Temporal-Spatial Adaption
    :house:project
• 开集识别
  • Operational Open-Set Recognition and PostMax Refinement
  • Open-Set Recognition in the Age of Vision-Language Models
  • Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection
    :star:code
• 目标定位
  • WildRefer: 3D Object Localization in Large-scale Dynamic Scenes with Multi-modal Visual Data and Natural Language
    :star:code
  • Pro2SAM: Mask Prompt to SAM with Grid Points for Weakly Supervised Object Localization

6.Object Tracking(目标跟踪)

• Local All-Pair Correspondence for Point Tracking
  :star:code
  :star:code
• Track Everything Everywhere Fast and Robustly
  :star:code
  :house:project
• CoTracker: It is Better to Track Together
  :house:project
• DINO-Tracker: Taming DINO for Self-Supervised Point Tracking in a Single Video
  :star:code
  :house:project
• Decomposition Betters Tracking Everything Everywhere
  :star:code
• Self-Supervised Any-Point Tracking by Contrastive Random Walks
  :house:project
  :star:code
• TAPTR: Tracking Any Point with Transformers as Detection
  :star:code
• MapTracker: Tracking with Strided Memory Fusion for Consistent Vector HD Mapping
  :house:project
• SPAMming Labels: Efficient Annotations for the Trackers of Tomorrow
• SLAck: Semantic, Location, and Appearance Aware Open-Vocabulary Tracking
  :star:code
• OneTrack: Demystifying the Conflict Between Detection and Tracking in End-to-End 3D Trackers
• Exploring the Feature Extraction and Relation Modeling For Light-Weight Transformer Tracking
  :star:code
• Empowering Embodied Visual Tracking with Visual Foundation Models and Offline RL
• 3D目标跟踪
  • 3D Single-object Tracking in Point Clouds with High Temporal Variation
  • Boosting 3D Single Object Tracking with 2D Matching Distillation and 3D Pre-training
• 多目标跟踪
  • Lost and Found: Overcoming Detector Failures in Online Multi-Object Tracking
    :star:code
  • Walker: Self-supervised Multiple Object Tracking by Walking on Temporal Object Appearance Graphs
  • Beyond MOT: Semantic Multi-Object Tracking
    :star:code
  • PapMOT: Exploring Adversarial Patch Attack against Multiple Object Tracking
  • VETRA: A Dataset for Vehicle Tracking in Aerial Imagery - New Challenges for Multi-Object Tracking
    :house:project
• 细胞跟踪
  • Trackastra: Transformer-based cell tracking for live-cell microscopy
    :star:code
  • Large-Scale Multi-Hypotheses Cell Tracking Using Ultrametric Contours Maps
    :star:code

5.OCR

• Parrot Captions Teach CLIP to Spot Text
  :star:code
• WeCromCL: Weakly Supervised Cross-Modality Contrastive Learning for Transcription-only Supervised Text Spotting
• FineMatch: Aspect-based Fine-grained Image and Text Mismatch Detection and Correction
  :house:project
• Bridging Synthetic and Real Worlds for Pre-training Scene Text Detectors
  :star:code
• 手写文本检测
  • Align, Minimize and Diversify: A Source-Free Unsupervised Domain Adaptation Method for Handwritten Text Recognition
  • PosFormer: Recognizing Complex Handwritten Mathematical Expression with Position Forest Transformer
    :star:code
    :Thumbsup:上交推出 PosFormer！优化位置识别任务来辅助表达式识别，复杂公式识别能力再创新SOTA！
  • Elegantly Written: Disentangling Writer and Character Styles for Enhancing Online Chinese Handwriting
  • NAMER: Non-Autoregressive Modeling for Handwritten Mathematical Expression Recognition
• 手写文本合成
  • One-Shot Diffusion Mimicker for Handwritten Text Generation
    :star:code
  • DiffusionPen: Towards Controlling the Style of Handwritten Text Generation
• 场景文本删除
  • Leveraging Text Localization for Scene Text Removal via Text-aware Masked Image Modeling
    :star:code
• 文档理解
  • Textual Grounding for Open-vocabulary Visual Information Extraction in Layout-diversified Documents
    :thumbsup:结合版式感知上下文学习和适用于文档的两阶段预训练，显著提高了模型对文档的理解能力
  • VisFocus: Prompt-Guided Vision Encoders for OCR-Free Dense Document Understanding
    🤗huggingface密集文档理解
• 文本分割
  • WAS: Dataset and Methods for Artistic Text Segmentation
    :star:code
• 文本合成
  • Visual Text Generation in the Wild
    :star:code
• 文本修复
  • LLMCO4MR: LLMs-aided Neural Combinatorial Optimization for Ancient Manuscript Restoration from Fragments with Case Studies on Dunhuang从碎片修复古代手稿以敦煌为例

4.Pose(姿态估计)

• X-Pose: Detecting Any Keypoints
  :star:code
• VQ-HPS: Human Pose and Shape Estimation in a Vector-Quantized Latent Space
  :house:project
• Expressive Whole-Body 3D Gaussian Avatar
  :star:code
• GTPT: Group-based Token Pruning Transformer for Efficient Human Pose Estimation
• PoseSOR: Human Pose Can Guide Our Attention
  :star:code
• COSMU: Complete 3D human shape from monocular unconstrained images
• Modeling and Driving Human Body Soundfields through Acoustic Primitives
  :house:project
• Domain-Adaptive 2D Human Pose Estimation via Dual Teachers in Extremely Low-Light Conditions
  :star:code
• SpaRP: Fast 3D Object Reconstruction and Pose Estimation from Sparse Views
• PoseEmbroider: Towards a 3D, Visual, Semantic-aware Human Pose Representation
• PoseAugment: Generative Human Pose Data Augmentation with Physical Plausibility for IMU-based Motion Capture
  :star:code
• HPE-Li: WiFi-enabled Lightweight Dual Selective Kernel Convolution for Human Pose Estimation人体姿势估计
• EgoPoser: Robust Real-Time Egocentric Pose Estimation from Sparse and Intermittent Observations Everywhere
  :house:project
• You Only Learn One Query: Learning Unified Human Query for Single-Stage Multi-Person Multi-Task Human-Centric Perception
  :star:code
• Within the Dynamic Context: Inertia-aware 3D Human Modeling with Pose Sequence
• Benchmarks and Challenges in Pose Estimation for Egocentric Hand Interactions with Objects
• Human Pose Recognition via Occlusion-Preserving Abstract Images
• 文本驱动的人体生成
  • InterFusion: Text-Driven Generation of 3D Human-Object Interaction
    :house:project
  • Towards Open Domain Text-Driven Synthesis of Multi-Person Motions
    :house:project
• 多人姿势预测
  • Multi-Person Pose Forecasting with Individual Interaction Perceptron and Prior Learning
    :star:code
• 3D人体姿态估计
  • MPL: Lifting 3D Human Pose from Multi-view 2D Poses
    :star:code
  • RT-Pose: A 4D Radar Tensor-based 3D Human Pose Estimation and Localization Benchmark
    :house:project
  • AvatarPose: Avatar-guided 3D Pose Estimation of Close Human Interaction from Sparse Multi-view Videos
    :star:code
    :house:project
  • Mask as Supervision: Leveraging Unified Mask Information for Unsupervised 3D Pose Estimation
    :star:code
  • 3D Human Pose Estimation via Non-Causal Retentive Networks
    :star:code
  • UPose3D: Uncertainty-Aware 3D Human Pose Estimation with Cross-View and Temporal Cues
  • Occlusion Handling in 3D Human Pose Estimation with Perturbed Positional Encoding
  • RePOSE: 3D Human Pose Estimation via Spatio-Temporal Depth Relational Consistency
    :star:code
  • RT-Pose: A 4D Radar-Tensor based 3D Human Pose Estimation and Localization Benchmark
    🤗huggingface
  • EgoPoseFormer: A Simple Baseline for Stereo Egocentric 3D Human Pose Estimation
    :star:code
  • 3DSA:Multi-View 3D Human Pose Estimation With 3D Space Attention Mechanisms
  • WorldPose: A World Cup Dataset for Global 3D Human Pose Estimation
  • Rotated Orthographic Projection for Self-Supervised 3D Human Pose Estimation
  • NICP: Neural ICP for 3D Human Registration at Scale
    :house:project
• 人体网格恢复
  • Multi-HMR: Multi-Person Whole-Body Human Mesh Recovery in a Single Shot
    :star:code
  • Divide and Fuse: Body Part Mesh Recovery from Partially Visible Human Images
  • Global-to-Pixel Regression for Human Mesh Recovery
  • WindPoly: Polygonal Mesh Reconstruction via Winding Numbers
    :house:project
  • Multi-RoI Human Mesh Recovery with Camera Consistency and Contrastive Losses
    :star:code
• 3D人体纹理生成
  • TexDreamer: Towards Zero-Shot High-Fidelity 3D Human Texture Generation
    :house:project
• 3D人体生成
  • StructLDM: Structured Latent Diffusion for 3D Human Generation
    :star:code
    :house:project
    :thumbsup:南洋理工三维数字人生成新范式：结构扩散模型
  • Text Motion Translator: A Bi-Directional Model for Enhanced 3D Human Motion Generation from Open-Vocabulary Descriptions
  • Text to Layer-wise 3D Clothed Human Generation
    :house:project
  • SemanticHuman-HD: High Resolution Semantic disentangled 3D Human Generation
    :house:project3D 人类生成
• 人体重建
  • ReLoo: Reconstructing Humans Dressed in Loose Garments from Monocular Video in the Wild
    :house:project
  • Learned Neural Physics Simulation for Articulated 3D Human Pose Reconstruction
• 动作捕捉
  • LiveHPS++: Robust and Coherent Motion Capture in Dynamic Free Environment
    :house:project
• 手语识别
  • EvSign: Sign Language Recognition and Translation with Streaming Events
    :star:code
  • Pose-Guided Fine-Grained Sign Language Video Generation
  • Visual Alignment Pre-training for Sign Language Translation
  • A Simple Baseline for Spoken Language to Sign Language Translation with 3D Avatars
    :star:code
• 手部网格
  • HandDGP: Camera-Space Hand Mesh Prediction with Differentiable Global Positioning
• 3D手部序列恢复
  • 3D Hand Sequence Recovery from Real Blurry Images and Event Stream
• 3D手部重建
  • Weakly-Supervised 3D Hand Reconstruction with Knowledge Prior and Uncertainty Guidance
  • AttentionHand: Text-driven Controllable Hand Image Generation for 3D Hand Reconstruction in the Wild
  • MLPHand: Real Time Multi-View 3D Hand Reconstruction via MLP Modeling
    :star:code
• 手部姿态估计
  • HandDAGT: A Denoising Adaptive Graph Transformer for 3D Hand Pose Estimation
    :star:code
  • 3D Hand Pose Estimation in Everyday Egocentric Images
• 手部运动预测
  • Prompting Future Driven Diffusion Model for Hand Motion Prediction
• 头部姿态估计
  • Event-based Head Pose Estimation: Benchmark and Method
    :star:code
• 手持物体重建
  • D-SCo: Dual-Stream Conditional Diffusion for Monocular Hand-Held Object Reconstruction手持物体重建
• 头部姿态估计
  • 6DoF Head Pose Estimation through Explicit Bidirectional Interaction with Face Geometry
    :star:code
• 4D 头部捕获
  • Topo4D: Topology-Preserving Gaussian Splatting for High-Fidelity 4D Head Capture
    :star:code
    :house:project4D 头部捕获
• 动作捕捉
  • MetaCap: Meta-learning Priors from Multi-View Imagery for Sparse-view Human Performance Capture and Rendering
    :star:code
• 手语视频生成
  • Pose Guided Fine-Grained Sign Language Video Generation
  • Co-speech Gesture Video Generation with 3D Human Meshes

3.Face(人脸)

• Task-adaptive Q-Face
• Faceptor: A Generalist Model for Face Perception
  :star:code
• A Light Stage on Every Desk
  :house:project
• Face Adapter for Pre-Trained Diffusion Models with Fine-Grained ID and Attribute Control
  :star:code
  :house:project
• ReSyncer: Rewiring Style-based Generator for Unified Audio-Visually Synced Facial Performer
  :star:code
• Facial Affective Behavior Analysis with Instruction Tuning
  :star:code
  :house:project
• Arc2Face: A Foundation Model for ID-Consistent Human Faces
  :star:code
  :house:project
• GAMMA-FACE: GAussian Mixture Models Amend Diffusion Models for Bias Mitigation in Face Images
• GRAPE: Generalizable and Robust Multi-view Facial Capture
• High-Quality Mesh Blendshape Generation from Face Videos via Neural Inverse Rendering
  :star:code
• 人脸交换
  • SelfSwapper: Self-Supervised Face Swapping via Shape Agnostic Masked AutoEncoder
    🤗huggingface
• 人脸模糊
  • Forbes: Face Obfuscation Rendering via Backpropagation Refinement Scheme
    :star:code
• 人脸识别
  • Towards Certifiably Robust Face Recognition
  • AdaDistill: Adaptive Knowledge Distillation for Deep Face Recognition
    :star:code
  • ARoFace: Alignment Robustness to Improve Low-Quality Face Recognition
    :star:code
  • Personalized Privacy Protection Mask Against Unauthorized Facial Recognition
  • MST-KD: Multiple Specialized Teachers Knowledge Distillation for Fair Face Recognition
  • dversariaLeak: External Information Leakage Attack Using Adversarial Samples on Face Recognition Systems
• 人脸聚类
  • VideoClusterNet: Self-Supervised and Adaptive Face Clustering for Videos人脸聚类
• 人脸重建
  • Face Reconstruction Transfer Attack as Out-of-Distribution Generalization
    :star:code
• 人脸表情
  • Norface: Improving Facial Expression Analysis by Identity Normalization
    :star:code
  • Generalizable Facial Expression Recognition
    :star:code
  • How Video Meetings Change Your Expression
    :house:project人脸
  • AnimateMe: 4D Facial Expressions via Diffusion Models
• 人脸编辑
  • GroupDiff: Diffusion-based Group Portrait Editing
    :star:code
  • Real-time 3D-aware Portrait Editing from a Single Image
    :star:code肖像编辑
  • Efficient 3D-Aware Facial Image Editing via Attribute-Specific Prompt Learning
    :star:code
• 人脸动画
  • KMTalk: Speech-Driven 3D Facial Animation with Key Motion Embedding
    :star:code
  • Fast Registration of Photorealistic Avatars for VR Facial Animation
    :house:project
  • UniTalker: Scaling up Audio-Driven 3D Facial Animation through A Unified Model
    🤗huggingface
• 说话头合成
  • ScanTalk: 3D Talking Heads from Unregistered Scans
    :star:code
  • EDTalk: Efficient Disentanglement for Emotional Talking Head Synthesis
    :house:project头部合成
  • EmoTalk3D: High-Fidelity Free-View Synthesis of Emotional 3D Talking Head
    :star:code
  • All You Need is Your Voice: Emotional Face Representation with Audio Perspective for Emotional Talking Face Generation
    :star:code
  • Audio-driven Talking Face Generation with Stabilized Synchronization Loss
  • Head360: Learning a Parametric 3D Full-Head for Free-View Synthesis in 360°
    :star:code
  • S^3D-NeRF: Single-Shot Speech-Driven Neural Radiance Field for High Fidelity Talking Head Synthesis
  • Gaussian3Diff: 3D Gaussian Diffusion for 3D Full Head Synthesis and Editing
    :house:project头部合成
  • Tri2-plane: Thinking Head Avatar via Feature Pyramid
    :house:project
  • Avatar Fingerprinting for Authorized Use of Synthetic Talking-Head Videos
    :house:project
  • TalkingGaussian: Structure-Persistent 3D Talking Head Synthesis via Gaussian Splatting
    :star:code3D 说话头合成
• 动画头部头像
  • HeadStudio: Text to Animatable Head Avatars with 3D Gaussian Splatting
    :star:code动画头部头像
  • HeadGaS: Real-Time Animatable Head Avatars via 3D Gaussian Splatting
  • 3D Gaussian Parametric Head Model
    :star:code头部
• 人脸超分辨
  • Kalman-Inspired Feature Propagation for Video Face Super-Resolution
    :house:project
• 人脸活体检测
  • TF-FAS: Twofold-Element Fine-Grained Semantic Guidance for Generalizable Face Anti-Spoofing
    :star:code
    :thumbsup:通过双重元素细粒度语义指导来增强泛化能力
  • DiffFAS: Face Anti-Spoofing via Generative Diffusion Models
    :star:code
  • Towards Unified Representation of Invariant-Specific Features in Missing Modality Face Anti-Spoofing
  • Bottom-Up Domain Prompt Tuning for Generalized Face Anti-Spoofing
• 头部合成
  • Portrait4D-v2: Pseudo Multi-View Data Creates Better 4D Head Synthesizer
    :star:code
    :house:project
  • Loc3Diff: Local Diffusion for 3D Human Head Synthesis and Editing
• 情绪识别
  • Upper-body Hierarchical Graph for Skeleton Based Emotion Recognition in Assistive Driving
    :star:code基于骨骼的情绪识别
• 人脸动作单元检测
  • AUFormer: Vision Transformers are Parameter-Efficient Facial Action Unit Detectors
    :star:code
• 假脸检测
  • Learning Natural Consistency Representation for Face Forgery Video Detection

2.3D Visual

• GroundUp: Rapid Sketch-Based 3D City Massing
  :house:project
• Ray-Distance Volume Rendering for Neural Scene Reconstruction
• HSR: Holistic 3D Human-Scene Reconstruction from Monocular Videos
  :house:project
• Decomposition of Neural Discrete Representations for Large-Scale 3D Mapping
  :star:code
• BlenderAlchemy: Editing 3D Graphics with Vision-Language Models
  :house:project
• Temporal Event Stereo via Joint Learning with Stereoscopic Flow
  :star:code
• GenRC: Generative 3D Room Completion from Sparse Image Collections
  :star:code
• Single-Photon 3D Imaging with Equi-Depth Photon Histograms
  :house:project
• Viewpoint textual inversion: discovering scene representations and 3D view control in 2D diffusion models
  :star:code
• SparseCraft: Few-Shot Neural Reconstruction through Stereopsis Guided Geometric Linearization
  :star:code
• 3D Congealing: 3D-Aware Image Alignment in the Wild
  :house:project
• ClusteringSDF: Self-Organized Neural Implicit Surfaces for 3D Decomposition
  :house:project
• BAGS: Blur Agnostic Gaussian Splatting through Multi-Scale Kernel Modeling
  :star:code
• Soft Shadow Diffusion (SSD): Physics-inspired Learning for 3D Computational Periscopy
• DPA-Net: Structured 3D Abstraction from Sparse Views via Differentiable Primitive Assembly
• An Optimization Framework to Enforce Multi-View Consistency for Texturing 3D Meshes
  :house:project
• Omniview-Tuning: Boosting Viewpoint Invariance of Vision-Language Pre-training Models
  :house:project
• Diffusion Model is a Good Pose Estimator from 3D RF-Vision
  :house:project
• Nuvo: Neural UV Mapping for Unruly 3D Representations
  :house:project
• MAP-ADAPT: Real-Time Quality-Adaptive Semantic 3D Maps
  :house:project
• MinD-3D: Reconstruct High-quality 3D objects in Human Brain
  :house:project3D
• UpFusion: Novel View Diffusion from Unposed Sparse View Observations
  :star:code3D
• MVS
  • Smoothness, Synthesis, and Sampling: Re-thinking Unsupervised Multi-View Stereo with DIV Loss
    :star:code
• 3D Visual Grounding
  • Empowering 3D Visual Grounding with Reasoning Capabilities
    :house:project
  • Multi-branch Collaborative Learning Network for 3D Visual Grounding
    :star:code
    :thumbsup:3DREC的Acc@0.5提高了 3.27%，3DRES的mIOU 提高了5.22%
  • ScanReason: Empowering 3D Visual Grounding with Reasoning Capabilities
    :star:code
  • Four Ways to Improve Verbo-visual Fusion for Dense 3D Visual Grounding
    :house:project
• Stereo Matching
  • Temporally Consistent Stereo Matching
    :star:code
  • Match-Stereo-Videos: Bidirectional Alignment for Consistent Dynamic Stereo Matching
  • Learning Representations from Foundation Models for Domain Generalized Stereo Matching立体匹配
• 3DGS
  • GaussReg: Fast 3D Registration with Gaussian Splatting
  • 3iGS: Factorised Tensorial Illumination for 3D Gaussian Splatting
  • Texture-GS: Disentangle the Geometry and Texture for 3D Gaussian Splatting Editing
    :star:code
  • Compact3D: Smaller and Faster Gaussian Splatting with Vector Quantization
    :star:code
    :house:project
  • CoR-GS: Sparse-View 3D Gaussian Splatting via Co-Regularization
    :star:code
    :house:project3DGS
  • End-to-End Rate-Distortion Optimized 3D Gaussian Representation
  • Deblurring 3D Gaussian Splatting
    :star:code
    :house:project
  • Per-Gaussian Embedding-Based Deformation for Deformable 3D Gaussian Splatting
    :house:project
  • HAC: Hash-grid Assisted Context for 3D Gaussian Splatting Compression
    :house:project
  • On the Error Analysis of 3D Gaussian Splatting and an Optimal Projection Strategy
    :star:code
  • Analytic-Splatting: Anti-Aliased 3D Gaussian Splatting via Analytic Integration
    :star:code
  • MVSGaussian: Fast Generalizable Gaussian Splatting Reconstruction from Multi-View Stereo
    :star:code
    :house:project
  • Street Gaussians: Modeling Dynamic Urban Scenes with Gaussian Splatting
    :star:code
  • DGD: Dynamic 3D Gaussians Distillation
    :house:project
  • EAGLES: Efficient Accelerated 3D Gaussians with Lightweight EncodingS
    :star:code
  • Revising Densification in Gaussian Splatting
  • HO-Gaussian: Hybrid Optimization of 3D Gaussian Splatting for Urban Scenes
  • RoDUS: Robust Decomposition of Static and Dynamic Elements in Urban Scenes
  • SWinGS: Sliding Windows for Dynamic 3D Gaussian Splatting
  • VEGS: View Extrapolation of Urban Scenes in 3D Gaussian Splatting using Learned Priors
    :star:code
  • MesonGS: Post-training Compression of 3D Gaussians via Efficient Attribute Transformation
  • MIGS: Multi-Identity Gaussian Splatting via Tensor Decomposition
    :star:code
  • SAGS: Structure-Aware 3D Gaussian Splatting
    :house:project
  • GGRt: Towards Generalizable 3D Gaussians without Pose Priors in Real-Time
    :house:project
  • Gaussian in the wild: 3D Gaussian Splatting for Unconstrained Image Collections
    :star:code
  • Pixel-GS Density Control with Pixel-aware Gradient for 3D Gaussian Splatting
    :star:code
  • WaSt-3D: Wasserstein-2 Distance for Scene-to-Scene Stylization on 3D Gaussians
    :house:project
  • MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Images
    :star:code
  • GS-LRM: Large Reconstruction Model for 3D Gaussian Splatting
    :house:project
  • DynMF: Neural Motion Factorization for Real-time Dynamic View Synthesis with 3D Gaussian Splatting
    :star:code
• 深度估计
  • Leveraging Near-Field Lighting for Monocular Depth Estimation from Endoscopy Videos
    :star:code
    :house:project
  • PatchRefiner: Leveraging Synthetic Data for Real-Domain High-Resolution Monocular Metric Depth Estimation
    :star:code
  • Remove Projective LiDAR Depthmap Artifacts via Exploiting Epipolar Geometry深度图伪影
  • Revisit Self-supervision with Local Structure-from-Motion
  • DoubleTake: Geometry Guided Depth Estimation
    :star:code
  • Physics-informed Knowledge Transfer for Underwater Monocular Depth Estimation
  • FutureDepth: Learning to Predict the Future Improves Video Depth Estimation
  • ProDepth: Boosting Self-Supervised Multi-Frame Monocular Depth with Probabilistic Fusion
    :star:code
  • Mono-ViFI: A Unified Learning Framework for Self-supervised Single- and Multi-frame Monocular Depth Estimation
    :star:code
  • Diffusion Models for Monocular Depth Estimation: Overcoming Challenging Conditions
    :star:code
    :star:code
  • High-Precision Self-Supervised Monocular Depth Estimation with Rich-Resource Prior
  • DiffusionDepth: Diffusion Denoising Approach for Monocular Depth Estimation
    :star:code
  • SEDiff: Structure Extraction for Domain Adaptive Depth Estimation via Denoising Diffusion Models
  • Camera Height Doesn't Change: Unsupervised Training for Metric Monocular Road-Scene Depth Estimation
    :house:project
  • GroCo: Ground Constraint for Metric Self-Supervised Monocular Depth
    :star:code深度估计
  • M2Depth: Self-supervised Two-Frame Multi-camera Metric Depth Estimation
    :house:project
  • Improving Domain Generalization in Self-Supervised Monocular Depth Estimation via Stabilized Adversarial Training
• 深度补全
  • Deep Cost Ray Fusion for Sparse Depth Video Completion
  • OGNI-DC: Robust Depth Completion with Optimization-Guided Neural Iterations
    :star:code
  • AugUndo: Scaling Up Augmentations for Monocular Depth Completion and Estimation
    :star:code
  • Sparse Beats Dense: Rethinking Supervision in Radar-Camera Depth Completion
    :star:code
• 表面重建
  • Surface Reconstruction from Gaussian Splatting via Novel Stereo Views
    :house:project
  • SG-NeRF: Neural Surface Reconstruction with Scene Graph Optimization
    :star:code
  • Sur2f: A Hybrid Representation for High-Quality and Efficient Surface Reconstruction from Multi-view Images
  • Improving Neural Surface Reconstruction with Feature Priors from Multi-View Image
  • DiffSurf: A Transformer-based Diffusion Model for Generating and Reconstructing 3D Surfaces in Pose
  • Rethinking Directional Parameterization in Neural Implicit Surface Reconstruction
  • PISR: Polarimetric Neural Implicit Surface Reconstruction for Textureless and Specular Objects
    :star:code
  • Surface Reconstruction for 3D Gaussian Splatting via Local Structural Hints
    :house:project
  • EMIE-MAP: Large-Scale Road Surface Reconstruction Based on Explicit Mesh and Implicit Encoding
  • GS2Mesh: Surface Reconstruction from Gaussian Splatting via Novel Stereo Views
    :house:project
  • Improving Neural Surface Reconstruction with Feature Priors from Multi-View Images
    :star:code
  • DynoSurf: Neural Deformation-based Temporally Consistent Dynamic Surface Reconstruction
    :star:code
  • Surface-Centric Modeling for High-Fidelity Generalizable Neural Surface Reconstruction
    :star:code
  • Parameterization-driven Neural Surface Reconstruction for Object-oriented Editing in Neural Rendering
    :house:project
• 三维重建
  • GSD: View-Guided Gaussian Splatting Diffusion for 3D Reconstruction
    :house:project
  • InfoNorm: Mutual Information Shaping of Normals for Sparse-View Reconstruction
    :star:code
  • fMRI-3D: A Comprehensive Dataset for Enhancing fMRI-based 3D Reconstruction
    :star:code
    :house:project
    :house:project
  • Reconstruction and Simulation of Elastic Objects with Spring-Mass 3D Gaussians
    :star:code
  • GRM: Large Gaussian Reconstruction Model for Efficient 3D Reconstruction and Generation
    :star:code
    :house:project
  • latentSplat: Autoencoding Variational Gaussians for Fast Generalizable 3D Reconstruction
    :star:code
  • MirrorGaussian: Reflecting 3D Gaussians for Reconstructing Mirror Reflections
    :house:project
  • Resolving Scale Ambiguity in Multi-view 3D Reconstruction using Dual-Pixel Sensors
    :star:code
  • 3D Reconstruction of Objects in Hands without Real World 3D Supervision
  • Human Hair Reconstruction with Strand-Aligned 3D Gaussians
    :house:project
  • SUP-NeRF: A Streamlined Unification of Pose Estimation and NeRF for Monocular 3D Object Reconstruction
    :star:code
  • MVDiffHD: A Dense High-resolution Multi-view Diffusion Model for Single or Sparse-view 3D Object Reconstruction(https://github.com/Tangshitao/MVDiffusion_plusplus)
  • NeuSDFusion: A Spatial-Aware Generative Model for 3D Shape Completion, Reconstruction, and Generation
    :house:project
  • Sketch2Vox: Learning 3D Reconstruction from a Single Monocular Sketch Image
    :sunflower:dataset
  • Analysis-by-Synthesis Transformer for Single-View 3D Reconstruction
    :star:code
• 三维形状
  • Synchronous Diffusion for Unsupervised Smooth Non-Rigid 3D Shape Matching
  • Transferable 3D Adversarial Shape Completion using Diffusion Models
  • Self-supervised Shape Completion via Involution and Implicit Correspondences
  • TetraDiffusion: Tetrahedral Diffusion Models for 3D Shape Generation
    :star:code
    :house:project
  • Learning Neural Deformation Representation for 4D Dynamic Shape Generation
  • AWOL: Analysis WithOut synthesis using Language3D shape
  • DiscoMatch: Fast Discrete Optimisation for Geometrically Consistent 3D Shape Matching
• 视频重建
  • Temporal Residual Guided Diffusion Framework for Event-Driven Video Reconstruction
• 四维重建
  • SplatFields: Neural Gaussian Splats for Sparse 3D and 4D Reconstruction
    :star:code
• 3D 纹理形状
  • High-Fidelity 3D Textured Shapes Generation by Sparse Encoding and Adversarial Decoding
    :star:code
    :house:project
  • Learning Pseudo 3D Guidance for View-consistent Texturing with 2D Diffusion文本驱动的 3D 纹理

1.Other(其它)

• Dataset Growth
  :star:code
• Adaptive Parametric Activation
  :star:code
• Nonverbal Interaction Detection
  :star:code
• Situated Instruction Following
  :house:project
  :house:project
• Optimizing Illuminant Estimation in Dual-Exposure HDR Imaging
• Unsupervised Exposure Correction
  :star:code
• Global Structure-from-Motion Revisited
  :star:code
• Fast Sprite Decomposition from Animated Graphics
  :house:project
• Sync from the Sea: Retrieving Alignable Videos from Large-Scale Datasets
  :house:project
• MERLiN: Single-Shot Material Estimation and Relighting for Photometric Stereo
• Enhancing Vectorized Map Perception with Historical Rasterized Maps
  :star:code
• Align before Collaborate: Mitigating Feature Misalignment for Robust Multi-Agent Perception
• DC-Solver: Improving Predictor-Corrector Diffusion Sampler via Dynamic Compensation
  :star:code
• Weight Conditioning for Smooth Optimization of Neural Networks
• Bones Can't Be Triangles: Accurate and Efficient Vertebrae Keypoint Estimation through Collaborative Error Revision
  :star:code
• SphereHead: Stable 3D Full-head Synthesis with Spherical Tri-plane Representation
  :house:project
• VisionLLaMA: A Unified LLaMA Backbone for Vision Tasks
  :star:code
• Global Counterfactual Directions
• Which Model Generated This Image? A Model-Agnostic Approach for Origin Attribution
• Computing the Lipschitz constant needed for fast scene recovery from CASSI measurements
• Pseudo-Labelling Should Be Aware of Disguising Channel Activations
• FMBoost: Boosting Latent Diffusion with Flow Matching
  :star:code
• Holodepth: Programmable Depth-Varying Projection via Computer-Generated Holography
  :house:project
• Adversarial Diffusion Distillation
• When and How do negative prompts take effect
• Preventing Catastrophic Forgetting through Memory Networks in Continuous Detection
• SCOD: From Heuristics to Theory
• Unsupervised Representation Learning by Balanced Self Attention Matching
  :star:code
• DualBEV: Unifying Dual View Transformation with Probabilistic Correspondences
  :star:code
• Linking in Style: Understanding learned features in deep learning models
  :star:code
• CliffPhys: Camera-based Respiratory Measurement using Clifford Neural Networks
• Synthesizing Environment-Specific People in Photographs
  :house:project
• Implicit Steganography Beyond the Constraints of Modality
• Energy-Clibrated VAE with Test Time Free Lunch
• Debiasing surgeon: fantastic weights and how to find them
• SparseRadNet: Sparse Perception Neural Network on Subsampled Radar Data
• Learning Where to Look: Self-supervised Viewpoint Selection for Active Localization using Geometrical Information
• Using My Artistic Style? You Must Obtain My Authorization
  :star:code
• IFTR: An Instance-Level Fusion Transformer for Visual Collaborative Perception
  :star:code
• Convex Relaxations for Manifold-Valued Markov Random Fields with Approximation Guarantees
• Adapting to Shifting Correlations with Unlabeled Data Calibration
• On Spectral Properties of Gradient-based Explanation Methods
• O2V-Mapping: Online Open-Vocabulary Mapping with Neural Implicit Representation
• A Framework for Efficient Model Evaluation through Stratification, Sampling, and Estimation
• Non-Line-of-Sight Estimation of Fast Human Motion with Slow Scanning Imagers
• Image Manipulation Detection With Implicit Neural Representation and Limited Supervision
• Adaptive Bounding Box Uncertainties via Two-Step Conformal Prediction
• GOEmbed: Gradient Origin Embeddings for Representation Agnostic 3D Feature Learning
  :house:project
• AddBiomechanics Dataset: Capturing the Physics of Human Motion at Scale
  :house:project
• Tight and Efficient Upper Bound on Spectral Norm of Convolutional Layers
• Learning Multimodal Latent Generative Models with Energy-Based Prior
• Hierarchical Conditioning of Diffusion Models Using Tree-of-Life for Studying Species Evolution
• Learning to Build by Building Your Own Instructions
  :star:code
• LNL+K: Enhancing Learning with Noisy Labels Through Noise Source Knowledge Integration
  :star:code
• Deep Online Probability Aggregation Clustering
• Camera Calibration using a Collimator System
  :star:code
• Asynchronous Bioplausible Neuron for Spiking Neural Networks for Event-Based Vision
• LITA: Language Instructed Temporal-Localization Assistant
  :star:code
• INTRA: Interaction Relationship-aware Weakly Supervised Affordance Grounding
  :house:project
• Elucidating the Hierarchical Nature of Behavior with Masked Autoencoders
  :star:code
• MetaAT: Active Testing for Label-Efficient Evaluation of Dense Recognition Tasks
• Generalizable Symbolic Optimizer Learning
  :star:code
• Training A Secure Model against Data-Free Model Extraction
• EraseDraw : Learning to Insert Objects by Erasing Them from Images
• AdaDiff: Accelerating Diffusion Models through Step-Wise Adaptive Computation
• Imaging with Confidence: Uncertainty Quantification for High-dimensional Undersampled MR Images
  :star:code
• Learning to Make Keypoints Sub-Pixel Accurate
  :star:code
• Explorative Inbetweening of Time and Space
  :house:project
• Salience-Based Adaptive Masking: Revisiting Token Dynamics for Enhanced Pre-training
  :star:code
• Instant Uncertainty Calibration of NeRFs Using a Meta-Calibrator
  :star:code
• Improving Robustness to Model Inversion Attacks via Sparse Coding Architectures
• Constructing Concept-based Models to Mitigate Spurious Correlations with Minimal Human Effort
• Neural Poisson Solver: A Universal and Continuous Framework for Natural Signal Blending
• Augmented Neural Fine-tuning for Efficient Backdoor Purification
• REDIR: Refocus-free Event-based De-occlusion Image Reconstruction
• Comprehensive Attribution: Inherently Explainable Vision Model with Feature Detector
  :star:code
• Pre-trained Visual Dynamics Representations for Efficient Policy Learning
• MMEarth: Exploring Multi-Modal Pretext Tasks For Geospatial Representation Learning
  :house:project
• MARs: Multi-view Attention Regularizations for Patch-based Feature Recognition of Space Terrain
  :house:project
• Diff-Reg: Diffusion Model in Doubly Stochastic Matrix Space for Registration Problem
  :star:code
• Hetecooper: Feature Collaboration Graph for Heterogeneous Collaborative Perception
• FARSE-CNN: Fully Asynchronous, Recurrent and Sparse Event-Based CNN
  :star:code
• Unmasking Bias in Diffusion Model Training
  :star:code
• Cross-Input Certified Training for Universal Perturbations
• Investigating Style Similarity in Diffusion Models
• Delving into Adversarial Robustness on Document Tampering Localization
  :star:code
• AMD: Automatic Multi-step Distillation of Large-scale Vision Models
• Learning Scalable Model Soup on a Single GPU: An Efficient Subspace Training Strategy
  :star:code
• JDT3D: Addressing the Gaps in LiDAR-Based Tracking-by-Attention
  :star:code
• SPARO: Selective Attention for Robust and Compositional Transformer Encodings for Vision
• Adaptive Annealing for Robust Averaging
• Lost in Translation: Modern Neural Networks Still Struggle With Small Realistic Image Transformations
• Generalizing to Unseen Domains via Text-guided Augmentation
• MO-EMT-NAS: Multi-Objective Continuous Transfer of Architectural Knowledge Between Tasks from Different Datasets
• Learning a Dynamic Privacy-preserving Camera Robust to Inversion Attacks
• MaxMI: A Maximal Mutual Information Criterion for Manipulation Concept Discovery
• CadVLM: Bridging Language and Vision in the Generation of Parametric CAD Sketches
• Towards Image Ambient Lighting Normalization
• Synthesizing Time-varying BRDFs via Latent Space
• HoloADMM: High-Quality Holographic Complex Field Recovery
• Fundamental Matrix Estimation Using Relative Depths
• MTaDCS: Moving Trace and Feature Density-based Confidence Sample Selection under Label Noise
  :star:code
• CipherDM: Secure Three-Party Inference for Diffusion Model Sampling
  :star:code
• Weighted Ensemble Models Are Strong Continual Learners
  :star:code
• Learning Equilibrium Transformation for Gamut Expansion and Color Restoration
  :star:code
• Implicit Neural Models to Extract Heart Rate from Video
  :house:project
• Learning Quantized Adaptive Conditions for Diffusion Models
• High-Fidelity Modeling of Generalizable Wrinkle Deformation
• Efficient Learning of Event-based Dense Representation using Hierarchical Memories with Adaptive Update
• SlimFlow: Training Smaller One-Step Diffusion Models with Rectified Flow
  :star:code
• DreamSampler: Unifying Diffusion Sampling and Score Distillation for Image Manipulation
  :star:code
• PosterLlama: Bridging Design Ability of Langauge Model to Content-Aware Layout Generation
• Integration of Global and Local Representations for Fine-grained Cross-modal Alignment
• Veil Privacy on Visual Data: Concealing Privacy for Humans, Unveiling for DNNs
• A high-quality robust diffusion framework for corrupted dataset
  :star:code
• FRDiff : Feature Reuse for Universal Training-free Acceleration of Diffusion Models
  :star:code
• Leveraging Imperfect Restoration for Data Availability Attack
  :star:code
• Oulu Remote-photoplethysmography Physical Domain Attacks Database (ORPDAD)
  :star:code
• Spiking Wavelet Transformer
  :star:code
• Hypernetworks for Generalizable BRDF Representation
  :house:project
• Solving the inverse problem of microscopy deconvolution with a residual Beylkin-Coifman-Rokhlin neural network
• Photon Inhibition for Energy-Efficient Single-Photon Imaging
  :house:project
• RANRAC: Robust Neural Scene Representations via Random Ray Consensus
  :house:project
• Characterizing Model Robustness via Natural Input Gradients
• Emerging Property of Masked Token for Effective Pre-training
• SWAG: Splatting in the Wild images with Appearance-conditioned Gaussians
• Curved Diffusion: A Generative Model With Optical Geometry Control
  :house:project
• Mini-Splatting: Representing Scenes with a Constrained Number of Gaussians
  :star:code
• Skeleton Recall Loss for Connectivity Conserving and Resource Efficient Segmentation of Thin Tubular Structures
  :star:code
• RAP: Retrieval-Augmented Planner for Adaptive Procedure Planning in Instructional Videos
• Rethinking Fast Adversarial Training: A Splitting Technique To Overcome Catastrophic Overfitting
• Optimization-based Uncertainty Attribution Via Learning Informative Perturbations
• CPT-VR: Improving Surface Rendering via Closest Point Transform with View-Reflection Appearance
• Think before Placement: Common Sense Enhanced Transformer for Object Placement
• Efficient Bias Mitigation Without Privileged Information
• Region-Native Visual Tokenization
  :star:code
• DiffCD: A Symmetric Differentiable Chamfer Distance for Neural Implicit Surface Fitting
  :star:code
• Efficient Neural Video Representation with Temporally Coherent Modulation
• Made to Order: Discovering monotonic temporal changes via self-supervised video ordering
  :star:code
• Concise Plane Arrangements for Low-Poly Surface and Volume Modelling
  :star:code
• ViPer: Visual Personalization of Generative Models via Individual Preference Learning
  :star:code
• How Far Can a 1-Pixel Camera Go? Solving Vision Tasks using Photoreceptors and Computationally Designed Visual Morphology
• Watching it in Dark: A Target-aware Representation Learning Framework for High-Level Vision Tasks in Low Illumination
  :star:code
• 3R-INN: How to be climate friendly while consuming/delivering videos
• Model Breadcrumbs: Scaling Multi-Task Model Merging with Sparse Masks
  :star:code
• Dynamic Guidance Adversarial Distillation with Enhanced Teacher Knowledge
  :star:code
• Idling Neurons, Appropriately Lenient Workload During Fine-tuning Leads to Better Generalization
• ConDense: Consistent 2D-3D Pre-training for Dense and Sparse Features from Multi-View Images
• Tokenize Anything via Prompting
  :star:code
  🤗huggingface
• Efficient Diffusion-Driven Corruption Editor for Test-Time Adaptation
  :star:code
• Long-CLIP: Unlocking the Long-Text Capability of CLIP
  :star:code
• Dolfin: Diffusion Layout Transformers without Autoencoder
• Beyond the Contact: Discovering Comprehensive Affordance for 3D Objects from Pre-trained 2D Diffusion Models
  :star:code
• Hybrid Video Diffusion Models with 2D Triplane and 3D Wavelet Representation
  :star:code
• Zero-Shot Image Feature Consensus with Deep Functional Maps
• LLaVA-Plus: Learning to Use Tools for Creating Multimodal Agents
  :house:project
• Scissorhands: Scrub Data Influence via Connection Sensitivity in Networks
  :star:code
• FuseTeacher: Modality-fused Encoders are Strong Vision Supervisors
  :star:code
• SmartControl: Enhancing ControlNet for Handling Rough Visual Conditions
  :house:project
• CoSIGN: Few-Step Guidance of ConSIstency Model to Solve General INverse Problems
  :star:code
• FedRA: A Random Allocation Strategy for Federated Tuning to Unleash the Power of Heterogeneous Clients
  :star:code
• Uncertainty Calibration with Energy Based Instance-wise Scaling in the Wild Dataset
  :star:code
• Learning to Enhance Aperture Phasor Field for Non-Line-of-Sight Imaging
  :star:code
• UniFS: Universal Few-shot Instance Perception with Point Representations
  :star:code
• Combining Generative and Geometry Priors for Wide-Angle Portrait Correction
  :star:code
• FlashTex: Fast Relightable Mesh Texturing with LightControlNet
  :house:project重新照明
• Consistent 3D Line Mapping
  :star:code
• RSL-BA: Rolling Shutter Line Bundle Adjustment
• Discovering Novel Actions from Open World Egocentric Videos with Object-Grounded Visual Commonsense Reasoning
• EAS-SNN: End-to-End Adaptive Sampling and Representation for Event-based Detection with Recurrent Spiking Neural Networks
  :star:code
• PairingNet: A Learning-based Pair-searching and -matching Network for Image Fragments
• Distributed Active Client Selection With Noisy Clients Using Model Association Scores
• Towards a Density Preserving Objective Function for Learning on Point Sets
• Task-Driven Uncertainty Quantification in Inverse Problems via Conformal Prediction
  :star:code
• SIGMA: Sinkhorn-Guided Masked Video Modeling
  :house:project
• LiDAR-Event Stereo Fusion with Hallucinations
  :star:code
  :house:project
• Dual-Camera Smooth Zoom on Mobile Phones
  :star:code
  :house:project
• Learning by Aligning 2D Skeleton Sequences and Multi-Modality Fusion
• Agent Attention: On the Integration of Softmax and Linear Attention
  :star:code
• Deep Feature Surgery: Towards Accurate and Efficient Multi-Exit Networks
  :star:code
• Grid-Attention: Enhancing Computational Efficiency of Large Vision Models without Fine-Tuning
  :star:code
• Customized Generation Reimagined: Fidelity and Editability Harmonized
  :star:code
• Temporal Residual Guided Diffusion Framework for Event-Driven Video Reconstruction视频重建
• Cascade-Zero123: One Image to Highly Consistent 3D with Self-Prompted Nearby Views
  :house:project
• Controlling the World by Sleight of Hand
• Probabilistic Weather Forecasting with Deterministic Guidance-based Diffusion Model
  :star:code概率天气预报
• Representing Topological Self-Similarity Using Fractal Feature Maps for Accurate Segmentation of Tubular Structures
  :star:code
• Functional Transform-Based Low-Rank Tensor Factorization for Multi-Dimensional Data Recovery
• G3R: Gradient Guided Generalizable Reconstruction
  :house:project
• SAIR: Learning Semantic-aware Implicit Representation
• Spectral Subsurface Scattering for Material Classification
• Instance-dependent Noisy-label Learning with Graphical Model Based Noise-rate Estimation
• Tracking Meets LoRA: Faster Training, Larger Model, Stronger Performance
  :star:code
• A Direct Approach to Viewing Graph Solvability
• Boosting the Power of Small Multimodal Reasoning Models to Match Larger Models with Self-Consistency Training
  :star:code
• Towards Multimodal Sentiment Analysis Debiasing via Bias Purification
• Improving Feature Stability during Upsampling -- Spectral Artifacts and the Importance of Spatial Context
• From Fake to Real: Pretraining on Balanced Synthetic Images to Prevent Spurious Correlations in Image Recognition
  :star:code
• Quantization-Friendly Winograd Transformations for Convolutional Neural Networks
• LetsMap: Unsupervised Representation Learning for Label-Efficient Semantic BEV Mapping
• M3DBench: Towards Omni 3D Assistant with Interleaved Multi-modal Instructions
• Leveraging Enhanced Queries of Point Sets for Vectorized Map Construction
  :star:code
• StereoGlue: Joint Feature Matching and Robust Estimation
  :star:code
• Factorized Diffusion: Perceptual Illusions by Noise Decomposition
  :house:project
• GIVT: Generative Infinite-Vocabulary Transformers
  :star:code
• Tiny Models are the Computational Saver for Large Models
• Unlocking Attributes' Contribution to Successful Camouflage: A Combined Textual and Visual Analysis Strategy
  :star:code
• SNeRV: Spectra-preserving Neural Representation for Video
  :star:code
• COMO: Compact Mapping and Odometry
  :star:code
• Multi-Sentence Grounding for Long-term Instructional Video
• Scene Coordinate Reconstruction: Posing of Image Collections via Incremental Learning of a Relocalizer
  :house:project
• Exact Diffusion Inversion via Bidirectional Integration Approximation
  :star:code
• McGrids: Monte Carlo-Driven Adaptive Grids for Iso-Surface Extraction
• Regulating Model Reliance on Non-Robust Features by Smoothing Input Marginal Density
  :star:code
• Dynamic Data Selection for Efficient SSL via Coarse-to-Fine Refinement
• ZeST: Zero-Shot Material Transfer from a Single Image
  :star:code
• PCF-Lift: Panoptic Lifting by Probabilistic Contrastive Fusion
• SemGrasp: Semantic Grasp Generation via Language Aligned Discretization
  :house:project
• DragAPart: Learning a Part-Level Motion Prior for Articulated Objects
  :house:project
• Superpixel-informed Implicit Neural Representation for Multi-Dimensional Data
• Physics-Free Spectrally Multiplexed Photometric Stereo under Unknown Spectral Composition
• Robust Fitting on a Gate Quantum Computer
• Edge-Guided Fusion and Motion Augmentation for Event-Image Stereo
• Mahalanobis Distance-based Multi-view Optimal Transport for Multi-view Crowd Localization
  :house:project
• On the Vulnerability of Skip Connections to Model Inversion Attacks
  :star:code
• Taming CLIP for Fine-grained and Structured Visual Understanding of Museum Exhibits
  :star:code
• GMM-IKRS: Gaussian Mixture Models for Interpretable Keypoint Refinement and Scoring
• ConDense: Consistent 2D/3D Pre-training for Dense and Sparse Features from Multi-View Images
• Does Data-Efficient Generalization Exacerbate Bias in Foundation Models?
• InfMAE: A Foundation Model in The Infrared Modality红外
• Teach CLIP to Develop a Number Sense for Ordinal Regression
  :star:code
• GlobalPointer: Large-Scale Plane Adjustment with Bi-Convex Relaxation
  :star:code
• ColorMAE: Exploring data-independent masking strategies in Masked AutoEncoders
• Scalar Function Topology Divergence: Comparing Topology of 3D Objects
• OneRestore: A Universal Restoration Framework for Composite Degradation
  :star:code
• RoofDiffusion: Constructing Roofs from Severely Corrupted Point Data via Diffusion
• Binomial Self-compensation for Motion Error in Dynamic 3D Scanning
• Encapsulating Knowledge in One Prompt
  :star:code
• iMatching: Imperative Correspondence Learning
• An Adaptive Screen-Space Meshing Approach for Normal Integration
• Efficient Pre-training for Localized Instruction Generation of Procedural Videos
  :star:code
• Shape from Heat Conduction
• Put Myself in Your Shoes: Lifting the Egocentric Perspective from Exocentric Videos
• Optimal Transport of Diverse Unsupervised Tasks for Robust Learning from Noisy Few-Shot Data
• Finding Visual Task Vectors
  :star:code
• Occupancy as Set of Points
  :star:code
• Learning to Robustly Reconstruct Dynamic Scenes from Low-light Spike Streams
• AID-AppEAL: Automatic Image Dataset and Algorithm for Content Appeal Enhancement and Assessment Labeling
  :star:code
• Retargeting Visual Data with Deformation Fields
• Delving Deep into Engagement Prediction of Short Videos
  :star:code
• Temporal-Mapping Photography for Event Cameras
  :star:code
• Six-Point Method for Multi-Camera Systems with Reduced Solution Space
  :star:code
• BK-SDM: A Lightweight, Fast, and Cheap Version of Stable Diffusion
  :star:code
• Physical-Based Event Camera Simulator
  :star:code
• REFRAME: Reflective Surface Real-Time Rendering for Mobile Devices
  :star:code
• Self-Training Room Layout via Geometry-aware Ray-casting
• Closed-Loop Unsupervised Representation Disentanglement with β-VAE Distillation and Diffusion Probabilistic Feedback
• UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding
  :star:code
• EventBind: Learning a Unified Representation to Bind Them All for Event-based Open-world Understanding
  :star:code
• Where am I? Scene Retrieval with Language
• Event Camera Data Dense Pre-training
• Unsqueeze [CLS] Bottleneck to Learn Rich Representations
  :star:code
• VeCLIP: Improving CLIP Training via Visual-enriched Captions
  :star:code
• Spike-Temporal Latent Representation for Energy-Efficient Event-to-Video Reconstruction
• Catastrophic Overfitting: A Potential Blessing in Disguise
• Diffusion Reward: Learning Rewards via Conditional Video Diffusion
  :house:project
• Data-to-Model Distillation: Data-Efficient Learning Framework
• Neural graphics texture compression supporting random access
• ReMatching: Low-Resolution Representations for Scalable Shape Correspondence
• EgoPet: Egomotion and Interaction Data from an Animal's Perspective
  :house:project
• This Probably Looks Exactly Like That: An Invertible Prototypical Network
  :star:code
• Revisiting Feature Disentanglement Strategy in Diffusion Training and Breaking Conditional Independence Assumption in Sampling
• ArtVLM: Attribute Recognition Through Vision-Based Prefix Language Modeling
  :star:code属性识别
• Stream Query Denoising for Vectorized HD-Map Construction
• PartCraft: Crafting Creative Objects by Parts
  :star:code
• ControlNet++: Improving Conditional Controls with Efficient Consistency Feedback
  :star:code
• Dropout Mixture Low-Rank Adaptation for Visual Parameters-Efficient Fine-Tuning
• UNIC: Universal Classification Models via Multi-teacher Distillation
• Efficient Training of Spiking Neural Networks with Multi-Parallel Implicit Stream Architecture
  :star:code尖峰神经网络
• Visual Prompting via Partial Optimal Transport
• E3V-K5: An Authentic Benchmark for Redefining Video-Based Energy Expenditure Estimation
  :star:code
• Understanding Physical Dynamics with Counterfactual World Modeling
  :house:project
• 4Diff: 3D-Aware Diffusion Model for Third-to-First Viewpoint Translation
  :house:project
• Revisiting Calibration of Wide-Angle Radially Symmetric Cameras
  :star:code相机校准
• STAG4D: Spatial-Temporal Anchored Generative 4D Gaussians
  :star:code
• Synchronization of Projective Transformations
• UniCal: Unified Neural Sensor Calibration
  :house:project
• Rawformer: Unpaired Raw-to-Raw Translation for Learnable Camera ISPs
  :star:code
• Robust Incremental Structure-from-Motion with Hybrid Features
• Any2Point: Empowering Any-modality Transformers for Efficient 3D Understanding
  :star:code
• CompGS: Smaller and Faster Gaussian Splatting with Vector Quantization
  :star:code
• Multiscale Graph Texture Network
  :star:code
• Enhancing Optimization Robustness in 1-bit Neural Networks through Stochastic Sign Descent
  :star:code
• Domain Reduction Strategy for Non-Line-of-Sight Imaging
  :star:code
• BI-MDRG: Bridging Image History in Multimodal Dialogue Response Generation
  :star:code
• Mamba-ND: Selective State Space Modeling for Multi-Dimensional Data
  :star:code
• Model Stock: All we need is just a few fine-tuned models
  :star:code
• DreamStruct: Understanding Slides and User Interfaces via Synthetic Data Generation
• DetailSemNet: Elevating Signature Verification through Detail-Semantic Integration
  :star:code
• SLIM: Spuriousness Mitigation with Minimal Human Annotations
  :star:code
• Scaling Backwards: Minimal Synthetic Pre-training?
  :star:code
• On the Evaluation Consistency of Attribution-based Explanations
  :star:code
• GTP-4o: Modality-prompted Heterogeneous Graph Learning for Omni-modal Biomedical Representation
  :star:code
• OvSW: Overcoming Silent Weights for Accurate Binary Neural Networks
  :star:code
• SHINE: Saliency-aware HIerarchical NEgative Ranking for Compositional Temporal Grounding
  :star:code
• ReGround: Improving Textual and Spatial Grounding at No Cost
  :house:project
• ComboVerse: Compositional 3D Assets Creation Using Spatially-Aware Diffusion Guidance
  :house:project
• WHAC: World-grounded Humans and Cameras
  :house:project
• Unlocking Attributes' Contribution to Successful Camouflage: A Combined Textual and VisualAnalysis Strategy
  :star:code
• Neural Metamorphosis
  :house:project
• Light-in-Flight for a World-in-Motion
• Learning with Unmasked Tokens Drives Stronger Vision Learners
  :star:code
• PSALM: Pixelwise Segmentation with Large Multi-modal Model
  :star:code
• InsMapper: Exploring Inner-instance Information for Vectorized HD Mapping
  :star:code
• The All-Seeing Project V2: Towards General Relation Comprehension of the Open World
  :star:code
• Refine, Discriminate and Align: Stealing Encoders via Sample-Wise Prototypes and Multi-Relational Extraction
• Multi-Task Domain Adaptation for Language Grounding with 3D Objects
  :house:project
• QueryCDR: Query-based Controllable Distortion Rectification Network for Fisheye Images
  :star:code鱼眼图像
• BAMM: Bidirectional Autoregressive Motion Model
  :house:project
• Handling The Non-Smooth Challenge in Tensor SVD: A Multi-Objective Tensor Recovery Framework
• Latent-INR: A Flexible Framework for Implicit Representations of Videos with Discriminative Semantics
• RPBG: Towards Robust Neural Point-based Graphics in the Wild
  :star:code
• Memory-Efficient Fine-Tuning for Quantized Diffusion Model
  :star:code
• Towards Architecture-Agnostic Untrained Networks Priors for Image Reconstruction with Frequency Regularization
  :star:code
• Similarity of Neural Architectures using Adversarial Attack Transferability
• NeuroPictor: Refining fMRI-to-Image Reconstruction via Multi-individual Pretraining and Multi-level Modulation
  :house:project
• Robustness Preserving Fine-tuning using Neuron Importance
• A Riemannian Approach for Spatiotemporal Analysis and Generation of 4D Tree-shaped Structures
• Dual-Path Adversarial Lifting for Domain Shift Correction in Online Test-time Adaptation
  :star:code
• FTBC: Forward Temporal Bias Correction for Optimizing ANN-SNN Conversion
• Test-time Model Adaptation for Image Reconstruction Using Self-supervised Adaptive Layers图像重建
• Unveiling Privacy Risks in Stochastic Neural Networks Training: Effective Image Reconstruction from Gradients图像重建
• CrossScore: A Multi-View Approach to Image Evaluation and Scoring
• ADMap: Anti-disturbance Framework for Vectorized HD Map Construction
• GaussianImage: 1000 FPS Image Representation and Compression by 2D Gaussian Splatting
  :star:code
• PYRA: Parallel Yielding Re-Activation for Training-Inference Efficient Task Adaptation
  :star:code
• ReALFRED: An Embodied Instruction Following Benchmark in Photo-Realistic Environments
  :star:code
• DVLO: Deep Visual-LiDAR Odometry with Local-to-Global Feature Fusion and Bi-Directional Structure Alignment
  :star:code
• UL-VIO: Ultra-lightweight Visual-Inertial Odometry with Noise Robust Test-time Adaptation视觉惯性里程计
• Real-data-driven 2000 FPS Color Video from Mosaicked Chromatic Spikes
  🤗huggingface
• RoGUENeRF: A Robust Geometry-Consistent Universal Enhancer for NeRF
  :house:project
• LaRa: Efficient Large-Baseline Radiance Fields
  :star:code
• Bi-TTA: Bidirectional Test-Time Adapter for Remote Physiological Measurement
  :house:project
• ELSE: Efficient Deep Neural Network Inference through Line-based Sparsity Exploration
• Open-World Dynamic Prompt and Continual Visual Representation Learning
• GeoCalib: Learning Single-image Calibration with Geometric Optimization
  :star:code
• LEIA: Latent View-invariant Embeddings for Implicit 3D Articulation
  :star:code
• Alignist: CAD-Informed Orientation Distribution Estimation by Fusing Shape and Correspondences
• Weakly-supervised Camera Localization by Ground-to-satellite Image Registration
• Learning Neural Volumetric Pose Features for Camera Localization
  :house:project
• SPVLoc: Semantic Panoramic Viewport Matching for 6D Camera Localization in Unseen Environments
  :star:code
• DECOLLAGE: 3D Detailization by Controllable, Localized, and Learned Geometry Enhancement
  :star:code
• Event-based Mosaicing Bundle Adjustment
  :star:code
• Reprojection Errors as Prompts for Efficient Scene Coordinate Regression
• Depth on Demand: Streaming Dense Depth from a Low Frame Rate Active Sensor
• AMEGO: Active Memory from long EGOcentric videos
  :star:code
• Vista3D: Unravel the 3D Darkside of a Single Image
  :star:code
• Agglomerative Token Clustering
  :house:project
• Formula-Supervised Visual-Geometric Pre-training
  :star:code
• Interpretability-Guided Test-Time Adversarial Defense
  :star:code
• Towards Model-Agnostic Dataset Condensation by Heterogeneous Models
  :star:code
• MVPGS: Excavating Multi-view Priors for Gaussian Splatting from Sparse Input Views
  :star:code
• Intrinsic Single-Image HDR Reconstruction
• Disentangled Generation and Aggregation for Robust Radiance Fields
  :star:code
• Mixture of Efficient Diffusion Experts Through Automatic Interval and Sub-Network Selection
• Commonly Interesting Images
• Sequential Representation Learning via Static-Dynamic Conditional Disentanglement
• QuasiSim: Parameterized Quasi-Physical Simulators for Dexterous Manipulations Transfer
  :star:code
  :house:project
• Dataset Distillation by Automatic Training Trajectories
  :star:code
• Neural Graphics Texture Compression Supporting Random Acces
• LookupViT: Compressing visual information to a limited number of tokens
• Bridging the Gap: Studio-like Avatar Creation from a Monocular Phone Capture
  :house:project
• Generating 3D House Wireframes with Semantics
  :star:code
  :house:project
• Flying with Photons: Rendering Novel Views of Propagating Light
  :star:code
• Deep Nets with Subsampling Layers Unwittingly Discard Useful Activations at Test-Time
  :star:code
• MobileNetV4: Universal Models for the Mobile Ecosystem
• Gravity-aligned Rotation Averaging with Circular Regression
  :star:code
• Concept Sliders: LoRA Adaptors for Precise Control in Diffusion Models
  :house:project
• HYPE: Hyperbolic Entailment Filtering for Underspecified Images and Texts
• DoughNet: A Visual Predictive Model for Topological Manipulation of Deformable Objects
  :house:project
• TrajPrompt: Aligning Color Trajectory with Vision-Language Representations
  :star:code
• DomainFusion: Generalizing To Unseen Domains with Latent Diffusion Models
• Toward Tiny and High-quality Facial Makeup with Data Amplify Learning
  :star:code
• Multi-Label Cluster Discrimination for Visual Representation Learning
• Preventing Catastrophic Overfitting in Fast Adversarial Training: A Bi-level Optimization Perspective
• MemBN: Robust Test-Time Adaptation via Batch Norm with Statistics Memory
• SeiT++: Masked Token Modeling Improves Storage-efficient Training
  :star:code
• MagicEraser: Erasing Any Objects via Semantics-Aware Control
• Reliable Spatial-Temporal Voxels For Multi-Modal Test-Time Adaptation
  :house:project
• A Cephalometric Landmark Regression Method based on Dual-encoder for High-resolution X-ray Image
  :star:code
• Resilience of Entropy Model in Distributed Neural Networks
  :star:code
• GeoWizard: Unleashing the Diffusion Priors for 3D Geometry Estimation from a Single Image
  :house:project
• MotionChain: Conversational Motion Controllers via Multimodal Prompts
• MacDiff: Unified Skeleton Modeling with Masked Conditional Diffusion
  :house:project
• Improving Intervention Efficacy via Concept Realignment in Concept Bottleneck Models
  :star:code
• Brain Netflix: Scaling Data to Reconstruct Videos from Brain Signals
• Latent-INR: A Flexible Framework for Implicit Representations of Videos with Discriminative Semantics
• Tensorial template matching for fast cross-correlation with rotations and its application for tomography
• SelfGeo: Self-supervised and Geodesic-consistent Estimation of Keypoints on Deformable Shapes
  :star:code
• Explain via Any Concept: Concept Bottleneck Model with Open Vocabulary Concepts
• Motion and Structure from Event-based Normal Flow
  :house:project
• SENC: Handling Self-collision in Neural Cloth Simulation
• Distribution Alignment for Fully Test-Time Adaptation with Dynamic Online Data Streams
• Animate Your Motion: Turning Still Images into Dynamic Videos
  :house:project
• Gaussian Splatting on the Move:Blur and Rolling Shutter Compensation for Natural Camera Motion
  :star:code
  :house:project
• Relightable Neural Actor with Intrinsic Decomposition and Pose Control
  :house:project
• Layer-Wise Relevance Propagation with Conservation Property for ResNet
  :house:project
• Self-Rectifying Diffusion Sampling with Perturbed-Attention Guidance
  :house:project
• SparseSSP: 3D Subcellular Structure Prediction from Sparse-View Transmitted Light Images
• ViG-Bias: Visually Grounded Bias Discovery and Mitigation
• DOCCI: Descriptions of Connected and Contrasting Images
  :house:project
• Geometry Fidelity for Spherical Images
• Efficient Inference of Vision Instruction-Following Models with Elastic Cache
  :star:code
• Mew: Multiplexed Immunofluorescence Image Analysis through an Efficient Multiplex Network
  :star:code
• Topology-Preserving Downsampling of Binary Images
• Quality Assured: Rethinking Annotation Strategies in Imaging AI
• Chronologically Accurate Retrieval for Temporal Grounding of Motion-Language Models
  :house:project
• Data Collection-free Masked Video Modeling
• Efficient Image Pre-Training with Siamese Cropped Masked Autoencoders
  :star:code
• Möbius Transform for Mitigating Perspective Distortions in Representation Learning
  :house:project
• Foster Adaptivity and Balance in Learning with Noisy Labels
  :star:code
  无需先验知识即可高效解决深度学习中的噪声标签问题，让模型性能和鲁棒性大幅提升！
• Solving Motion Planning Tasks with a Scalable Generative Model
  :star:code
• 4D Contrastive Superflows are Dense 3D Representation Learners
  :star:code
• Learning to Complement and to Defer to Multiple Users
  :star:code
• Shedding More Light on Robust Classifiers under the lens of Energy-based Models
  :star:code
• TIP: Tabular-Image Pre-training for Multimodal Classification with Incomplete Data
  :star:code
• UMBRAE: Unified Multimodal Brain Decoding
  :star:code
  :house:project
• Trainable Highly-expressive Activation Functions
  :star:code
• Controllable Navigation Instruction Generation with Chain of Thought Prompting
• Recursive Visual Programming
  :star:code
• Reshaping the Online Data Buffering and Organizing Mechanism for Continual Test-Time Adaptation
  :star:code
• Imaging Interiors: An Implicit Solution to Electromagnetic Inverse Scattering Problems
  :star:code
• The Gaussian Discriminant Variational Autoencoder (GdVAE): A Self-Explainable Model with Counterfactual Explanations
  :star:code
  :house:project
• HyperSpaceX: Radial and Angular Exploration of HyperSpherical Dimensions
• DataDream: Few-shot Guided Dataset Generation
  :star:code
• Aligning Neuronal Coding of Dynamic Visual Scenes with Foundation Vision Models
  :star:code
  :house:project
• Towards Robust Event-based Networks for Nighttime via Unpaired Day-to-Night Event Translation
  :star:code
• FRI-Net: Floorplan Reconstruction via Room-wise Implicit Representation
• Hierarchically Structured Neural Bones for Reconstructing Animatable Objects from Casual Videos
  :star:code
• Deep Diffusion Image Prior for Efficient OOD Adaptation in 3D Inverse Problems
  :star:code
• Pathformer3D: A 3D Scanpath Transformer for 360° Images
  :star:code
• Kinetic Typography Diffusion Model
  :star:code
• PolyRoom: Room-aware Transformer for Floorplan Reconstruction
  :star:code
• Tree-D Fusion: Simulation-Ready Tree Dataset from Single Images with Diffusion Priors
• Multiscale Sliced Wasserstein Distances as Perceptual Color Difference Measures
  :star:code
• Augmented Neural Fine-Tuning for Efficient Backdoor Purification
• Improving Hyperbolic Representations via Gromov-Wasserstein Regularization
• Mutual Learning for Acoustic Matching and Dereverberation via Visual Scene-driven Diffusion
  :star:code
• Efficient Training with Denoised Neural Weights
  :star:code
• SpaceJAM: a Lightweight and Regularization-free Method for Fast Joint Alignment of Images
  :star:code
• Discover-then-Name: Task-Agnostic Concept Bottlenecks via Automated Concept Discovery
  :star:code
• SceneScript: Reconstructing Scenes With An Autoregressive Structured Language Model
  :house:project
• Multi-modal Relation Distillation for Unified 3D Representation Learning
• Continual Learning for Remote Physiological Measurement: Minimize Forgetting and Simplify Inference
  :star:code
• TreeSBA: Tree-Transformer for Self-Supervised Sequential Brick Assembly
• SIGMA:Sinkhorn-Guided Masked Video Modeling
  :star:code
• Attention Beats Linear for Fast Implicit Neural Representation Generation
  :star:code
• Text2Place: Affordance-aware Text Guided Human Placement
  :star:code
• RoadPainter: Points Are Ideal Navigators for Topology transformER
• STAMP: Outlier-Aware Test-Time Adaptation with Stable Memory Replay
  :star:code
• Differentiable Convex Polyhedra Optimization from Multi-view Images
  :star:code
• A Diffusion Model for Simulation Ready Coronary Anatomy with Morpho-skeletal Control
  :star:code
• Power Variable Projection for Initialization-Free Large-Scale Bundle Adjustment
• Multi-label Cluster Discrimination for Visual Representation Learning
• SINDER: Repairing the Singular Defects of DINOv2
  :star:code
• SHIC: Shape-Image Correspondences with no Keypoint Supervision
  :house:project
• Semicalibrated Relative Pose from an Affine Correspondence and Monodepth相对位姿半校准
• Scalable Group Choreography via Variational Phase Manifold Learning
• Deep Companion Learning: Enhancing Generalization Through Historical Consistency
• Revisit Event Generation Model: Self-Supervised Learning of Event-to-Video Reconstruction with Implicit Neural Representations
  :star:code
• Neural Surface Detection for Unsigned Distance Fields
• Merging and Splitting Diffusion Paths for Semantically Coherent Panoramas
  :star:code
• Platypus: A Generalized Specialist Model for Reading Text in Various Forms
  :star:code
• RAW-Adapter: Adapting Pre-trained Visual Model to Camera RAW Images
  :star:code
• Learning Differentially Private Diffusion Models via Stochastic Adversarial Distillation
• Affine steerers for structured keypoint description
• SeA: Semantic Adversarial Augmentation for Last Layer Features from Unsupervised Representation Learning
  :star:code
• MMBench: Is Your Multi-modal Model an All-around Player?
  :star:code
  :house:project
• DenseNets Reloaded: Paradigm Shift Beyond ResNets and ViTs
  :star:code
• PreLAR: World Model Pre-training with Learnable Action Representation
  :star:code
• Dataset Enhancement with Instance-Level Augmentations
  :star:code
• Non-parametric Sensor Noise Modeling and Synthesis
• Stripe Observation Guided Inference Cost-free Attention Mechanism
  :star:code
• Leveraging Hierarchical Feature Sharing for Efficient Dataset Condensation
• Object-Aware NIR-to-Visible Translation
  :star:code
  :sunflower:datasetLow-level Vision

2020 年论文分类汇总戳这里

↘️CVPR-2020-Papers ↘️ECCV-2020-Papers

2021 年论文分类汇总戳这里

↘️ICCV-2021-Papers ↘️CVPR-2021-Papers

2022 年论文分类汇总戳这里

↘️CVPR-2022-Papers ↘️WACV-2022-Papers ↘️ECCV-2022-Papers

2023 年论文分类汇总戳这里

↘️CVPR-2023-Papers ↘️WACV-2023-Papers ↘️ICCV-2023-Papers

扫码CV君微信(注明：CVPR)入微信交流群：

# ECCV-2024-Papers