ECCV-2022-Papers

官网链接：https://eccv2022.ecva.net/

截稿日期：2022年3月7日(9:59PM CET, 11:59AM PST)

会议日期：2022年10月24日-2022年10月28日

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

2022 年论文分类汇总戳这里

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

2021年论文分类汇总戳这里

↘️ICCV-2021-Papers ↘️CVPR-2021-Papers

2020 年论文分类汇总戳这里

↘️CVPR-2020-Papers ↘️ECCV-2020-Papers

❣❣❣另外打包下载ECCV 2022论文，可在【我爱计算机视觉】微信公众号后台回复“paper”。共计 1645 篇。分类完成

:cat:	:dog:	:tiger:	:wolf:
1.其它	2.Image Segmentation(图像分割)	3.Image Progress(图像处理)	4.Image Captioning(图像字幕)
5.Image/Video Retrieval(图像/视频检索)	6.Object Detection(目标检测)	7.Object Tracking(目标跟踪)	8.3D(三维视觉)
9.Human Pose Estimation(人体姿态估计)	10.Pose Estimation(物体姿势估计)	11.Video	12.Action Detection(人体动作检测与识别)
13.Human-Object Interaction(人物交互)	14.Visual Answer Questions(视觉问答)	15.Vision-Language(视觉语言)	16.Transformer
17.GAN	18.Image-to-Image Translation(图像到图像翻译)	19.Image Synthesis/Generation(图像合成)	20.Face(人脸)
21.Semi/self-supervised learning(半/自监督)	22.OCR	23.Medical Image(医学影像)	24.UAV/Remote Sensing/Satellite Image(无人机/遥感/卫星图像)
25.Autonomous vehicles(自动驾驶)	26.Video/Image Super-Resolution(视频/图像超分辨率)	27.Image Classification(图像分类)	28.Neural Architecture Search(神经架构搜索)
29.Re-identification(重识别)	30.Optical Flow(光流)	31.SLAM/Augmented Reality/Virtual Reality/Robotics(增强/虚拟现实/机器人)	32.Point Cloud(点云)
33.Model Compression/Knowledge Distillation/Pruning(模型压缩/知识蒸馏/剪枝)	34.Meta-Learning(元学习)	35.Feature Learning(联邦学习)	36.Machine Learning(机器学习)
37.Open-set Recognition(开集识别)	38.Contrastive Learning(对比学习)	39.Transfer Learning(迁移学习)	40.Adversarial Learning(对抗学习)
41.Incremental Learning(增量学习)	42.Reinforcement Learning(强化学习)	43.Lifelong Learning(终生学习)	44.Active Learning(主动学习)
45.Metric Learning(度量学习)	46.Continual Learning(持续学习)	47.GNN/GCN(图神经网络)	48.Semantic Correspondence(语义对应)
49.Few/Zero-Shot Learning/Domain Generalization/Adaptation(小/零样本/域泛化/适应)	50.Neural Rendering(渲染)	51.Anomaly Detection(异常检测)	52.Scene Flow Estimation(场景流估计)
53.Dataset(数据集)	54.View Generation(视图生成)	55.Style Transfer(风格迁移)	56.Sound
57.Scene Graph Generation(场景图生成)	58.Human Motion Prediction(人体动作预测)	59.Image Matching(图像匹配)	60.Data Augmentation(数据增强)
61.Light Field(光学、几何、光场成像)

:trophy::trophy::trophy: 获奖论文

• 最佳论文奖
  • On the Versatile Uses of Partial Distance Correlation in Deep Learning
    :star:code
• 最佳论文荣誉奖
  • A Level Set Theory for Neural Implicit Evolution under Explicit Flows
  • Pose-NDF: Modeling Human Pose Manifolds with Neural Distance Fields
    :star:code
• Koenderink Prize (test of time)
  • A naturalistic open source movie for optical flow evaluation
  • Indoor Segmentation and Support Inference from RGBD Images
• Best Demo Award
  • [Using a Smartphone for Augmented Reality in a Classroom]
    :tv:video
• Everingham Prize
  • 【The UCF101 and HMD51 dataset teams】&【Walter J. Scheirer 】

61.Light Field(光学、几何、光场成像)

• 相机相关
  • Learned Monocular Depth Priors in Visual-Inertial Initialization
  • 相机姿态估计
    • E-Graph: Minimal Solution for Rigid Rotation with Extensibility Graphs
    • Camera Pose Estimation and Localization with Active Audio Sensing
• 相机姿势
  • Camera Pose Auto-Encoders for Improving Pose Regression
    :star:code
• 相机估计
  • A Reliable Online Method for Joint Estimation of Focal Length and Camera Rotation
    :star:code
• 相机自动校准
  • Camera Auto-Calibration from the Steiner Conic of the Fundamental Matrix
• 事件相机
  • DVS-Voltmeter: Stochastic Process-Based Event Simulator for Dynamic Vision Sensors
    :star:code
  • Selection and Cross Similarity for Event-Image Deep Stereo
    :star:code
• 相机重识别
  • SC-wLS: Towards Interpretable Feed-forward Camera Re-localization
    :star:code
• 相机定位
  • Towards Accurate Active Camera Localization
    :star:code
• 光场
  • Neural Light Field Estimation for Street Scenes with Differentiable Virtual Object Insertion
    :house:project
  • Synthesizing Light Field Video from Monocular Video
    :star:code
  • NeILF: Neural Incident Light Field for Physically-Based Material Estimation

60.Data Augmentation(数据增强)

• TokenMix: Rethinking Image Mixing for Data Augmentation in Vision Transformers
  :star:code
• Neuromorphic Data Augmentation for Training Spiking Neural Networks
• 3D Random Occlusion and Multi-layer Projection for Deep Multi-Camera Pedestrian Localization
  :star:code

59.Image Matching(图像匹配)

• ASpanFormer: Detector-Free Image Matching with Adaptive Span Transformer
  :house:project
• ECO-TR: Efficient Correspondences Finding Via Coarse-to-Fine Refinement
  :star:code:house:project

58.Human Motion Prediction(人体动作预测)

• ERA: Expert Retrieval and Assembly for Early Action Prediction
• Overlooked Poses Actually Make Sense: Distilling Privileged Knowledge for Human Motion Prediction
• GIMO: Gaze-Informed Human Motion Prediction in Context
  :star:code
• Diverse Human Motion Prediction Guided by Multi-level Spatial-Temporal Anchors
  :star:code
• 行动预测
  • Rethinking Learning Approaches for Long-Term Action Anticipation
    :star:code
• 运动估计
  • PREF: Predictability Regularized Neural Motion Fields
    :open_mouth:oral
• 人体运动合成
  • Learning Uncoupled-Modulation CVAE for 3D Action-Conditioned Human Motion Synthesis
  • MotionCLIP: Exposing Human Motion Generation to CLIP Space
    :star:code

57.Scene Graph Generation(场景图生成)

• Panoptic Scene Graph Generation
  :star:code:house:project
• Meta Spatio-Temporal Debiasing for Video Scene Graph Generation
• Hierarchical Memory Learning for Fine-Grained Scene Graph Generation
• Fine-Grained Scene Graph Generation with Data Transfer
  :star:code
• Towards Open-Vocabulary Scene Graph Generation with Prompt-Based Finetuning

56.Sound

• Learning Visual Styles from Audio-Visual Associations
  :house:project
• Active Audio-Visual Separation of Dynamic Sound Sources
  :house:project
• 声源定位
  • Localizing Visual Sounds the Easy Way
    :star:code
• 有源扬声器检测
  • End-to-End Active Speaker Detection
• 音频驱动的视频肖像生成
  • Semantic-Aware Implicit Neural Audio-Driven Video Portrait Generation
    :open_mouth:oral:house:project
• 视听分割
  • Audio-Visual Segmentation
    :star:code
  • Audio—Visual Segmentation
    :star:code
• 语音合成
  • VisageSynTalk: Unseen Speaker Video-to-Speech Synthesis via Speech-Visage Feature Selection
• 声音分离
  • AudioScopeV2: Audio-Visual Attention Architectures for Calibrated Open-Domain On-Screen Sound Separation
  • VoViT: Low Latency Graph-Based Audio-Visual Voice Separation Transformer
    :house:project

55.Style Transfer(风格迁移)

• CCPL: Contrastive Coherence Preserving Loss for Versatile Style Transfer
  :open_mouth:oral:star:code
• Learning Graph Neural Networks for Image Style Transfer
• ARF: Artistic Radiance Fields
  :house:project
• 图像风格化
  • WISE: Whitebox Image Stylization by Example-based Learning
    :star:code
• 发型迁移
  • Style Your Hair: Latent Optimization for Pose-Invariant Hairstyle Transfer via Local-Style-Aware Hair Alignment
    :star:code

54.View Generation(视图生成)

• InfiniteNature-Zero: Learning Perpetual View Generation of Natural Scenes from Single Images
  :open_mouth:oral
• CompNVS: Novel View Synthesis with Scene Completion
• HDR-Plenoxels: Self-Calibrating High Dynamic Range Radiance Fields
  :star:code
• Neural Radiance Transfer Fields for Relightable Novel-View Synthesis with Global Illumination
• R2L: Distilling Neural Radiance Field to Neural Light Field for Efficient Novel View Synthesis
  :house:project
• NeXT: Towards High Quality Neural Radiance Fields via Multi-Skip Transformer
  :star:code

53.Dataset(数据集)

• The Abduction of Sherlock Holmes: A Dataset for Visual Abductive Reasoning
  :sunflower:dataset
• Responsive Listening Head Generation: A Benchmark Dataset and Baseline
  :sunflower:dataset
• Online Segmentation of LiDAR Sequences: Dataset and Algorithm
  :sunflower:dataset
• COO: Comic Onomatopoeia Dataset for Recognizing Arbitrary or Truncated Texts
  :star:code
  用于识别任意或截断文本的漫画拟声词数据集
• BRACE: The Breakdancing Competition Dataset for Dance Motion Synthesis
  :sunflower:dataset
  用于舞蹈动作合成的霹雳舞比赛数据集
• CelebV-HQ: A Large-Scale Video Facial Attributes Dataset
  :sunflower:dataset:house:project
  一个大规模的视频人脸属性数据集
• UnrealEgo: A New Dataset for Robust Egocentric 3D Human Motion Capture
  :star:code:house:project
  用于鲁棒性以自我为中心的三维人类运动捕捉的新数据集
• BEAT: A Large-Scale Semantic and Emotional Multi-modal Dataset for Conversational Gestures Synthesis
  :sunflower:dataset
  :newspaper:ECCV 2022 | 76小时动捕，最大规模数字人多模态数据集开源
• MovieCuts: A New Dataset and Benchmark for Cut Type Recognition
  :sunflower:dataset
  剪切类型识别
• A Real World Dataset for Multi-View 3D Reconstruction
  :sunflower:dataset
  三维重建
• Capturing, Reconstructing, and Simulating: The UrbanScene3D Dataset
  :sunflower:dataset
  城市场景重建
• PartImageNet: A Large, High-Quality Dataset of Parts
  分割
• A-OKVQA: A Benchmark for Visual Question Answering Using World Knowledge
  :sunflower:dataset
  VQA
• OOD-CV: A Benchmark for Robustness to Out-of-Distribution Shifts of Individual Nuisances in Natural Images
• The Anatomy of Video Editing: A Dataset and Benchmark Suite for AI-Assisted Video Editing
  :sunflower:dataset
  视频编辑
• ClearPose: Large-Scale Transparent Object Dataset and Benchmark
  :sunflower:dataset
  深度估计
• AnimeCeleb: Large-Scale Animation CelebHeads Dataset for Head Reenactment
  :sunflower:dataset
  动画名人头像数据集
• A Dense Material Segmentation Dataset for Indoor and Outdoor Scene Parsing
  用于室内和室外场景解析的密集材料分割数据集
• MimicME: A Large Scale Diverse 4D Database for Facial Expression Analysis
  用于面部表情分析的大规模多样化4D数据库
• Delving into Universal Lesion Segmentation: Method, Dataset, and Benchmark
  :sunflower:dataset
  病变分割

52.Scene Flow Estimation(场景流估计)

• Bi-PointFlowNet: Bidirectional Learning for Point Cloud Based Scene Flow Estimation
  :star:code
• What Matters for 3D Scene Flow Network
  :star:code
• MonoPLFlowNet: Permutohedral Lattice FlowNet for Real-Scale 3D Scene Flow Estimation with Monocular Images

51.Anomaly Detection(异常检测)

• Registration based Few-Shot Anomaly Detection
  :open_mouth:oral:star:code
• Dynamic Local Aggregation Network with Adaptive Clusterer for Anomaly Detection
  :star:code
• DSR – A Dual Subspace Re-Projection Network for Surface Anomaly Detection
  :star:code
• Locally Varying Distance Transform for Unsupervised Visual Anomaly Detection
• SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentation
  :star:code
• HaloAE: An HaloNet based Local Transformer Auto-Encoder for Anomaly Detection and Localization
• Hierarchical Semi-Supervised Contrastive Learning for Contamination-Resistant Anomaly Detection
  :star:code
• 表面异常检测
  • DSR -- A dual subspace re-projection network for surface anomaly detection
    :star:code

50.Neural Rendering(渲染)

• Relighting4D: Neural Relightable Human from Videos
  :star:code:house:project:tv:video
• MPIB: An MPI-Based Bokeh Rendering Framework for Realistic Partial Occlusion Effects
  :star:code:house:project
• NeuMan: Neural Human Radiance Field from a Single Video
  :star:code
• Approximate Differentiable Rendering with Algebraic Surfaces
  :star:code:house:project
• AdaNeRF: Adaptive Sampling for Real-time Rendering of Neural Radiance Fields
  :star:code:house:project
• Generalizable Patch-Based Neural Rendering
  :open_mouth:oral:star:code:house:project
• Deforming Radiance Fields with Cages
  :star:code:house:project
• NeuMesh: Learning Disentangled Neural Mesh-based Implicit Field for Geometry and Texture Editing
  :open_mouth:oral:star:code:house:project
• ActiveNeRF: Learning where to See with Uncertainty Estimation
  :star:code
• ARAH: Animatable Volume Rendering of Articulated Human SDFs
  :star:code:house:project
• LaTeRF: Label and Text Driven Object Radiance Fields
• MoFaNeRF: Morphable Facial Neural Radiance Field
  :star:code
• Conditional-Flow NeRF: Accurate 3D Modelling with Reliable Uncertainty Quantification
• Sem2NeRF: Converting Single-View Semantic Masks to Neural Radiance Fields
  :star:code
• KeypointNeRF: Generalizing Image-Based Volumetric Avatars Using Relative Spatial Encoding of Keypoints
  :house:project
• ViewFormer: NeRF-Free Neural Rendering from Few Images Using Transformers
  :star:code
• GeoAug: Data Augmentation for Few-Shot NeRF with Geometry Constraints
• SinNeRF: Training Neural Radiance Fields on Complex Scenes from a Single Image
  :star:code
• BungeeNeRF: Progressive Neural Radiance Field for Extreme Multi-Scale Scene Rendering

49.Few/Zero-Shot Learning/Domain Generalization/Adaptation(小/零样本/域泛化/适应)

• 小样本
  • Cross-Domain Cross-Set Few-Shot Learning via Learning Compact and Aligned Representations
    :star:code
  • Self-Supervision Can Be a Good Few-Shot Learner
    :star:code
  • VizWiz-FewShot: Locating Objects in Images Taken by People With Visual Impairments
    :house:project
  • Contrastive Prototypical Network with Wasserstein Confidence Penalty
    :star:code
  • tSF: Transformer-Based Semantic Filter for Few-Shot Learning
  • Worst Case Matters for Few-Shot Recognition
  • Learning Instance and Task-Aware Dynamic Kernels for Few-Shot Learning
  • Self-Promoted Supervision for Few-Shot Transformer
    :star:code
  • Coarse-to-Fine Incremental Few-Shot Learning
  • Improving Few-Shot Learning through Multi-task Representation Learning Theory
  • TransVLAD: Focusing on Locally Aggregated Descriptors for Few-Shot Learning
  • Kernel Relative-Prototype Spectral Filtering for Few-Shot Learning
    :star:code
  • Uncertainty-DTW for Time Series and Sequences
• 零样本
  • Temporal and cross-modal attention for audio-visual zero-shot learning
    :star:code
  • 3D Compositional Zero-Shot Learning with DeCompositional Consensus
  • Learning Invariant Visual Representations for Compositional Zero-Shot Learning
    :star:code
• 域适应
  • Prior Knowledge Guided Unsupervised Domain Adaptation
    :star:code
  • MoDA: Map Style Transfer for Self-Supervised Domain Adaptation of Embodied Agents
  • CoSMix: Compositional Semantic Mix for Domain Adaptation in 3D LiDAR Segmentation
    :star:code
  • GIPSO: Geometrically Informed Propagation for Online Adaptation in 3D LiDAR Segmentation
    :star:code
  • Prototype-Guided Continual Adaptation for Class-Incremental Unsupervised Domain Adaptation
    :star:code
  • MemSAC: Memory Augmented Sample Consistency for Large Scale Domain Adaptation
    :star:code:house:project
  • Concurrent Subsidiary Supervision for Unsupervised Source-Free Domain Adaptation
    :star:code:house:project
  • Combating Label Distribution Shift for Active Domain Adaptation
  • Uncertainty-guided Source-free Domain Adaptation
    :star:code
  • Learning Unbiased Transferability for Domain Adaptation by Uncertainty Modeling
  • Unknown-Oriented Learning for Open Set Domain Adaptation
  • Burn after Reading: Online Adaptation for Cross-Domain Streaming Data
    :star:code
  • Adversarial Partial Domain Adaptation by Cycle Inconsistency
  • A Broad Study of Pre-training for Domain Generalization and Adaptation
  • Interpretable Open-Set Domain Adaptation via Angular Margin Separation
  • Contrastive Vicinal Space for Unsupervised Domain Adaptation
    :star:code
  • Incomplete Multi-View Domain Adaptation via Channel Enhancement and Knowledge Transfer
  • BMD: A General Class-Balanced Multicentric Dynamic Prototype Strategy for Source-Free Domain Adaptation
    :star:code
• 域泛化
  • Grounding Visual Representations with Texts for Domain Generalization
    :star:code
  • Improving Test-Time Adaptation via Shift-agnostic Weight Regularization and Nearest Source Prototypes
  • Attention Diversification for Domain Generalization
    :star:code
  • Cross-Domain Ensemble Distillation for Domain Generalization
  • Domain Generalization by Mutual-Information Regularization with Pre-trained Models
    :star:code
  • MVDG: A Unified Multi-View Framework for Domain Generalization
    :star:code

48.Semantic Correspondence(语义对应)

• Demystifying Unsupervised Semantic Correspondence Estimation
  :star:code:house:project
• Learning Semantic Correspondence with Sparse Annotations

47.GNN/GCN(图神经网络)

• GCN
  • End-to-end Graph-constrained Vectorized Floorplan Generation with Panoptic Refinement
  • Learning Self-Prior for Mesh Denoising Using Dual Graph Convolutional Networks
    :star:code
  • An Efficient Person Clustering Algorithm for Open Checkout-Free Groceries
• GNN
  • Adversarial Label Poisoning Attack on Graph Neural Networks via Label Propagation
  • Equivariant Hypergraph Neural Networks
    :star:code

46.Continual Learning(持续学习)

• Balancing Stability and Plasticity through Advanced Null Space in Continual Learning
  :open_mouth:oral
• Online Continual Learning with Contrastive Vision Transformer
• Helpful or Harmful: Inter-Task Association in Continual Learning
• Theoretical Understanding of the Information Flow on Continual Learning Performance
  :star:code
• Transfer without Forgetting
  :star:code
• incDFM: Incremental Deep Feature Modeling for Continual Novelty Detection
• Online Task-Free Continual Learning with Dynamic Sparse Distributed Memory
  :star:code
• Discriminability-Transferability Trade-Off: An Information-Theoretic Perspective
• CoSCL: Cooperation of Small Continual Learners Is Stronger than a Big One
  :star:code
• DualPrompt: Complementary Prompting for Rehearsal-Free Continual Learning
  :star:code

45.Metric Learning(度量学习)

• DAS: Densely-Anchored Sampling for Deep Metric Learning
  :star:code
• Posterior Refinement on Metric Matrix Improves Generalization Bound in Metric Learning
• A Non-Isotropic Probabilistic Take On Proxy-Based Deep Metric Learning
  :star:code

44.Active Learning(主动学习)

• When Active Learning Meets Implicit Semantic Data Augmentation
• PT4AL: Using Self-Supervised Pretext Tasks for Active Learning
  :star:code

43.Lifelong Learning(终生学习)

• Anti-Retroactive Interference for Lifelong Learning
  :star:code

42.Reinforcement Learning(强化学习)

• Style-Agnostic Reinforcement Learning
  :star:code
• StARformer: Transformer with State-Action-Reward Representations for Visual Reinforcement Learning
  :star:code
• Learning Efficient Multi-agent Cooperative Visual Exploration
  :house:project
• DexMV: Imitation Learning for Dexterous Manipulation from Human Videos
  :house:project

41.Incremental Learning(增量学习)

• Learning with Recoverable Forgetting
• Incremental Task Learning with Incremental Rank Updates
  :star:code
• DLCFT: Deep Linear Continual Fine-Tuning for General Incremental Learning
• 类增量
  • Class-incremental Novel Class Discovery
    :star:code
  • Long-Tailed Class Incremental Learning
    :star:code
  • Few-Shot Class-Incremental Learning via Entropy-Regularized Data-Free Replay
  • Few-Shot Class-Incremental Learning from an Open-Set Perspective
    :star:code
  • Class-Incremental Learning with Cross-Space Clustering and Controlled Transfer
    :star:code:house:project
  • R-DFCIL: Relation-Guided Representation Learning for Data-Free Class Incremental Learning
    :star:code
  • FOSTER: Feature Boosting and Compression for Class-Incremental Learning
  • S3C: Self-Supervised Stochastic Classifiers for Few-Shot Class-Incremental Learning
    :star:code

40.Adversarial Learning(对抗学习)

• Prior-Guided Adversarial Initialization for Fast Adversarial Training
  :star:code
  :newspaper:ECCV 2022 | 一种基于先验指导的对抗样本初始化方法
• BIPS: Bi-modal Indoor Panorama Synthesis via Residual Depth-Aided Adversarial Learning
  :star:code
• Decoupled Adversarial Contrastive Learning for Self-supervised Adversarial Robustness
  :open_mouth:oral:star:code
• RIBAC: Towards Robust and Imperceptible Backdoor Attack against Compact DNN
  :star:code
• Adversarial Coreset Selection for Efficient Robust Training
• Shape Matters: Deformable Patch Attack
• Enhanced Accuracy and Robustness via Multi-Teacher Adversarial Distillation
  :star:code
• GradAuto: Energy-Oriented Attack on Dynamic Neural Networks
  :star:code
• Learning Energy-Based Models with Adversarial Training
• Revisiting Outer Optimization in Adversarial Training
• One Size Does NOT Fit All: Data-Adaptive Adversarial Training
  :star:code
• UniCR: Universally Approximated Certified Robustness via Randomized Smoothing
• ℓ∞-Robustness and Beyond:Unleashing Efficient Adversarial Training
• Towards Efficient Adversarial Training on Vision Transformers
• FrequencyLowCut Pooling - Plug & Play against Catastrophic Overfitting
  :star:code
• TAFIM: Targeted Adversarial Attacks against Facial Image Manipulations
  :house:project
• 对抗攻击
  • Frequency Domain Model Augmentation for Adversarial Attack
    :star:code
  • Watermark Vaccine: Adversarial Attacks to Prevent Watermark Removal
    :star:code
  • SegPGD: An Effective and Efficient Adversarial Attack for Evaluating and Boosting Segmentation Robustness
  • Scaling Adversarial Training to Large Perturbation Bounds
    :star:code
  • Towards Effective and Robust Neural Trojan Defenses via Input Filtering
  • Exploiting the Local Parabolic Landscapes of Adversarial Losses to Accelerate Black-Box Adversarial Attack
    :star:code
  • Robust Network Architecture Search via Feature Distortion Restraining
  • Triangle Attack: A Query-Efficient Decision-Based Adversarial Attack
    :star:code
  • Adaptive Image Transformations for Transfer-Based Adversarial Attack
• 黑盒
  • Black-Box Dissector: Towards Erasing-Based Hard-Label Model Stealing Attack
  • An Invisible Black-Box Backdoor Attack through Frequency Domain
    :star:code
• 白盒
  • Harmonizer: Learning to Perform White-Box Image and Video Harmonization
• 对抗样本
  • Boosting Transferability of Targeted Adversarial Examples via Hierarchical Generative Networks
    :star:code

39.Transfer Learning(迁移学习)

• Factorizing Knowledge in Neural Networks
  :star:code
• SecretGen: Privacy Recovery on Pre-trained Models via Distribution Discrimination
  :star:code
• How Stable Are Transferability Metrics Evaluations?
• Language-Driven Artistic Style Transfer
• MultiMAE: Multi-modal Multi-task Masked Autoencoders
  :house:project

38.Contrastive Learning(对比学习)

• Network Binarization via Contrastive Learning
  :star:code
• Adversarial Contrastive Learning via Asymmetric InfoNCE
  :star:code
• Fast-MoCo: Boost Momentum-based Contrastive Learning with Combinatorial Patches
  :star:code
• Contrastive Learning for Diverse Disentangled Foreground Generation
• Decoupled Contrastive Learning
• Joint Learning of Localized Representations from Medical Images and Reports
• Contrasting Quadratic Assignments for Set-Based Representation Learning
• Generative Subgraph Contrast for Self-Supervised Graph Representation Learning
  :star:code

37.Open-set Recognition(开集识别)

• DenseHybrid: Hybrid Anomaly Detection for Dense Open-set Recognition
• Difficulty-Aware Simulator for Open Set Recognition
  :star:code

36.Machine Learning(机器学习)

• Predicting is not Understanding: Recognizing and Addressing Underspecification in Machine Learning

35.Feature Learning(联邦学习)

• SphereFed: Hyperspherical Federated Learning
• Image Coding for Machines with Omnipotent Feature Learning
• Addressing Heterogeneity in Federated Learning via Distributional Transformation
  :star:code
• FedLTN: Federated Learning for Sparse and Personalized Lottery Ticket Networks
• Improving Generalization in Federated Learning by Seeking Flat Minima
• AdaBest: Minimizing Client Drift in Federated Learning via Adaptive Bias Estimation

34.Meta-Learning(元学习)

• Bitwidth-Adaptive Quantization-Aware Neural Network Training: A Meta-Learning Approach
• Meta-Learning with Less Forgetting on Large-Scale Non-Stationary Task Distributions
• Learning to Weight Samples for Dynamic Early-exiting Networks
  :star:code
• Rethinking Clustering-Based Pseudo-Labeling for Unsupervised Meta-Learning
  :star:code

33.Model Compression/Knowledge Distillation/Pruning(模型压缩/知识蒸馏/剪枝)

• 知识蒸馏
  • Knowledge Condensation Distillation
    :star:code
  • FedX: Unsupervised Federated Learning with Cross Knowledge Distillation
    :star:code
  • Black-box Few-shot Knowledge Distillation
    :star:code
  • Efficient One Pass Self-distillation with Zipf's Label Smoothing
    :star:code
  • MixSKD: Self-Knowledge Distillation from Mixup for Image Recognition
    :star:code
  • Switchable Online Knowledge Distillation
    :star:code
  • Distilling the Undistillable: Learning from a Nasty Teacher
    :star:code
  • Masked Generative Distillation
    :star:code
  • DistPro: Searching a Fast Knowledge Distillation Process via Meta Optimization
    :star:code
  • Personalized Education: Blind Knowledge Distillation
    :star:code
  • Prune Your Model before Distill It
    :star:code
  • IDa-Det: An Information Discrepancy-Aware Distillation for 1-Bit Detectors
    :star:code
  • Deep Ensemble Learning by Diverse Knowledge Distillation for Fine-Grained Object Classification
  • A Fast Knowledge Distillation Framework for Visual Recognition
    :house:project
  • Self-Regulated Feature Learning via Teacher-Free Feature Distillation
    :house:project
• 量化
  • Synergistic Self-supervised and Quantization Learning
    :open_mouth:oral:star:code
  • PalQuant: Accelerating High-precision Networks on Low-precision Accelerators
    :star:code
  • Fine-Grained Data Distribution Alignment for Post-Training Quantization
    :star:code
  • Symmetry Regularization and Saturating Nonlinearity for Robust Quantization
  • Mixed-Precision Neural Network Quantization via Learned Layer-Wise Importance
  • Non-uniform Step Size Quantization for Accurate Post-Training Quantization
    :star:code
  • Towards Accurate Network Quantization with Equivalent Smooth Regularizer
  • Explicit Model Size Control and Relaxation via Smooth Regularization for Mixed-Precision Quantization
  • BASQ: Branch-Wise Activation-Clipping Search Quantization for Sub-4-Bit Neural Networks
    :star:code
  • RDO-Q: Extremely Fine-Grained Channel-Wise Quantization via Rate-Distortion Optimization
  • PTQ4ViT: Post-Training Quantization for Vision Transformers with Twin Uniform Quantization
• 剪枝
  • FairGRAPE: Fairness-aware GRAdient Pruning mEthod for Face Attribute Classification
    :star:code
  • Bayesian Optimization with Clustering and Rollback for CNN Auto Pruning
    :star:code
  • Trainability Preserving Neural Structured Pruning
    :star:code
  • Interpretations Steered Network Pruning via Amortized Inferred Saliency Maps
    :star:code
  • Data-Free Backdoor Removal Based on Channel Lipschitzness
    :star:code
  • Multi-Granularity Pruning for Model Acceleration on Mobile Devices
  • Ensemble Knowledge Guided Sub-network Search and Fine-Tuning for Filter Pruning
  • Soft Masking for Cost-Constrained Channel Pruning
    :star:code
  • Towards Ultra Low Latency Spiking Neural Networks for Vision and Sequential Tasks Using Temporal Pruning
  • CPrune: Compiler-Informed Model Pruning for Efficient Target-Aware DNN Execution
  • Filter Pruning via Feature Discrimination in Deep Neural Networks
• 轻量级
  • Learning Extremely Lightweight and Robust Model with Differentiable Constraints on Sparsity and Condition Number
• MC
  • Patch Similarity Aware Data-Free Quantization for Vision Transformers
    :star:code
  • Disentangled Differentiable Network Pruning
  • Weight Fixing Networks
  • SPViT: Enabling Faster Vision Transformers via Latency-Aware Soft Token Pruning
    :star:code

32.Point Cloud(点云)

• Few 'Zero Level Set'-Shot Learning of Shape Signed Distance Functions in Feature Space
• FH-Net: A Fast Hierarchical Network for Scene Flow Estimation on Real-World Point Clouds
  :star:code
• Dynamic 3D Scene Analysis by Point Cloud Accumulation
  :star:code:house:project
• Point Cloud Compression with Sibling Context and Surface Priors
• LESS: Label-Efficient Semantic Segmentation for LiDAR Point Clouds
• Point MixSwap: Attentional Point Cloud Mixing via Swapping Matched Structural Divisions
  :star:code
• MORE: Multi-Order RElation Mining for Dense Captioning in 3D Scenes
  :star:code
• Bottom Up Top down Detection Transformers for Language Grounding in Images and Point Clouds
  :house:project
• PointTree: Transformation-Robust Point Cloud Encoder with Relaxed K-D Trees
  :star:code
• Learning to Generate Realistic LiDAR Point Clouds
  :house:project
• PseudoAugment: Learning to Use Unlabeled Data for Data Augmentation in Point Clouds
• SPE-Net: Boosting Point Cloud Analysis via Rotation Robustness Enhancement
  :star:code
• Resolution-Free Point Cloud Sampling Network with Data Distillation
  :star:code
• diffConv: Analyzing Irregular Point Clouds with an Irregular View
  :star:code
• GraphFit: Learning Multi-Scale Graph-Convolutional Representation for Point Cloud Normal Estimation
  :star:code
• Quasi-Balanced Self-Training on Noise-Aware Synthesis of Object Point Clouds for Closing Domain Gap
  :star:code
• PD-Flow: A Point Cloud Denoising Framework with Normalizing Flows
  :star:code
• Shape-Pose Disentanglement Using SE(3)-Equivariant Vector Neurons
• Revisiting Point Cloud Simplification: A Learnable Feature Preserving Approach
• Masked Autoencoders for Point Cloud Self-Supervised Learning
  :star:code
• Masked Discrimination for Self-Supervised Learning on Point Clouds
  :star:code
• Meta-Sampler: Almost-Universal yet Task-Oriented Sampling for Point Clouds
  :star:code
• Efficient Point Cloud Analysis Using Hilbert Curve
• RFNet-4D: Joint Object Reconstruction and Flow Estimation from 4D Point Clouds
  :star:code
• 3D点云
  • Autoregressive 3D Shape Generation via Canonical Mapping
    :star:code
  • Point Cloud Domain Adaptation via Masked Local 3D Structure Prediction
    :star:code
  • Exploring the Devil in Graph Spectral Domain for 3D Point Cloud Attacks
    :star:code
  • Unsupervised Learning of 3D Semantic Keypoints with Mutual Reconstruction
    :star:code
  • Few-Shot Class-Incremental Learning for 3D Point Cloud Objects
    :star:code
  • Image2Point: 3D Point-Cloud Understanding with 2D Image Pretrained Models
    :star:code
  • Manifold Adversarial Learning for Cross-Domain 3D Shape Representation
• 点云定位
  • CPO: Change Robust Panorama to Point Cloud Localization
• 点云分割
  • Dual Adaptive Transformations for Weakly Supervised Point Cloud Segmentation
  • Efficient Point Cloud Segmentation with Geometry-Aware Sparse Networks
• 点云补全
  • SeedFormer: Patch Seeds based Point Cloud Completion with Upsample Transformer
    :star:code
  • FBNet: Feedback Network for Point Cloud Completion
    :open_mouth:oral:star:code
  • Optimization over Disentangled Encoding: Unsupervised Cross-Domain Point Cloud Completion via Occlusion Factor Manipulation
    :star:code
• 点云配准
  • SuperLine3D: Self-supervised Line Segmentation and Description for LiDAR Point Cloud
    :star:code
  • Improving RGB-D Point Cloud Registration by Learning Multi-scale Local Linear Transformation
    :star:code
  • PointCLM: A Contrastive Learning-Based Framework for Multi-Instance Point Cloud Registration
    :star:code
  • PCR-CG: Point Cloud Registration via Deep Explicit Color and Geometry
• 点云重建
  • Learning to Train a Point Cloud Reconstruction Network without Matching
    :star:code
• 点云分类
  • Improving Adversarial Robustness of 3D Point Cloud Classification Models
    :star:code
• 点云理解
  • PointMixer: MLP-Mixer for Point Cloud Understanding
    :star:code

31.SLAM/Augmented Reality/Virtual Reality/Robotics(增强/虚拟现实/机器人)

• 增强现实
  • LaMAR: Benchmarking Localization and Mapping for Augmented Reality
    :star:code:house:project
• VR
  • LiP-Flow: Learning Inference-Time Priors for Codec Avatars via Normalizing Flows in Latent Space
  • human volumetric capture(容积捕获)
    • AvatarCap: Animatable Avatar Conditioned Monocular Human Volumetric Capture
      :star:code:house:project
• 虚拟试穿
  • Single Stage Virtual Try-on via Deformable Attention Flows
  • Dress Code: High-Resolution Multi-Category Virtual Try-On
    :star:code
  • High-Resolution Virtual Try-On with Misalignment and Occlusion-Handled Conditions
    :star:code
• 视觉定位(相机姿势估计)
  • MeshLoc: Mesh-Based Visual Localization
    :star:code
• 机器人
  • Visual Cross-View Metric Localization with Dense Uncertainty Estimates
    :star:code

30.Optical Flow(光流)

• Secrets of Event-Based Optical Flow
  :star:code
• Deep 360∘ Optical Flow Estimation Based on Multi-Projection Fusion
• Learning Omnidirectional Flow in 360-degree Video via Siamese Representation
  :house:project
• Video Interpolation by Event-driven Anisotropic Adjustment of Optical Flow
• Learning Omnidirectional Flow in 360° Video via Siamese Representation
  :house:project
• FlowFormer: A Transformer Architecture for Optical Flow
• Complementing Brightness Constancy with Deep Networks for Optical Flow Prediction
• Disentangling Architecture and Training for Optical Flow
  :house:project
• A Perturbation-Constrained Adversarial Attack for Evaluating the Robustness of Optical Flow
  :open_mouth:oral:star:code
• Optical Flow Training under Limited Label Budget via Active Learning
  :star:code
• S2F2: Single-Stage Flow Forecasting for Future Multiple Trajectories Prediction
• Semi-Supervised Learning of Optical Flow by Flow Supervisor
  :star:code
• Deep 360° Optical Flow Estimation Based on Multi-Projection Fusion

29.Re-identification(重识别)

• 重识别
  • Negative Samples are at Large: Leveraging Hard-distance Elastic Loss for Re-identification
  • PASS: Part-Aware Self-Supervised Pre-training for Person Re-identification
    :star:code
  • Adaptive Cross-Domain Learning for Generalizable Person Re-identification
    :star:code
  • Dynamically Transformed Instance Normalization Network for Generalizable Person Re-identification
  • Mimic Embedding via Adaptive Aggregation: Learning Generalizable Person Re-identification
    :star:code
  • Modality Synergy Complement Learning with Cascaded Aggregation for Visible-Infrared Person Re-identification
    :star:code
  • Cross-Modality Transformer for Visible-Infrared Person Re-identification
  • Optimal Transport for Label-Efficient Visible-Infrared Person Re-identification
    :star:code
  • Counterfactual Intervention Feature Transfer for Visible-Infrared Person Re-identification
• 行人搜索
  • OIMNet++: Prototypical Normalization and Localization-aware Learning for Person Search
    :house:project
  • A Simple and Robust Correlation Filtering Method for Text-Based Person Search
    :star:code
  • Domain Adaptive Person Search
    :open_mouth:oral:star:code
• 人群计数
  • An End-to-End Transformer Model for Crowd Localization
    :house:project
  • Completely Self-Supervised Crowd Counting via Distribution Matching
  • Calibration-Free Multi-View Crowd Counting
• Visual Search
  • Target-Absent Human Attention
    :star:code
• 步态识别
  • MetaGait: Learning to Learn an Omni Sample Adaptive Representation for Gait Recognition
  • GaitEdge: Beyond Plain End-to-End Gait Recognition for Better Practicality
    :star:code

28.Neural Architecture Search(神经架构搜索)

• SuperTickets: Drawing Task-Agnostic Lottery Tickets from Supernets via Jointly Architecture Searching and Parameter Pruning
  :star:code
• UniNet: Unified Architecture Search with Convolution, Transformer, and MLP
  :star:code
• ScaleNet: Searching for the Model to Scale
  :star:code
• CLOSE: Curriculum Learning On the Sharing Extent Towards Better One-shot NAS
  :star:code
• Towards Regression-Free Neural Networks for Diverse Compute Platforms
• LidarNAS: Unifying and Searching Neural Architectures for 3D Point Clouds
• U-Boost NAS: Utilization-Boosted Differentiable Neural Architecture Search
  :star:code
• A Max-Flow Based Approach for Neural Architecture Search
• ViTAS: Vision Transformer Architecture Search
• Learning Where to Look – Generative NAS Is Surprisingly Efficient
  :star:code
• Neural Architecture Search for Spiking Neural Networks
• Data-Free Neural Architecture Search via Recursive Label Calibration
  :star:code

27.Image Classification(图像分类)

• Exploring Fine-Grained Audiovisual Categorization with the SSW60 Dataset
  :star:code
• Centrality and Consistency: Two-Stage Clean Samples Identification for Learning with Instance-Dependent Noisy Labels
  :star:code
• Not All Models Are Equal: Predicting Model Transferability in a Self-Challenging Fisher Space
  :star:code
• Constructing Balance from Imbalance for Long-tailed Image Recognition
  :star:code
• No Token Left Behind: Explainability-Aided Image Classification and Generation
  :star:code
• Interpretable Image Classification with Differentiable Prototypes Assignment
  :star:code
• Rotation Regularization without Rotation
  :star:code
• Revisiting a kNN-based Image Classification System with High-capacity Storage
• In Defense of Image Pre-training for Spatiotemporal Recognition
  :star:code
• Augmenting Deep Classifiers with Polynomial Neural Networks
• A Dataset Generation Framework for Evaluating Megapixel Image Classifiers & their Explanations
• Cartoon Explanations of Image Classifiers
• Exploring Hierarchical Graph Representation for Large-Scale Zero-Shot Image Classification
  :house:project
• SSBNet: Improving Visual Recognition Efficiency by Adaptive Sampling
• AutoMix: Unveiling the Power of Mixup for Stronger Classifiers
• MaxViT: Multi-axis Vision Transformer
  :star:code
• Self-Feature Distillation with Uncertainty Modeling for Degraded Image Recognition
• Three Things Everyone Should Know about Vision Transformers
• RealPatch: A Statistical Matching Framework for Model Patching with Real Samples
  :star:code
• TDAM: Top-Down Attention Module for Contextually Guided Feature Selection in CNNs
  :star:code
• Automatic Check-Out via Prototype-Based Classifier Learning from Single-Product Exemplars
  :star:code
• Embedding Contrastive Unsupervised Features to Cluster in- and Out-of-Distribution Noise in Corrupted Image Datasets
• Unsupervised Few-Shot Image Classification by Learning Features into Clustering Space
• 小样本图像分类
  • Tree Structure-Aware Few-Shot Image Classification via Hierarchical Aggregation
    :star:code
  • Tip-Adapter: Training-free Adaption of CLIP for Few-shot Classification
    :star:code
  • Adversarial Feature Augmentation for Cross-domain Few-shot Classification
    :star:code
  • Few-Shot Classification with Contrastive Learning
• 多标签分类
  • Hyperspherical Learning in Multi-Label Classification
    :star:code
• 长尾分类
  • SAFA: Sample-Adaptive Feature Augmentation for Long-Tailed Image Classification
  • Invariant Feature Learning for Generalized Long-Tailed Classification
    :star:code
  • Tackling Long-Tailed Category Distribution Under Domain Shifts
    :star:code:house:project
  • Identifying Hard Noise in Long-Tailed Sample Distribution
    :open_mouth:oral:star:code
  • On Multi-Domain Long-Tailed Recognition, Imbalanced Domain Generalization and Beyond
    :star:code
• 视觉分类
  • Visual Knowledge Tracing
    :star:code
• 细粒度识别
  • Improving Fine-Grained Visual Recognition in Low Data Regimes via Self-Boosting Attention Mechanism
    :star:code
  • Zero-Shot Attribute Attacks on Fine-Grained Recognition Models
  • Where to Focus: Investigating Hierarchical Attention Relationship for Fine-Grained Visual Classification
    :star:code
• 长尾识别
  • Towards Calibrated Hyper-Sphere Representation via Distribution Overlap Coefficient for Long-tailed Learning
    :star:code
  • Breadcrumbs: Adversarial Class-Balanced Sampling for Long-Tailed Recognition
    :star:code
  • VL-LTR: Learning Class-Wise Visual-Linguistic Representation for Long-Tailed Visual Recognition
    :star:code

26.Video/Image Super-Resolution(视频/图像超分辨率)

• 跨模态超分辨率
  • Learning Mutual Modulation for Self-Supervised Cross-Modal Super-Resolution
    :star:code
• 图像超分辨率
  • Image Super-Resolution with Deep Dictionary
    :star:code
  • CADyQ: Content-Aware Dynamic Quantization for Image Super-Resolution
    :star:code
  • Reference-based Image Super-Resolution with Deformable Attention Transformer
    :star:code
  • KXNet: A Model-Driven Deep Neural Network for Blind Super-Resolution
    :star:code
  • Super-Resolution by Predicting Offsets: An Ultra-Efficient Super-Resolution Network for Rasterized Images
  • Boosting Event Stream Super-Resolution with a Recurrent Neural Network
  • Learning Series-Parallel Lookup Tables for Efficient Image Super-Resolution
    :star:code
  • Efficient Long-Range Attention Network for Image Super-Resolution
    :star:code
  • Metric Learning Based Interactive Modulation for Real-World Super-Resolution
    :star:code
  • Dynamic Dual Trainable Bounds for Ultra-Low Precision Super-Resolution Networks
    :star:code
  • Perception-Distortion Balanced ADMM Optimization for Single-Image Super-Resolution
    :star:code
  • Uncertainty Learning in Kernel Estimation for Multi-stage Blind Image Super-Resolution
  • MuLUT: Cooperating Multiple Look-Up Tables for Efficient Image Super-Resolution
  • Adaptive Patch Exiting for Scalable Single Image Super-Resolution
    :star:code
  • From Face to Natural Image: Learning Real Degradation for Blind Image Super-Resolution
    :star:code
  • Unfolded Deep Kernel Estimation for Blind Image Super-Resolution
    :star:code
  • Efficient and Degradation-Adaptive Network for Real-World Image Super-Resolution
    :star:code
  • Self-Supervised Learning for Real-World Super-Resolution from Dual Zoomed Observations
    :star:code
  • Restore Globally, Refine Locally: A Mask-Guided Scheme to Accelerate Super-Resolution Networks
    :star:code
  • Compiler-Aware Neural Architecture Search for On-Mobile Real-Time Super-Resolution
    :star:code
  • KXNet: A Model-Driven Deep Neural Network for Blind Super-Resolution
    :star:code
  • ARM: Any-Time Super-Resolution Method
    :star:code
  • D2C-SR: A Divergence to Convergence Approach for Real-World Image Super-Resolution
    :star:code
  • RRSR:Reciprocal Reference-Based Image Super-Resolution with Progressive Feature Alignment and Selection
• 视频超分辨率
  • Towards Interpretable Video Super-Resolution via Alternating Optimization
    :star:code
  • Learning Spatiotemporal Frequency-Transformer for Compressed Video Super-Resolution
    :star:code
  • Real-RawVSR: Real-World Raw Video Super-Resolution with a Benchmark Dataset
    :star:code
  • A Codec Information Assisted Framework for Efficient Compressed Video Super-Resolution

25.Autonomous vehicles(自动驾驶)

• 车辆轨迹预测
  • Hierarchical Latent Structure for Multi-Modal Vehicle Trajectory Forecasting
    :star:code
  • Action-based Contrastive Learning for Trajectory Prediction
  • D2-TPred: Discontinuous Dependency for Trajectory Prediction under Traffic Lights
    :star:code
  • AdvDO: Realistic Adversarial Attacks for Trajectory Prediction
    :house:project
• 自动驾驶
  • ST-P3: End-to-end Vision-based Autonomous Driving via Spatial-Temporal Feature Learning
    :star:code
  • Resolving Copycat Problems in Visual Imitation Learning via Residual Action Prediction
  • Dfferentiable Raycasting for Self-supervised Occupancy Forecasting
    :star:code
  • Self-Distillation for Robust LiDAR Semantic Segmentation in Autonomous Driving
    :star:code
  • V2X-ViT: Vehicle-to-Everything Cooperative Perception with Vision Transformer
    :star:code
  • Radatron: Accurate Detection Using Multi-Resolution Cascaded MIMO Radar
    :house:project
  • Rethinking Closed-Loop Training for Autonomous Driving
  • Motion Inspired Unsupervised Perception and Prediction in Autonomous Driving
  • KING: Generating Safety-Critical Driving Scenarios for Robust Imitation via Kinematics Gradients
    :open_mouth:oral:house:project
  • InAction: Interpretable Action Decision Making for Autonomous Driving
    :star:code
  • CODA: A Real-World Road Corner Case Dataset for Object Detection in Autonomous Driving
    :house:project
  • Unsupervised Semantic Segmentation of Urban Scenes via Cross-Modal Distillation
    :open_mouth:oral:house:project
  • StretchBEV: Stretching Future Instance Prediction Spatially and Temporally
    :house:project
  • BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers
    :star:code
  • Point Cloud Compression with Range Image-Based Entropy Model for Autonomous Driving
• 轨迹预测
  • Learning Pedestrian Group Representations for Multi-modal Trajectory Prediction
    :star:code
  • Aware of the History: Trajectory Forecasting with the Local Behavior Data
    :star:code
  • Social-SSL: Self-Supervised Cross-Sequence Representation Learning Based on Transformers for Multi-agent Trajectory Prediction
    :star:code
  • Sequential Multi-View Fusion Network for Fast LiDAR Point Motion Estimation
  • Social-Implicit: Rethinking Trajectory Prediction Evaluation and the Effectiveness of Implicit Maximum Likelihood Estimation
    :star:code
  • View Vertically: A Hierarchical Network for Trajectory Prediction via Fourier Spectrums
    :star:code
  • PreTraM: Self-Supervised Pre-training via Connecting Trajectory and Map
    :star:code
• 车道线检测
  • RCLane: Relay Chain Prediction for Lane Detection
  • PersFormer: 3D Lane Detection via Perspective Transformer and the OpenLane Benchmark
    :open_mouth:oral:star:code
  • Lane Detection Transformer Based on Multi-Frame Horizontal and Vertical Attention and Visual Transformer Module
• 行人轨迹预测
  • Human Trajectory Prediction via Neural Social Physics
    :star:code
  • SocialVAE: Human Trajectory Prediction Using Timewise Latents
    :star:code:house:project
• 车辆重识别
  • Unstructured Feature Decoupling for Vehicle Re-identification
    :star:code

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

• 遥感
  • Tomography of Turbulence Strength Based on Scintillation Imaging
  • TD-Road: Top-Down Road Network Extraction with Holistic Graph Construction
• 航空视频识别
• FAR: Fourier Aerial Video Recognition
  :house:project

23.Medical Image(医学影像)

• The Surprisingly Straightforward Scene Text Removal Method With Gated Attention and Region of Interest Generation: A Comprehensive Prominent Model Analysis
  :star:code:house:project
• 医学图像分割
  • Personalizing Federated Medical Image Segmentation via Local Calibration
    :star:code
  • Learning Topological Interactions for Multi-Class Medical Image Segmentation
    :open_mouth:oral:star:code
  • Generalizable Medical Image Segmentation via Random Amplitude Mixup and Domain-Specific Image Restoration
    :star:code
  • PointScatter: Point Set Representation for Tubular Structure Extraction
    :open_mouth:oral:star:code
  • Dual Contrastive Learning with Anatomical Auxiliary Supervision for Few-Shot Medical Image Segmentation
    :star:code
  • Auto-FedRL: Federated Hyperparameter Optimization for Multi-Institutional Medical Image Segmentation
    :star:code
  • Med-DANet: Dynamic Architecture Network for Efficient Medical Volumetric Segmentation
    :star:code
  • CXR Segmentation by AdaIN-Based Domain Adaptation and Knowledge Distillation
• 放射科报告生成
  • Cross-modal Prototype Driven Network for Radiology Report Generation
    :star:code
• 密集预测
  • ConCL: Concept Contrastive Learning for Dense Prediction Pre-training in Pathology Images
    :star:code
• retinal image matching(视网膜图像匹配)
  • Semi-Supervised Keypoint Detector and Descriptor for Retinal Image Matching
    :star:code
• 支架追踪
  • Robust Landmark-based Stent Tracking in X-ray Fluoroscopy
• 病变检测
  • Check and Link: Pairwise Lesion Correspondence Guides Mammogram Mass Detection
• 医学图像分析
  • UniMiSS: Universal Medical Self-Supervised Learning via Breaking Dimensionality Barrier
    :star:code
  • K-SALSA: K-Anonymous Synthetic Averaging of Retinal Images via Local Style Alignment
    :star:code
• 医学图像分类
  • Differentiable Zooming for Multiple Instance Learning on Whole-Slide Images
    :star:code
  • 疾病分类
    • RadioTransformer: A Cascaded Global-Focal Transformer for Visual Attention-Guided Disease Classification
      :star:code
• 医学关键点定位
  • One-Shot Medical Landmark Localization by Edge-Guided Transform and Noisy Landmark Refinement
    :star:code

22.OCR

• Levenshtein OCR
• 文本识别
  • Text-DIAE: Degradation Invariant Autoencoders for Text Recognition and Document Enhancement
• 手写数学表达式识别
  • CoMER: Modeling Coverage for Transformer-based Handwritten Mathematical Expression Recognition
    :star:code
  • When Counting Meets HMER: Counting-Aware Network for Handwritten Mathematical Expression Recognition
    :star:code
• 场景文本检测
  • Scene Text Recognition with Permuted Autoregressive Sequence Models
    :star:code
  • Dynamic Low-Resolution Distillation for Cost-Efficient End-to-End Text Spotting
    :star:code
  • SGBANet: Semantic GAN and Balanced Attention Network for Arbitrarily Oriented Scene Text Recognition
  • Optimal Boxes: Boosting End-to-End Scene Text Recognition by Adjusting Annotated Bounding Boxes via Reinforcement Learning
  • Contextual Text Block Detection towards Scene Text Understanding
    :house:project
  • Toward Understanding WordArt: Corner-Guided Transformer for Scene Text Recognition
    :open_mouth:oral:star:code
  • GLASS: Global to Local Attention for Scene-Text Spotting
    :star:code
  • Multi-Granularity Prediction for Scene Text Recognition
  • Pure Transformer with Integrated Experts for Scene Text Recognition
  • Background-Insensitive Scene Text Recognition with Text Semantic Segmentation
  • Detecting Tampered Scene Text in the Wild
    :star:code
  • Language Matters: A Weakly Supervised Vision-Language Pre-training Approach for Scene Text Detection and Spotting
  • TextAdaIN: Paying Attention to Shortcut Learning in Text Recognizers
  • Multi-modal Text Recognition Networks: Interactive Enhancements between Visual and Semantic Features
    :star:code
  • OCR-Free Document Understanding Transformer
    :star:code
• 视频文本检测
  • Real-time End-to-End Video Text Spotter with Contrastive Representation Learning
    :star:code
• 文本检测
  • Unitail: Detecting, Reading, and Matching in Retail Scene"
    :house:project
• 文件图像矫正
  • Geometric Representation Learning for Document Image Rectification
    :star:code
• document unwarping
  • Learning an Isometric Surface Parameterization for Texture Unwrapping
    :star:code

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

• 无监督
  • Contrastive Positive Mining for Unsupervised 3D Action Representation Learning
  • Dense Siamese Network for Dense Unsupervised Learning
    :star:code
  • Contrastive Positive Mining for Unsupervised 3D Action Representation Learning
  • Relative Contrastive Loss for Unsupervised Representation Learning
  • DiffuseMorph: Unsupervised Deformable Image Registration Using Diffusion Model
• 弱监督
  • Acknowledging the Unknown for Multi-Label Learning with Single Positive Labels
    :star:code
• 自监督
  • GOCA: Guided Online Cluster Assignment for Self-Supervised Video Representation Learning
    :star:code
  • RegionCL: Exploring Contrastive Region Pairs for Self-Supervised Representation Learning
    :star:code
  • Domain Knowledge-Informed Self-Supervised Representations for Workout Form Assessment
  • Differentiable Raycasting for Self-Supervised Occupancy Forecasting
  • How Severe Is Benchmark-Sensitivity in Video Self-Supervised Learning?
    :star:code
  • MaCLR: Motion-Aware Contrastive Learning of Representations for Videos
  • Making the Most of Text Semantics to Improve Biomedical Vision-Language Processing
  • Mc-BEiT: Multi-Choice Discretization for Image BERT Pre-training
    :star:code
  • What to Hide from Your Students: Attention-Guided Masked Image Modeling
    :star:code
  • Constrained Mean Shift Using Distant Yet Related Neighbors for Representation Learning
    :star:code
  • Semantic-Aware Fine-Grained Correspondence
  • Self-Supervised Classification Network
    :star:code
  • Dual-Domain Self-Supervised Learning and Model Adaption for Deep Compressive Imaging
  • SdAE: Self-distillated Masked Autoencoder
    :star:code
  • RDA: Reciprocal Distribution Alignment for Robust SSL
    :star:code
  • Motion Sensitive Contrastive Learning for Self-supervised Video Representation
  • Towards Efficient and Effective Self-Supervised Learning of Visual Representations
    :star:code
  • Unifying Visual Contrastive Learning for Object Recognition from a Graph Perspective
  • The Challenges of Continuous Self-Supervised Learning
  • GeoRefine: Self-Supervised Online Depth Refinement for Accurate Dense Mapping
  • Fusion from Decomposition: A Self-Supervised Decomposition Approach for Image Fusion
  • DNA: Improving Few-Shot Transfer Learning with Low-Rank Decomposition and Alignment
    :star:code
  • Self-Supervised Learning of Visual Graph Matching
    :star:code
  • DisCo: Remedying Self-Supervised Learning on Lightweight Models with Distilled Contrastive Learning
    :star:code
  • SLIP: Self-Supervision Meets Language-Image Pre-training
    :star:code
  • Domain Invariant Masked Autoencoders for Self-Supervised Learning from Multi-Domains
  • Improving Self-Supervised Lightweight Model Learning via Hard-Aware Metric Distillation
    :star:code
  • Masked Siamese Networks for Label-Efficient Learning
    :star:code
  • Natural Synthetic Anomalies for Self-Supervised Anomaly Detection and Localization
    :star:code
  • Understanding Collapse in Non-Contrastive Siamese Representation Learning
  • Discovering Deformable Keypoint Pyramids
    :star:code
• 半监督
  • Towards Realistic Semi-Supervised Learning
  • OpenLDN: Learning to Discover Novel Classes for Open-World Semi-Supervised Learning
  • Semi-Leak: Membership Inference Attacks Against Semi-supervised Learning
    :star:code
  • ConMatch: Semi-Supervised Learning with Confidence-Guided Consistency Regularization
    :star:code
  • Vibration-Based Uncertainty Estimation for Learning from Limited Supervision
  • Unsupervised Selective Labeling for More Effective Semi-Supervised Learning
  • RDA: Reciprocal Distribution Alignment for Robust Semi-Supervised Learning
    :star:code
  • Semi-Supervised Vision Transformers
    :star:code
  • CA-SSL: Class-Agnostic Semi-Supervised Learning for Detection and Segmentation
  • RVSL: Robust Vehicle Similarity Learning in Real Hazy Scenes Based on Semi-supervised Learning
    :star:code
  • Semi-Supervised Keypoint Detector and Descriptor for Retinal Image Matching
  • PSS: Progressive Sample Selection for Open-World Visual Representation Learning
    :star:code
  • Stochastic Consensus: Enhancing Semi-Supervised Learning with Consistency of Stochastic Classifiers
• 监督学习
  • Supervised Attribute Information Removal and Reconstruction for Image Manipulation
    :star:code
  • Tailoring Self-Supervision for Supervised Learning
    :star:code

20.Face(人脸)

• Effective Presentation Attack Detection Driven by Face Related Task
  :star:code
• Facial Depth and Normal Estimation Using Single Dual-Pixel Camera
  :star:code
• StyleFace: Towards Identity-Disentangled Face Generation on Megapixels
• Augmentation of rPPG Benchmark Datasets: Learning to Remove and Embed rPPG Signals via Double Cycle Consistent Learning from Unpaired Facial Videos
  :star:code
• Custom Structure Preservation in Face Aging
• deepfake检测
  • Detecting and Recovering Sequential DeepFake Manipulation
    :star:code:house:project
  • Explaining Deepfake Detection by Analysing Image Matching
    :star:code
  • Hierarchical Contrastive Inconsistency Learning for Deepfake Video Detection
• 三维人脸
  • Structure-aware Editable Morphable Model for 3D Facial Detail Animation and Manipulation
    :star:code
• 活体检测
  • Generative Domain Adaptation for Face Anti-Spoofing
  • Multi-domain Learning for Updating Face Anti-spoofing Models
    :star:code
  • Source-Free Domain Adaptation with Contrastive Domain Alignment and Self-Supervised Exploration for Face Anti-Spoofing
    :star:code
  • Adaptive Transformers for Robust Few-Shot Cross-Domain Face Anti-Spoofing
• 人脸识别
  • Controllable and Guided Face Synthesis for Unconstrained Face Recognition
    :star:code:house:project
  • Towards Robust Face Recognition with Comprehensive Search
  • BoundaryFace: A mining framework with noise label self-correction for Face Recognition
    :star:code
  • Privacy-Preserving Face Recognition with Learnable Privacy Budgets in Frequency Domain
    :star:code
  • OneFace: One Threshold for All
  • AgeTransGAN for Facial Age Transformation with Rectified Performance Metrics
    :star:code
  • Teaching Where to Look: Attention Similarity Knowledge Distillation for Low Resolution Face Recognition
    :star:code
  • CoupleFace: Relation Matters for Face Recognition Distillation
  • Towards Racially Unbiased Skin Tone Estimation via Scene Disambiguation
    :house:project
  • Pre-training Strategies and Datasets for Facial Representation Learning
  • Unsupervised and Semi-Supervised Bias Benchmarking in Face Recognition
• 人脸聚类
  • On Mitigating Hard Clusters for Face Clustering
    :open_mouth:oral:star:code
• 说话人脸合成
  • StyleHEAT: One-Shot High-Resolution Editable Talking Face Generation via Pre-trained StyleGAN
    :star:code
• 谈话头像合成
  • Learning Dynamic Facial Radiance Fields for Few-Shot Talking Head Synthesis
    :star:code:house:project
• 人脸姿势估计
  • Towards Unbiased Label Distribution Learning for Facial Pose Estimation Using Anisotropic Spherical Gaussian
• 人脸交换
  • StyleSwap: Style-Based Generator Empowers Robust Face Swapping
    :star:code:house:project
  • Designing One Unified Framework for High-Fidelity Face Reenactment and Swapping
    :star:code
• 假脸检测
  • UIA-ViT: Unsupervised Inconsistency-Aware Method based on Vision Transformer for Face Forgery Detection
    :open_mouth:oral
  • An Information Theoretic Approach for Attention-Driven Face Forgery Detection
  • Exploring Disentangled Content Information for Face Forgery Detection
  • Adaptive Face Forgery Detection in Cross Domain
• 人脸捕捉
  • Practical and Scalable Desktop-Based High-Quality Facial Capture
• 人脸表情识别
  • How to Synthesize a Large-Scale and Trainable Micro-Expression Dataset?
    :star:code
  • Teaching with Soft Label Smoothing for Mitigating Noisy Labels in Facial Expressions
    :star:code
  • Order Learning Using Partially Ordered Data via Chainization
    :star:code
  • Emotion-Aware Multi-View Contrastive Learning for Facial Emotion Recognition
    :star:code
  • Learn-to-Decompose: Cascaded Decomposition Network for Cross-Domain Few-Shot Facial Expression Recognition
    :star:code
  • Learn from All: Erasing Attention Consistency for Noisy Label Facial Expression Recognition
    :star:code
• 三维人脸重建
  • REALY: Rethinking the Evaluation of 3D Face Reconstruction
    :star:code:house:project
  • AU-Aware 3D Face Reconstruction through Personalized AU-Specific Blendshape Learning
  • 3D Face Reconstruction with Dense Landmarks
  • Towards Metrical Reconstruction of Human Faces
    :house:project
• 人脸重现
  • Face2Faceρ: Real-Time High-Resolution One-Shot Face Reenactment
• 人脸身份操作
  • MFIM: Megapixel Facial Identity Manipulation
• 人脸纹理合成与重建
  • Unsupervised High-Fidelity Facial Texture Generation and Reconstruction
    :star:code
• 人脸恢复
  • VQFR: Blind Face Restoration with Vector-Quantized Dictionary and Parallel Decoder
    :star:code
• 表情识别
  • Emotion Recognition for Multiple Context Awareness
    :house:project
  • S2-VER: Semi-Supervised Visual Emotion Recognition
    :star:code

19.Image Synthesis/Generation(图像合成)

• Injecting 3D Perception of Controllable NeRF-GAN into StyleGAN for Editable Portrait Image Synthesis
  :star:code:house:project
• GALA: Toward Geometry-and-Lighting-Aware Object Search for Compositing
• Generalized Brain Image Synthesis with Transferable Convolutional Sparse Coding Networks
• Auto-regressive Image Synthesis with Integrated Quantization
  :open_mouth:oral
• Paint2Pix: Interactive Painting based Progressive Image Synthesis and Editing
  :star:code
• Improved Masked Image Generation with Token-Critic
• Weakly-Supervised Stitching Network for Real-World Panoramic Image Generation
• SCAM! Transferring humans between images with Semantic Cross Attention Modulation
  :house:project
• PixelFolder: An Efficient Progressive Pixel Synthesis Network for Image Generation
  :star:code
• Adaptive Feature Interpolation for Low-Shot Image Generation
• Few-Shot Image Generation with Mixup-Based Distance Learning
  :star:code
• Multimodal Conditional Image Synthesis with Product-of-Experts GANs
  :house:project
• Any-Resolution Training for High-Resolution Image Synthesis
  :house:project
• 3D-Aware Indoor Scene Synthesis with Depth Priors
  :house:project
• 图像生成
  • DeltaGAN: Towards Diverse Few-shot Image Generation with Sample-Specific Delta
    :star:code
  • Scraping Textures from Natural Images for Synthesis and Editing
    :house:project
  • Word-Level Fine-Grained Story Visualization
  • CoGS: Controllable Generation and Search from Sketch and Style
  • Unsupervised Learning of Efficient Geometry-Aware Neural Articulated Representations
  • Unleashing Transformers: Parallel Token Prediction with Discrete Absorbing Diffusion for Fast High-Resolution Image Generation from Vector-Quantized Codes
    :star:code
• 样本引导下的图像生成
  • DynaST: Dynamic Sparse Transformer for Exemplar-Guided Image Generation
    :star:code
• 文本-图像合成
  • StoryDALL-E: Adapting Pretrained Text-to-Image Transformers for Story Continuation
    :star:code
  • TISE: Bag of Metrics for Text-to-Image Synthesis Evaluation
  • Make-a-Scene: Scene-Based Text-to-Image Generation with Human Priors
    :tv:video
  • Trace Controlled Text to Image Generation
• 从文本描述中生成不同的人类动作
  • TEMOS: Generating Diverse Human Motions from Textual Descriptions
    :open_mouth:oral:star:code:house:project

18.Image-to-Image Translation(图像到图像翻译)

• VecGAN: Image-to-Image Translation with Interpretable Latent Directions
• Vector Quantized Image-to-Image Translation
  :star:code:house:project
• Ultra-high-resolution unpaired stain transformation via Kernelized Instance Normalization
  :star:code
• Unpaired Image Translation via Vector Symbolic Architectures
  :open_mouth:oral:star:code
• Bi-Level Feature Alignment for Versatile Image Translation and Manipulation
• ManiFest: Manifold Deformation for Few-Shot Image Translation
  :star:code
• 图像翻译
  • BayesCap: Bayesian Identity Cap for Calibrated Uncertainty in Frozen Neural Networks
    :star:code
  • Multi-Curve Translator for High-Resolution Photorealistic Image Translation

17.GAN

• VQGAN-CLIP: Open Domain Image Generation and Editing with Natural Language Guidance
• Quantized GAN for Complex Music Generation from Dance Videos
  :star:code
• RepMix: Representation Mixing for Robust Attribution of Synthesized Images
  :star:code
• FakeCLR: Exploring Contrastive Learning for Solving Latent Discontinuity in Data-Efficient GANs
  :star:code
• Generative Multiplane Images: Making a 2D GAN 3D-Aware
  :star:code:house:project
• Generator Knows What Discriminator Should Learn in Unconditional GANs
  :star:code
• Hierarchical Semantic Regularization of Latent Spaces in StyleGANs
  :star:code:house:project
• Mind the Gap in Distilling StyleGANs
  :star:code
• FurryGAN: High Quality Foreground-aware Image Synthesis
  :house:project
• Improving GANs for Long-Tailed Data through Group Spectral Regularization
  :star:code:house:project
• 3D-FM GAN: Towards 3D-Controllable Face Manipulation
  :house:project
• Exploring Gradient-based Multi-directional Controls in GANs
  :star:code
• Studying Bias in GANs through the Lens of Race
• FairStyle: Debiasing StyleGAN2 with Style Channel Manipulations
  :house:project
• FingerprintNet: Synthesized Fingerprints for Generated Image Detection
• Detecting Generated Images by Real Images
  :star:code
• High-Fidelity GAN Inversion with Padding Space
  :house:project
• A Style-Based GAN Encoder for High Fidelity Reconstruction of Images and Videos
  :star:code
• BlobGAN: Spatially Disentangled Scene Representations
  :house:project
• GAN with Multivariate Disentangling for Controllable Hair Editing
  :star:code
• StyleGAN-Human: A Data-Centric Odyssey of Human Generation
  :star:code
• EAGAN: Efficient Two-Stage Evolutionary Architecture Search for GANs
  :star:code
• JoJoGAN: One Shot Face Stylization
• HairNet: Hairstyle Transfer with Pose Changes
• EleGANt: Exquisite and Locally Editable GAN for Makeup Transfer
  :star:code
• Editing Out-of-Domain GAN Inversion via Differential Activations
  :star:code
• On the Robustness of Quality Measures for GANs
  :star:code
• Diverse Generation from a Single Video Made Possible
  :house:project
• Rayleigh EigenDirections (REDs): Nonlinear GAN Latent Space Traversals for Multidimensional Features
• Generating Natural Images with Direct Patch Distributions Matching
  :star:code
• TREND: Truncated Generalized Normal Density Estimation of Inception Embeddings for GAN Evaluation
• Neural Scene Decoration from a Single Photograph
  :star:code
• ChunkyGAN: Real Image Inversion via Segments
• GAN Cocktail: Mixing GANs without Dataset Access
  :house:project
• DuelGAN: A Duel between Two Discriminators Stabilizes the GAN Training
  :star:code
• 线稿上色
  • Eliminating Gradient Conflict in Reference-based Line-Art Colorization
    :star:code
• 图像生成
  • WaveGAN: Frequency-aware GAN for High-Fidelity Few-shot Image Generation
    :star:code
• GAN逆映射
  • IntereStyle: Encoding an Interest Region for Robust StyleGAN Inversion
• 妆发迁移
  • RamGAN: Region Attentive Morphing GAN for Region-Level Makeup Transfer
• 文本消除
  • Don't Forget Me: Accurate Background Recovery for Text Removal via Modeling Local-Global Context
    :star:code

16.Transformer

• k-means Mask Transformer
  :star:code
• Outpainting by Queries
  :star:code
• Laplacian Mesh Transformer: Dual Attention and Topology Aware Network for 3D Mesh Classification and Segmentation
• Locality Guidance for Improving Vision Transformers on Tiny Datasets
  :star:code
• ParC-Net: Position Aware Circular Convolution with Merits from ConvNets and Transformer
  :star:code
• MTFormer: Multi-task Learning via Transformer and Cross-Task Reasoning
• TinyViT: Fast Pretraining Distillation for Small Vision Transformers
  :star:code
• MeshMAE: Masked Autoencoders for 3D Mesh Data Analysis
• An Impartial Take to the CNN vs Transformer Robustness Contest
• Ghost-free High Dynamic Range Imaging with Context-aware Transformer
  :star:code
• EdgeViTs: Competing Light-Weight CNNs on Mobile Devices with Vision Transformers
  :star:code
• Adaptive Token Sampling for Efficient Vision Transformers
  :open_mouth:oral:house:project
• Self-Slimmed Vision Transformer
  :star:code
• Are Vision Transformers Robust to Patch Perturbations?
• Selective TransHDR: Transformer-Based Selective HDR Imaging Using Ghost Region Mask
• BLT: Bidirectional Layout Transformer for Controllable Layout Generation
  :house:project
• Convolutional Embedding Makes Hierarchical Vision Transformer Stronger
• AMixer: Adaptive Weight Mixing for Self-Attention Free Vision Transformers
  :star:code
• Doubly-Fused ViT: Fuse Information from Vision Transformer Doubly with Local Representation
• VIP: Unified Certified Detection and Recovery for Patch Attack with Vision Transformers
• Improving Vision Transformers by Revisiting High-Frequency Components
  :star:code
• VSA: Learning Varied-Size Window Attention in Vision Transformers
  :star:code
• DaViT: Dual Attention Vision Transformers
  :star:code
• KVT: k-NN Attention for Boosting Vision Transformers
  :star:code
• ScalableViT: Rethinking the Context-Oriented Generalization of Vision Transformer
  :star:code
• DeiT III: Revenge of the ViT
• Sliced Recursive Transformer
  :star:code
• Improving Closed and Open-Vocabulary Attribute Prediction Using Transformers
  :house:project
• Training Vision Transformers with Only 2040 Images

15.Vision-Language(视觉语言)

• FashionViL: Fashion-Focused Vision-and-Language Representation Learning
  :star:code
• NewsStories: Illustrating articles with visual summaries
  :star:code
• Can Shuffling Video Benefit Temporal Bias Problem: A Novel Training Framework for Temporal Grounding
  :star:code
• Frozen CLIP Models are Efficient Video Learners
  :star:code
• Generative Negative Text Replay for Continual Vision-Language Pretraining
• This Is My Unicorn, Fluffy”: Personalizing Frozen Vision-Language Representations
  :star:code
• Contrastive Vision-Language Pre-training with Limited Resources
  :star:code
• ASSISTER: Assistive Navigation via Conditional Instruction Generation
• X-DETR: A Versatile Architecture for Instance-Wise Vision-Language Tasks
  :star:code
• UniTAB: Unifying Text and Box Outputs for Grounded Vision-Language Modeling
• Single-Stream Multi-level Alignment for Vision-Language Pretraining
• Most and Least Retrievable Images in Visual-Language Query Systems
• 视觉表征学习
  • Wave-ViT: Unifying Wavelet and Transformers for Visual Representation Learning
    :star:code
  • Unsupervised Visual Representation Learning by Synchronous Momentum Grouping
  • Learning Visual Representation from Modality-Shared Contrastive Language-Image Pre-training
    :star:code
• VLN
  • Learning from Unlabeled 3D Environments for Vision-and-Language Navigation
    :house:project
  • Sim-2-Sim Transfer for Vision-and-Language Navigation in Continuous Environments
    :house:project
  • Bridging the visual gap in VLN via semantically richer instructions
  • A Dataset for Interactive Vision-Language Navigation with Unknown Command Feasibility
  • Learning Disentanglement with Decoupled Labels for Vision-Language Navigation
    :star:code
  • Multimodal Transformer with Variable-Length Memory for Vision-and-Language Navigation
    :star:code
  • FedVLN: Privacy-Preserving Federated Vision-and-Language Navigation
    :star:code
  • Bridging the Visual Semantic Gap in VLN via Semantically Richer Instructions
• 视觉重定位
  • Map-free Visual Relocalization: Metric Pose Relative to a Single Image

14.Visual Answer Questions(视觉问答)

• Weakly Supervised Grounding for VQA in Vision-Language Transformers
  :star:code
• Rethinking Data Augmentation for Robust Visual Question Answering
  :star:code
• Reliable Visual Question Answering: Abstain Rather Than Answer Incorrectly
  :star:code
• New Datasets and Models for Contextual Reasoning in Visual Dialog
  :star:code
• Classification-Regression for Chart Comprehension
• AssistQ: Affordance-Centric Question-Driven Task Completion for Egocentric Assistant
  :house:project
• Video-QA
  • Video Graph Transformer for Video Question Answering
    :star:code
  • Video Question Answering with Iterative Video-Text Co-Tokenization
    :star:code

13.Human-Object Interaction(人物交互)

• Towards Hard-Positive Query Mining for DETR-based Human-Object Interaction Detection
  :star:code
• Geometric Features Informed Multi-person Human-object Interaction Recognition in Videos
  :star:code
• IGFormer: Interaction Graph Transformer for Skeleton-based Human Interaction Recognition
• Mining Cross-Person Cues for Body-Part Interactiveness Learning in HOI Detection
  :star:code
• Iwin: Human-Object Interaction Detection via Transformer with Irregular Windows
• SAGA: Stochastic Whole-Body Grasping with Contact
  :house:project
• Chairs Can Be Stood On: Overcoming Object Bias in Human-Object Interaction Detection
• Discovering Human-Object Interaction Concepts via Self-Compositional Learning
  :star:code
• 交互式物体分割
  • Self-Supervised Interactive Object Segmentation Through a Singulation-and-Grasping Approach
    :house:project
• HOS
  • Fine-Grained Egocentric Hand-Object Segmentation: Dataset, Model, and Applications
    :star:code
• 手物交互
  • TOCH: Spatio-Temporal Object-to-Hand Correspondence for Motion Refinement
  • 抓握合成(手物交互)
    • Grasp'D: Differentiable Contact-Rich Grasp Synthesis for Multi-Fingered Hands
      :house:project
• 人椅互动
  • COUCH: Towards Controllable Human-Chair Interactions

12.Action Detection(人体动作检测与识别)

• 动作识别
  • PrivHAR: Recognizing Human Actions from Privacy-Preserving Lens
    :house:project
  • Source-Free Video Domain Adaptation by Learning Temporal Consistency for Action Recognition
    :star:code
  • Efficient Video Transformers with Spatial-Temporal Token Selection
    :star:code
  • Learn2Augment: Learning to Composite Videos for Data Augmentation in Action Recognition
    :house:project
  • Hierarchically Self-Supervised Transformer for Human Skeleton Representation Learning
    :star:code
  • An Efficient Spatio-Temporal Pyramid Transformer for Action Detection
  • Spatiotemporal Self-attention Modeling with Temporal Patch Shift for Action Recognition
    :star:code
  • Privacy-Preserving Action Recognition via Motion Difference Quantization
    :star:code
  • SOS! Self-Supervised Learning over Sets of Handled Objects in Egocentric Action Recognition
  • Real-time Online Video Detection with Temporal Smoothing Transformers
    :star:code
  • CycDA: Unsupervised Cycle Domain Adaptation to Learn from Image to Video
  • Uncertainty-Based Spatial-Temporal Attention for Online Action Detection
  • Is Appearance Free Action Recognition Possible?
  • Panoramic Human Activity Recognition
  • Delving into Details: Synopsis-to-Detail Networks for Video Recognition
    :star:code
  • 细粒度动作识别
    • Dynamic Spatio-Temporal Specialization Learning for Fine-Grained Action Recognition
    • Dynamic Spatio-Temporal Specialization Learning for Fine-Grained Action Recognition
  • 零样本动作识别
    • CLASTER: Clustering with Reinforcement Learning for Zero-Shot Action Recognition
      :house:project
    • Rethinking Zero-Shot Action Recognition: Learning from Latent Atomic Actions
  • 小样本动作识别
    • Few-Shot Action Recognition with Hierarchical Matching and Contrastive Learning
    • Learning Spatial-Preserved Skeleton Representations for Few-Shot Action Recognition
  • 3D动作识别
    • Collaborating Domain-shared and Target-specific Feature Clustering for Cross-domain 3D Action Recognition
    • CMD: Self-supervised 3D Action Representation Learning with Cross-modal Mutual Distillation
      :open_mouth:oral:star:code
    • Continual 3D Convolutional Neural Networks for Real-Time Processing of Videos
      :star:code
    • Egocentric Activity Recognition and Localization on a 3D Map
  • 基于骨架动作识别
    • Global-local Motion Transformer for Unsupervised Skeleton-based Action Learning
      :star:code
• 社会群体活动识别
  • Hunting Group Clues with Transformers for Social Group Activity Recognition
  • Entry-Flipped Transformer for Inference and Prediction of Participant Behavior
  • Hunting Group Clues with Transformers for Social Group Activity Recognition
  • Self-Supervised Social Relation Representation for Human Group Detection
    :star:code
  • COMPOSER: Compositional Reasoning of Group Activity in Videos with Keypoint-Only Modality
    :star:code
• 时序动作检测
  • Semi-Supervised Temporal Action Detection with Proposal-Free Masking
    :star:code
  • Temporal Action Detection with Global Segmentation Mask Learning
    :star:code
  • ReAct: Temporal Action Detection with Relational Queries
    :star:code
  • Zero-Shot Temporal Action Detection via Vision-Language Prompting
    :star:code
  • Weakly-Supervised Temporal Action Detection for Fine-Grained Videos with Hierarchical Atomic Actions
    :star:code
  • TALLFormer: Temporal Action Localization with a Long-Memory Transformer
    :star:code
  • A Sliding Window Scheme for Online Temporal Action Localization
  • Proposal-Free Temporal Action Detection via Global Segmentation Mask Learning
    :star:code
• 时序动作定位
  • Dual-Evidential Learning for Weakly-Supervised Temporal Action Localization
    :star:code
  • ActionFormer: Localizing Moments of Actions with Transformers
    :star:code
• 时序动作分割
  • Unified Fully and Timestamp Supervised Temporal Action Segmentation via Sequence to Sequence Translation
  • Leveraging Action Affinity and Continuity for Semi-Supervised Temporal Action Segmentation
• Action Quality Assessment(行动质量评估)
  • Action Quality Assessment with Temporal Parsing Transformer
  • Pairwise Contrastive Learning Network for Action Quality Assessment
    :star:code
  • A Generalized & Robust Framework For Timestamp Supervision in Temporal Action Segmentation
• 动作定位
  • Actor-Centered Representations for Action Localization in Streaming Videos

11.Video

• Dynamic Temporal Filtering in Video Models
  :star:code
• Delta Distillation for Efficient Video Processing
• TDViT: Temporal Dilated Video Transformer for Dense Video Tasks
  :star:code
• 视频合成
  • Layered Controllable Video Generation
    :house:project
  • Sound-Guided Semantic Video Generation
    :house:project
  • Controllable Video Generation through Global and Local Motion Dynamics
    :house:project
  • Long Video Generation with Time-Agnostic VQGAN and Time-Sensitive Transformer
    :house:project
• 视频-视频合成
  • Fast-Vid2Vid: Spatial-Temporal Compression for Video-to-Video Synthesis
    :star:code:house:project
• 视频帧插值
  • A Perceptual Quality Metric for Video Frame Interpolation
    :star:code
• 视频生成
  • RealFlow: EM-based Realistic Optical Flow Dataset Generation from Videos
    :open_mouth:oral:star:code
• 视频质量评估
  • FAST-VQA: Efficient End-to-end Video Quality Assessment with Fragment Sampling
    :star:code
  • Telepresence Video Quality Assessment
• 视频修复
  • Error Compensation Framework for Flow-Guided Video Inpainting
  • Flow-Guided Transformer for Video Inpainting
    :star:code
  • Video Restoration Framework and Its Meta-Adaptations to Data-Poor Conditions
    :star:code
• 视频去模糊
  • Spatio-Temporal Deformable Attention Network for Video Deblurring
    :star:code:house:project
  • Efficient Video Deblurring Guided by Motion Magnitude
    :star:code
• 视频对话
  • Video Dialog as Conversation about Objects Living in Space-Time
    :star:code
• 有源扬声器检测(视频会议)
  • Learning Long-Term Spatial-Temporal Graphs for Active Speaker Detection
    :star:code
• VOS
  • XMem: Long-Term Video Object Segmentation with an Atkinson-Shiffrin Memory Model
    :star:code:house:project:tv:video
  • Tackling Background Distraction in Video Object Segmentation
    :star:code
  • BATMAN: Bilateral Attention Transformer in Motion-Appearance Neighboring Space for Video Object Segmentation
  • Hierarchical Feature Alignment Network for Unsupervised Video Object Segmentation
    :star:code
  • Learning Quality-aware Dynamic Memory for Video Object Segmentation
    :star:code
  • Global Spectral Filter Memory Network for Video Object Segmentation
    :star:code
• VIS
  • In Defense of Online Models for Video Instance Segmentation
    :open_mouth:oral:star:code
  • Instance As Identity: A Generic Online Paradigm for Video Instance Segmentation
    :star:code
  • Video Instance Segmentation via Multi-Scale Spatio-Temporal Split Attention Transformer
  • Less than Few: Self-Shot Video Instance Segmentation
    :star:code
  • Video Mask Transfiner for High-Quality Video Instance Segmentation
  • SeqFormer: Sequential Transformer for Video Instance Segmentation
    :star:code
• VSS
  • Mining Relations among Cross-Frame Affinities for Video Semantic Segmentation
    :star:code
  • Domain Adaptive Video Segmentation via Temporal Pseudo Supervision
    :star:code
  • Is It Necessary to Transfer Temporal Knowledge for Domain Adaptive Video Semantic Segmentation?
    :star:code
• VPS
  • Waymo Open Dataset: Panoramic Video Panoptic Segmentation
    :house:project
  • PolyphonicFormer: Unified Query Learning for Depth-Aware Video Panoptic Segmentation
    :star:code
• 视频抠图
  • One-Trimap Video Matting
    :star:code:tv:video
• 视频表征
  • E-NeRV: Expedite Neural Video Representation with Disentangled Spatial-Temporal Context
    :star:code
  • Static and Dynamic Concepts for Self-supervised Video Representation Learning
• 视频传输
  • Efficient Meta-Tuning for Content-aware Neural Video Delivery
    :star:code
• 运动分割
  • ParticleSfM: Exploiting Dense Point Trajectories for Localizing Moving Cameras in the Wild
    :star:code:house:project
• 视频异常检测
  • Video Anomaly Detection by Solving Decoupled Spatio-Temporal Jigsaw Puzzles
    :star:code
  • Towards Open Set Video Anomaly Detection
  • Scale-Aware Spatio-Temporal Relation Learning for Video Anomaly Detection
    :star:code
  • Self-Supervised Sparse Representation for Video Anomaly Detection
    :star:code
• 视频识别
  • Temporal Saliency Query Network for Efficient Video Recognition
    :house:project
  • NSNet: Non-saliency Suppression Sampler for Efficient Video Recognition
    :house:project
  • Expanding Language-Image Pretrained Models for General Video Recognition
    :open_mouth:oral:star:code
  • AdaFocusV3: On Unified Spatial-temporal Dynamic Video Recognition
  • DualFormer: Local-Global Stratified Transformer for Efficient Video Recognition
    :star:code
  • K-Centered Patch Sampling for Efficient Video Recognition
• 视频理解
  • Spotting Temporally Precise, Fine-Grained Events in Video
    :star:code:house:project
  • Point Primitive Transformer for Long-Term 4D Point Cloud Video Understanding
  • Panoramic Vision Transformer for Saliency Detection in 360° Videos
    :star:code
  • Streaming Multiscale Deep Equilibrium Models
    :house:project
  • Learning Shadow Correspondence for Video Shadow Detection
  • Federated Self-Supervised Learning for Video Understanding
    :star:code
  • Prompting Visual-Language Models for Efficient Video Understanding
  • GraphVid: It Only Takes a Few Nodes to Understand a Video
• 视频分类
  • Inductive and Transductive Few-Shot Video Classification via Appearance and Temporal Alignments
    :star:code
  • MorphMLP: An Efficient MLP-Like Backbone for Spatial-Temporal Representation Learning
    :star:code
• 视频卷帘快门(Rolling shutter)
  • Combining Internal and External Constraints for Unrolling Shutter in Videos
• Video Transition Effects(视频转场特效)
  • AutoTransition: Learning to Recommend Video Transition Effects
    :star:code
• 图像-视频编解码
  • AlphaVC: High-Performance and Efficient Learned Video Compression
  • A Cloud 3D Dataset and Application-Specific Learned Image Compression in Cloud 3D
    :star:code
  • CANF-VC: Conditional Augmented Normalizing Flows for Video Compression
    :star:code
  • Expanded Adaptive Scaling Normalization for End to End Image Compression
  • Coarse-to-Fine Sparse Transformer for Hyperspectral Image Reconstruction
    :star:code
  • Content Adaptive Latents and Decoder for Neural Image Compression
  • Contextformer: A Transformer with Spatio-Channel Attention for Context Modeling in Learned Image Compression
  • RAWtoBit: A Fully End-to-End Camera ISP Network
  • Content-Oriented Learned Image Compression
  • Implicit Neural Representations for Image Compression
  • Neural Video Compression Using GANs for Detail Synthesis and Propagation
• 视频摘要
  • TL;DW? Summarizing Instructional Videos with Task Relevance & Cross-Modal Saliency
    :star:code:house:project
• Video Grounding
  • Graph2Vid: Flow graph to Video Grounding forWeakly-supervised Multi-Step Localization
    :open_mouth:oral
• 帧插值
  • FILM: Frame Interpolation for Large Motion
    :house:project
  • Real-Time Intermediate Flow Estimation for Video Frame Interpolation
    :star:code
  • Deep Bayesian Video Frame Interpolation
    :star:code
  • Improving the Perceptual Quality of 2D Animation Interpolation
• 视频分析
  • Event Neural Networks
  • Sports Video Analysis on Large-Scale Data
    :star:code
• 视频编辑
  • Temporally Consistent Semantic Video Editing
• 视频增强
  • Learning Cross-Video Neural Representations for High-Quality Frame Interpolation
    :house:project
• 视频目标重识别
  • CAViT: Contextual Alignment Vision Transformer for Video Object Re-identification
    :star:code
• 图像视频编辑
  • Text2LIVE: Text-Driven Layered Image and Video Editing
• 视频升格
  • Learning Spatio-Temporal Downsampling for Effective Video Upscaling
• 视频色彩传播
  • Learned Variational Video Color Propagation
    :star:code
• 视听事件定位
  • Dual Perspective Network for Audio-Visual Event Localization
• 视频活动定位
  • Video Activity Localisation with Uncertainties in Temporal Boundary
• 视听视频解析
  • Joint-Modal Label Denoising for Weakly-Supervised Audio-Visual Video Parsing
    :star:code
• Video Highlight Detection
  • PAC-Net: Highlight Your Video via History Preference Modeling
• 视频片段分类
  • Long Movie Clip Classification with State-Space Video Models
    :star:code
• Video Relation Grounding
  • Asymmetric Relation Consistency Reasoning for Video Relation Grounding
• 视频片段检索
  • Selective Query-Guided Debiasing for Video Corpus Moment Retrieval

10.Pose Estimation(物体姿势估计)

• 物体姿势
  • Neural Correspondence Field for Object Pose Estimation
    :star:code:house:project
  • Zero-Shot Category-Level Object Pose Estimation
    :star:code
  • Fusing Local Similarities for Retrieval-based 3D Orientation Estimation of Unseen Objects
    :star:code:house:project
  • A Visual Navigation Perspective for Category-Level Object Pose Estimation
    :star:code
  • Polarimetric Pose Prediction
  • RayTran: 3D Pose Estimation and Shape Reconstruction of Multiple Objects from Videos with Ray-Traced Transformers
  • Gaussian Activated Neural Radiance Fields for High Fidelity Reconstruction & Pose Estimation
• 物体姿势变换
  • General Object Pose Transformation Network from Unpaired Data
• 抓取物体姿势估计
  • TransGrasp: Grasp Pose Estimation of a Category of Objects by Transferring Grasps from Only One Labeled Instance
    :star:code
• 4D
  • HuMMan: Multi-modal 4D Human Dataset for Versatile Sensing and Modeling
    :house:project
• 6D
  • Category-Level 6D Object Pose and Size Estimation using Self-Supervised Deep Prior Deformation Networks
    :star:code
  • Sim-to-Real 6D Object Pose Estimation via Iterative Self-Training for Robotic Bin Picking
    :house:project
  • Gen6D: Generalizable Model-Free 6-DoF Object Pose Estimation from RGB Images
    :house:project
  • Affine Correspondences between Multi-Camera Systems for 6DOF Relative Pose Estimation
    :star:code
  • ShAPO: Implicit Representations for Multi-Object Shape, Appearance, and Pose Optimization
    :house:project
  • RBP-Pose: Residual Bounding Box Projection for Category-Level Pose Estimation
    :star:code
  • Robust Category-Level 6D Pose Estimation with Coarse-to-Fine Rendering of Neural Features
  • Learning-based Point Cloud Registration for 6D Object Pose Estimation in the Real World
    :star:code
  • Perspective Flow Aggregation for Data-Limited 6D Object Pose Estimation
    :star:code
  • Object Level Depth Reconstruction for Category Level 6D Object Pose Estimation from Monocular RGB Image
  • DProST: Dynamic Projective Spatial Transformer Network for 6D Pose Estimation
    :star:code
  • WeLSA: Learning to Predict 6D Pose from Weakly Labeled Data Using Shape Alignment
  • DCL-Net: Deep Correspondence Learning Network for 6D Pose Estimation
    :star:code
  • DISP6D: Disentangled Implicit Shape and Pose Learning for Scalable 6D Pose Estimation
    :star:code
  • Vote from the Center: 6 DoF Pose Estimation in RGB-D Images by Radial Keypoint Voting
    :star:code
• 9D
  • CATRE: Iterative Point Clouds Alignment for Category-level Object Pose Refinement
    :star:code

9.Human Pose Estimation(人体姿态估计)

• Self-Constrained Inference Optimization on Structural Groups for Human Pose Estimation
• Pose for Everything: Towards Category-Agnostic Pose Estimation
  :open_mouth:oral:star:code
• BodySLAM: Joint Camera Localisation, Mapping, and Human Motion Tracking
• PoseTrans: A Simple Yet Effective Pose Transformation Augmentation for Human Pose Estimation
• Learning Visibility for Robust Dense Human Body Estimation
  :star:code
• D&D: Learning Human Dynamics from Dynamic Camera
  :open_mouth:oral:star:code
• PPT: token-Pruned Pose Transformer for monocular and multi-view human pose estimation
  :star:code
• DeciWatch: A Simple Baseline for 10× Efficient 2D and 3D Pose Estimation
  :star:code
• SmoothNet: A Plug-and-Play Network for Refining Human Poses in Videos
  :star:code
• Poseur: Direct Human Pose Regression with Transformers
  :star:code
• SimCC: A Simple Coordinate Classification Perspective for Human Pose Estimation
  :star:code
• Regularizing Vector Embedding in Bottom-Up Human Pose Estimation
  :star:code
• Hallucinating Pose-Compatible Scenes
• A Unified Framework for Domain Adaptive Pose Estimation
  :star:code
• 运动捕捉
  • TM2T: Stochastic and Tokenized Modeling for the Reciprocal Generation of 3D Human Motions and Texts
    :star:code:house:project
  • HULC: 3D HUman Motion Capture with Pose Manifold SampLing and Dense Contact Guidance
    :house:project
• 基于点的衣着人体建模
  • Learning Implicit Templates for Point-Based Clothed Human Modeling
    :star:code:house:project
• 动态人体数字化
  • NDF: Neural Deformable Fields for Dynamic Human Modelling
    :star:code
• 人体姿势与形状估计
  • CLIFF: Carrying Location Information in Full Frames into Human Pose and Shape Estimation
    :open_mouth:oral:star:code
  • Super-Resolution 3D Human Shape from a Single Low-Resolution Image
    :star:code:house:project
• 三维人体姿势估计
  • DH-AUG: DH Forward Kinematics Model Driven Augmentation for 3D Human Pose Estimation
    :star:code
  • Faster VoxelPose: Real-time 3D Human Pose Estimation by Orthographic Projection
  • Explicit Occlusion Reasoning for Multi-person 3D Human Pose Estimation
    :star:code
  • PoseScript: 3D Human Poses from Natural Language
    :house:project
  • Multi-Person 3D Pose and Shape Estimation via Inverse Kinematics and Refinement
  • 3D Human Pose Estimation Using Möbius Graph Convolutional Networks
  • P-STMO: Pre-trained Spatial Temporal Many-to-One Model for 3D Human Pose Estimation
    :star:code
  • C3P: Cross-Domain Pose Prior Propagation for Weakly Supervised 3D Human Pose Estimation
    :star:code
  • Structural Triangulation: A Closed-Form Solution to Constrained 3D Human Pose Estimation
    :star:code
  • VirtualPose: Learning Generalizable 3D Human Pose Models from Virtual Data
    :star:code
  • Learning to Fit Morphable Models
  • EgoBody: Human Body Shape and Motion of Interacting People from Head-Mounted Devices
    :house:project
  • AutoAvatar: Autoregressive Neural Fields for Dynamic Avatar Modeling
    :house:project
  • FLEX: Extrinsic Parameters-Free Multi-View 3D Human Motion Reconstruction
    :house:project
• Mul-Pose
  • Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation
    :star:code
• 三维人体重建
  • 3D Clothed Human Reconstruction in the Wild
    :star:code
  • DiffuStereo: High Quality Human Reconstruction via Diffusion-Based Stereo Using Sparse Cameras
  • UNIF: United Neural Implicit Functions for Clothed Human Reconstruction and Animation
    :star:code
  • The One Where They Reconstructed 3D Humans and Environments in TV Shows
    :star:code:house:project
  • Neural Capture of Animatable 3D Human from Monocular Video
  • SUPR: A Sparse Unified Part-Based Human Representation
    :star:code:house:project
  • IntegratedPIFu: Integrated Pixel Aligned Implicit Function for Single-view Human Reconstruction
    :star:code
  • Learned Vertex Descent:A New Direction for 3D Human Model Fitting
    :star:code:house:project
• 三维交互式手部姿势估计
  • 3D Interacting Hand Pose Estimation by Hand De-occlusion and Removal
    :star:code
  • S2Contact: Graph-based Network for 3D Hand-Object Contact Estimation with Semi-Supervised Learning
    :star:code:house:project
• 姿势合成
  • TIPS: Text-Induced Pose Synthesis
    :star:code:house:project
• 手物重建
  • AlignSDF: Pose-Aligned Signed Distance Fields for Hand-Object Reconstruction
    :star:code:house:project
• 人体与场景的交互
  • Compositional Human-Scene Interaction Synthesis with Semantic Control
    :star:code
• 人体姿势建模
  • Pose-NDF: Modeling Human Pose Manifolds with Neural Distance Fields
    :open_mouth:oral:house:project
• 姿势跟踪
  • AvatarPoser: Articulated Full-Body Pose Tracking from Sparse Motion Sensing
    :star:code
• 三维人体网格恢复
  • Cross-Attention of Disentangled Modalities for 3D Human Mesh Recovery with Transformers
    :star:code
• 三维人体运动预测与生成
  • Skeleton-Parted Graph Scattering Networks for 3D Human Motion Prediction
    :star:code
  • PoseGPT: Quantization-based 3D Human Motion Generation and Forecasting
    :house:project
• 姿势迁移
  • Skeleton-free Pose Transfer for Stylized 3D Characters
    :star:code:house:project
  • Cross Attention Based Style Distribution for Controllable Person Image Synthesis
    :star:code
• 人体姿势预测
  • Pose Forecasting in Industrial Human-Robot Collaboration
    :star:code
• 4D
  • LoRD: Local 4D Implicit Representation for High-Fidelity Dynamic Human Modeling
    :star:code:house:project
• 人体网格恢复
  • Self-supervised Human Mesh Recovery with Cross-Representation Alignment
• 手部网格估计
  • Identity-Aware Hand Mesh Estimation and Personalization from RGB Images
    :star:code
• 头部网格重建
  • Realistic One-Shot Mesh-Based Head Avatars
    :house:project
• 人体网格动画
  • CLIP-Actor: Text-Driven Recommendation and Stylization for Animating Human Meshes
• 音频驱动的风格化手势生成
  • Audio-Driven Stylized Gesture Generation with Flow-Based Model

8.3D(三维视觉)

• DeepPS2: Revisiting Photometric Stereo Using Two Differently Illuminated Images
  :star:code
• Towards High-Fidelity Single-view Holistic Reconstruction of Indoor Scenes
  :star:code
• Self-calibrating Photometric Stereo by Neural Inverse Rendering
  :star:code
• 3DG-STFM: 3D Geometric Guided Student-Teacher Feature Matching
  :star:code
• Learning Online Multi-sensor Depth Fusion
  :star:code
• Stereo Matching
  • PCW-Net: Pyramid Combination and Warping Cost Volume for Stereo Matching
  • EASNet: Searching Elastic and Accurate Network Architecture for Stereo Matching
    :star:code
• MVS
  • MVSTER: Epipolar Transformer for Efficient Multi-View Stereo
    :star:code
  • KD-MVS: Knowledge Distillation Based Self-Supervised Learning for Multi-View Stereo
    :star:code
  • RC-MVSNet: Unsupervised Multi-View Stereo with Neural Rendering
    :star:code
  • Multiview Stereo with Cascaded Epipolar RAFT
    :star:code
  • MVPS
    • PS-NeRF: Neural Inverse Rendering for Multi-view Photometric Stereo
      :house:project
• 3D场景合成
  • Simple and Effective Synthesis of Indoor 3D Scenes
    :tv:video
• 场景重建
  • Initialization and Alignment for Adversarial Texture Optimization
    :star:code:house:project
• 深度估计
  • Physical Attack on Monocular Depth Estimation with Optimal Adversarial Patches
  • Relationship Spatialization for Depth Estimation
  • BRNet: Exploring Comprehensive Features for Monocular Depth Estimation
    :star:code
  • Towards Scale-Aware, Robust, and Generalizable Unsupervised Monocular Depth Estimation by Integrating IMU Motion Dynamics
    :star:code
  • Stereo Depth Estimation with Echoes
  • JPerceiver: Joint Perception Network for Depth, Pose and Layout Estimation in Driving Scenes
    :star:code
  • RA-Depth: Resolution Adaptive Self-Supervised Monocular Depth Estimation
    :star:code
  • PointFix: Learning to Fix Domain Bias for Robust Online Stereo Adaptation
  • Depth Field Networks for Generalizable Multi-view Scene Representation
    :house:project
  • Structure and Motion from Casual Videos
  • MODE: Multi-View Omnidirectional Depth Estimation with 360° Cameras
    :star:code
  • Gradient-based Uncertainty for Monocular Depth Estimation
    :star:code
  • DevNet: Self-supervised Monocular Depth Learning via Density Volume Construction
    :star:code
  • Self-distilled Feature Aggregation for Self-supervised Monocular Depth Estimation
    :star:code
  • 3D-PL: Domain Adaptive Depth Estimation with 3D-aware Pseudo-Labeling
    :star:code
  • DELTAR: Depth Estimation from a Light-weight ToF Sensor and RGB Image
    :star:code:house:project
  • FloatingFusion: Depth from ToF and Image-stabilized Stereo Cameras
  • Context-Enhanced Stereo Transformer
  • Adaptive Co-Teaching for Unsupervised Monocular Depth Estimation
    :star:code
  • PanoFormer: Panorama Transformer for Indoor 360° Depth Estimation
    :star:code
  • Towards Comprehensive Representation Enhancement in Semantics-Guided Self-Supervised Monocular Depth Estimation
  • LocalBins: Improving Depth Estimation by Learning Local Distributions
    :star:code
  • Depth Map Decomposition for Monocular Depth Estimation
  • Uncertainty Quantification in Depth Estimation via Constrained Ordinal Regression
    :star:code
  • Spike Transformer: Monocular Depth Estimation for Spiking Camera
    :star:code
  • Learning Phase Mask for Privacy-Preserving Passive Depth Estimation
• 深度补全
  • GraphCSPN: Geometry-Aware Depth Completion via Dynamic GCNs
    :star:code
  • RigNet: Repetitive Image Guided Network for Depth Completion
  • Multi-modal Masked Pre-training for Monocular Panoramic Depth Completion
  • Monitored Distillation for Positive Congruent Depth Completion
    :star:code
  • CostDCNet: Cost Volume Based Depth Completion for a Single RGB-D Image
    :star:code
• 三维视觉
  • A Closer Look at Invariances in Self-supervised Pre-training for 3D Vision
  • Neural Density-Distance Fields
    :star:code
  • DANBO: Disentangled Articulated Neural Body Representations via Graph Neural Networks
    :house:project
  • CryoAI: Amortized Inference of Poses for Ab Initio Reconstruction of 3D Molecular Volumes from Real Cryo-EM Images
• 三维房间布局
  • 3D Room Layout Estimation from a Cubemap of Panorama Image via Deep Manhattan Hough Transform
    :star:code
• 三维重建
  • Object-Compositional Neural Implicit Surfaces
    :star:code:house:project:tv:video
  • Perspective Phase Angle Model for Polarimetric 3D Reconstruction
    :star:code
  • Monocular 3D Object Reconstruction with GAN Inversion
    :star:code:house:project
  • Structural Causal 3D Reconstruction
  • 2D GANs Meet Unsupervised Single-view 3D Reconstruction
    :star:code:house:project
  • Few-shot Single-view 3D Reconstruction with Memory Prior Contrastive Network
  • NeuRIS: Neural Reconstruction of Indoor Scenes Using Normal Priors
    :house:project
  • SparseNeuS: Fast Generalizable Neural Surface Reconstruction from Sparse Views
    :house:project
  • Disentangling Object Motion and Occlusion for Unsupervised Multi-Frame Monocular Depth
    :star:code
  • SNeS: Learning Probably Symmetric Neural Surfaces from Incomplete Data
  • CIRCLE: Convolutional Implicit Reconstruction and Completion for Large-Scale Indoor Scene
  • IS-MVSNet: Importance Sampling-Based MVSNet
    :star:code
  • Unbiased Gradient Estimation for Differentiable Surface Splatting via Poisson Sampling
    :star:code
  • Towards Learning Neural Representations from Shadows
  • PlaneFormers: From Sparse View Planes to 3D Reconstruction
    :star:code:house:project:tv:video
  • SimpleRecon: 3D Reconstruction Without 3D Convolutions
    :star:code
  • Share with Thy Neighbors: Single-View Reconstruction by Cross-Instance Consistency
  • SketchSampler: Sketch-Based 3D Reconstruction via View-Dependent Depth Sampling
  • Semi-Supervised Single-View 3D Reconstruction via Prototype Shape Priors
    :star:code
  • Bilateral Normal Integration
    :star:code
  • CHORE: Contact, Human and Object REconstruction from a Single RGB Image
    :star:code:house:project
  • Directed Ray Distance Functions for 3D Scene Reconstruction
    :house:project
  • Object Wake-Up: 3D Object Rigging from a Single Image
    :star:code:house:project
  • Latent Partition Implicit with Surface Codes for 3D Representation
    :star:code
  • 3D Equivariant Graph Implicit Functions
  • Projective Parallel Single-Pixel Imaging to Overcome Global Illumination in 3D Structure Light Scanning
  • EvAC3D: From Event-Based Apparent Contours to 3D Models via Continuous Visual Hulls
    :house:project
  • 3D CoMPaT: Composition of Materials on Parts of 3D Things
    :house:project
• 三维形状
  • Texturify: Generating Textures on 3D Shape Surfaces
  • Implicit Field Supervision for Robust Non-rigid Shape Matching
    :star:code
  • 3D Shape Sequence of Human Comparison and Classification using Current and Varifolds
    :star:code
  • The Shape Part Slot Machine: Contact-Based Reasoning for Generating 3D Shapes from Parts
• 3D形状匹配
  • Unsupervised Deep Multi-Shape Matching
• 3D形状合成
  • Cross-Modal 3D Shape Generation and Manipulation
    :star:code:house:project
• 形状补全
  • PatchRD: Detail-Preserving Shape Completion by Learning Patch Retrieval and Deformation
    :star:code
• 形状解析
  • ExtrudeNet: Unsupervised Inverse Sketch-and-Extrude for Shape Parsing
    :star:code
• 形状修补
  • DeepMend: Learning Occupancy Functions to Represent Shape for Repair
    :star:code
• depth restoration
  • Domain Randomization-Enhanced Depth Simulation and Restoration for Perceiving and Grasping Specular and Transparent Objects
    :star:code
• 场景理解
  • Spatially Invariant Unsupervised 3D Object-Centric Learning and Scene Decomposition
  • Pose2Room: Understanding 3D Scenes from Human Activities
  • Inverted Pyramid Multi-task Transformer for Dense Scene Understanding
    :star:code
  • Panoptic-PartFormer: Learning a Unified Model for Panoptic Part Segmentation
    :star:code
  • Weakly Supervised 3D Scene Segmentation with Region-Level Boundary Awareness and Instance Discrimination
  • 4DContrast: Contrastive Learning with Dynamic Correspondences for 3D Scene Understanding
  • Point Scene Understanding via Disentangled Instance Mesh Reconstruction
    :star:code
  • PIP: Physical Interaction Prediction via Mental Simulation with Span Selection
    :house:project

7.Object Tracking(目标跟踪)

• Towards Grand Unification of Object Tracking
  :open_mouth:oral:star:code
  :newspaper:ECCV 2022 Oral《Unicorn》首次统一了四项目标跟踪任务的网络结构与学习范式，在8个富有挑战性的数据集上SOTA
• HVC-Net: Unifying Homography, Visibility, and Confidence Learning for Planar Object Tracking
• Tracking by Associating Clips
• ByteTrack: Multi-Object Tracking by Associating Every Detection Box
  :star:code
• Joint Feature Learning and Relation Modeling for Tracking: A One-Stream Framework
  :star:code
• Backbone Is All Your Need: A Simplified Architecture for Visual Object Tracking
  :star:code
• Robust Visual Tracking by Segmentation
  :star:code
• FEAR: Fast, Efficient, Accurate and Robust Visual Tracker
  :star:code
• 3D跟踪
  • 3D Siamese Transformer Network for Single Object Tracking on Point Clouds
    :star:code
  • SpOT: Spatiotemporal Modeling for 3D Object Tracking
  • Large-displacement 3D Object Tracking with Hybrid Non-local Optimization
    :star:code
  • CMT: Context-Matching-Guided Transformer for 3D Tracking in Point Clouds
  • Towards Generic 3D Tracking in RGBD Videos: Benchmark and Baseline
    :star:code
• 多目标跟踪
  • Tracking Objects as Pixel-wise Distributions
    :open_mouth:oral
  • The Caltech Fish Counting Dataset: A Benchmark for Multiple-Object Tracking and Counting
  • MOTCOM: The Multi-Object Tracking Dataset Complexity Metric
    :star:code:house:project
  • Tracking Every Thing in the Wild
  • PolarMOT: How Far Can Geometric Relations Take Us in 3D Multi-Object Tracking?
  • SOMPT22: A Surveillance Oriented Multi-Pedestrian Tracking Dataset
  • Robust Multi-Object Tracking by Marginal Inference
  • MOTR: End-to-End Multiple-Object Tracking with TRansformer
    :star:code
  • Large Scale Real-World Multi-person Tracking
    :star:code
  • Particle Video Revisited: Tracking through Occlusions Using Point Trajectories
    :house:project
• 视觉跟踪
  • AiATrack: Attention in Attention for Transformer Visual Tracking
    :star:code
  • Towards Sequence-Level Training for Visual Tracking
    :star:code
  • Hierarchical Feature Embedding for Visual Tracking
    :star:code
• 细胞跟踪
  • Graph Neural Network for Cell Tracking in Microscopy Videos
    :star:code

6.Object Detection(目标检测)

• Should All Proposals be Treated Equally in Object Detection?
  :star:code
• TIDEE: Tidying Up Novel Rooms Using Visuo-Semantic Commonsense Priors
  :house:project
• TALISMAN: Targeted Active Learning for Object Detection with Rare Classes and Slices Using Submodular Mutual Information
  :star:code
• HEAD: HEtero-Assists Distillation for Heterogeneous Object Detectors
  :star:code
• Adversarially-Aware Robust Object Detector
  :open_mouth:oral:star:code
• ObjectBox: From Centers to Boxes for Anchor-Free Object Detection
  :open_mouth:oral:star:code
• Point-to-Box Network for Accurate Object Detection via Single Point Supervision
  :star:code
• You Should Look at All Objects
  :star:code
• Class-agnostic Object Detection with Multi-modal Transformer
  :star:code
  使用多模态 ViTs 和人类可理解的文本查询来生成高质量的OP
• Exploiting Unlabeled Data with Vision and Language Models for Object Detection
  :star:code
• PoserNet: Refining Relative Camera Poses Exploiting Object Detections
  :star:code
• Robust Object Detection With Inaccurate Bounding Boxes
  :star:code
• UC-OWOD: Unknown-Classified Open World Object Detection
  :star:code
• Exploring Resolution and Degradation Clues as Self-supervised Signal for Low Quality Object
  :star:code
• Unifying Visual Perception by Dispersible Points Learning
  :star:code
• A Large-scale Multiple-objective Method for Black-box Attack against Object Detection
  :star:code
• Distilling Object Detectors With Global Knowledge
  :star:code
• PANDORA: A Panoramic Detection Dataset for Object with Orientation
  :star:code
• Exploring Plain Vision Transformer Backbones for Object Detection
  :star:code
• Long-Tail Detection with Effective Class-Margins
  :star:code
• Detecting Twenty-Thousand Classes Using Image-Level Supervision
  :star:code
• Exploring Resolution and Degradation Clues As Self-Supervised Signal for Low Quality Object Detection
  :star:code
• Translation, Scale and Rotation: Cross-Modal Alignment Meets RGB-Infrared Vehicle Detection
• MTTrans: Cross-Domain Object Detection with Mean Teacher Transformer
• PromptDet: Towards Open-Vocabulary Detection Using Uncurated Images
  :house:project
• Cornerformer: Purifying Instances for Corner-Based Detectors
• Efficient Decoder-Free Object Detection with Transformers
  :star:code
• W2N: Switching from Weak Supervision to Noisy Supervision for Object Detection
  :star:code
• Towards Data-Efficient Detection Transformers
  :star:code
• Open-Vocabulary DETR with Conditional Matching
  :star:code
• Prediction-Guided Distillation for Dense Object Detection
  :star:code
• Multimodal Object Detection via Probabilistic Ensembling
  :star:code
• Open Vocabulary Object Detection with Pseudo Bounding-Box Labels
• GLAMD: Global and Local Attention Mask Distillation for Object Detectors
• Object Detection As Probabilistic Set Prediction
• Out-of-Distribution Identification: Let Detector Tell Which I Am Not Sure
• Simple Open-Vocabulary Object Detection with Vision Transformers
  :star:code
• A Simple Approach and Benchmark for 21,000-Category Object Detection
  :star:code
• EAutoDet: Efficient Architecture Search for Object Detection
  :star:code
• Few-Shot End-to-End Object Detection via Constantly Concentrated Encoding across Heads
• 3D目标检测
  • DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection
    :star:code
  • Rethinking IoU-based Optimization for Single-stage 3D Object Detection
    :star:code
  • Densely Constrained Depth Estimator for Monocular 3D Object Detection
    :star:code
  • Learning Ego 3D Representation As Ray Tracing
    :house:project
  • LiDAR Distillation: Bridging the Beam-Induced Domain Gap for 3D Object Detection
    :star:code
  • SpatialDETR: Robust Scalable Transformer-Based 3D Object Detection from Multi-View Camera Images with Global Cross-Sensor Attention
    :star:code
  • AutoAlignV2: Deformable Feature Aggregation for Dynamic Multi-Modal 3D Object Detection
    :star:code
  • DEVIANT: Depth EquiVarIAnt NeTwork for Monocular 3D Object Detection
    :star:code
  • Label-Guided Auxiliary Training Improves 3D Object Detector
    :star:code
  • Monocular 3D Object Detection with Depth from Motion
    :open_mouth:oral:star:code
  • MV-FCOS3D++: Multi-View Camera-Only 4D Object Detection with Pretrained Monocular Backbones
    :open_mouth:oral:star:code
  • Graph R-CNN: Towards Accurate 3D Object Detection with Semantic-Decorated Local Graph
    :open_mouth:oral:star:code
  • Multimodal Transformer for Automatic 3D Annotation and Object Detection
    :star:code
  • Semi-Supervised 3D Object Detection with Proficient Teachers
    :star:code
  • ProposalContrast: Unsupervised Pre-training for LiDAR-Based 3D Object Detection
    :star:code
  • CenterFormer: Center-based Transformer for 3D Object Detection
    :open_mouth:oral:star:code
  • SWFormer: Sparse Window Transformer for 3D Object Detection in Point Clouds
  • Autoregressive Uncertainty Modeling for 3D Bounding Box Prediction
    :star:code
  • CramNet: Camera-Radar Fusion with Ray-Constrained Cross-Attention for Robust 3D Object Detection
  • Homogeneous Multi-modal Feature Fusion and Interaction for 3D Object Detection
  • Plausibility Verification For 3D Object Detectors Using Energy-Based Optimization
  • Cross-Modality Knowledge Distillation Network for Monocular 3D Object Detection
    :star:code
  • PETR: Position Embedding Transformation for Multi-View 3D Object Detection
    :star:code
  • Lidar Point Cloud Guided Monocular 3D Object Detection
    :star:code
  • INT: Towards Infinite-Frames 3D Detection with an Efficient Framework
  • Semi-Supervised Monocular 3D Object Detection by Multi-View Consistency
  • Unsupervised Domain Adaptation for Monocular 3D Object Detection via Self-Training
    :star:code
  • MPPNet: Multi-Frame Feature Intertwining with Proxy Points for 3D Temporal Object Detection
    :star:code
  • PillarNet: Real-Time and High-Performance Pillar-Based 3D Object Detection
    :star:code
  • Improving the Intra-Class Long-Tail in 3D Detection via Rare Example Mining
  • 3D Object Detection with a Self-Supervised Lidar Scene Flow Backbone
    :star:code
  • DetMatch: Two Teachers Are Better than One for Joint 2D and 3D Semi-Supervised Object Detection
    :star:code
  • FCAF3D: Fully Convolutional Anchor-Free 3D Object Detection
    :star:code
  • Enhancing Multi-modal Features Using Local Self-Attention for 3D Object Detection
• 半监督目标检测
  • Dense Teacher: Dense Pseudo-Labels for Semi-supervised Object Detection
    :star:code
  • Semi-Supervised Object Detection via Virtual Category Learning
    :star:code
  • Open-Set Semi-Supervised Object Detection
    :star:code:house:project
  • PseCo: Pseudo Labeling and Consistency Training for Semi-Supervised Object Detection
    :star:code
  • Diverse Learner: Exploring Diverse Supervision for Semi-Supervised Object Detection
• 小样本目标检测
  • Rethinking Few-Shot Object Detection on a Multi-Domain Benchmark
    :star:code
  • Multi-Faceted Distillation of Base-Novel Commonality for Few-shot Object Detection
    :star:code
  • AcroFOD: An Adaptive Method for Cross-domain Few-shot Object Detection
    :star:code
  • Time-rEversed diffusioN tEnsor Transformer: A new TENET of Few-Shot Object Detection
    :star:code
  • AirDet: Few-Shot Detection without Fine-Tuning for Autonomous Exploration
    :star:code
  • Few-Shot Object Detection by Knowledge Distillation Using Bag-of-Visual-Words Representations
  • Few-Shot Object Detection with Model Calibration
    :star:code
  • Few-Shot Video Object Detection
    :star:code
  • Mutually Reinforcing Structure with Proposal Contrastive Consistency for Few-Shot Object Detection
    :star:code
• 显著目标检测
  • SESS: Saliency Enhancing with Scaling and Sliding
    :star:code
  • SPSN: Superpixel Prototype Sampling Network for RGB-D Salient Object Detection
    :star:code
  • Salient Object Detection for Point Clouds
    :star:code
  • KD-SCFNet: Towards More Accurate and Efficient Salient Object Detection via Knowledge Distillation
    :star:code
  • Saliency Hierarchy Modeling via Generative Kernels for Salient Object Detection
  • MVSalNet:Multi-View Augmentation for RGB-D Salient Object Detection
• 弱监督目标检测
  • Active Learning Strategies for Weakly-supervised Object Detection
    :star:code
  • W2N:Switching From Weak Supervision to Noisy Supervision for Object Detection
    :star:code
  • Object Discovery via Contrastive Learning for Weakly Supervised Object Detection
    :star:code
  • End-to-End Weakly Supervised Object Detection with Sparse Proposal Evolution
• 目标定位
  • Object Manipulation via Visual Target Localization
    :house:project
  • On Label Granularity and Object Localization
  • 弱监督目标定位
    • Bagging Regional Classification Activation Maps for Weakly Supervised Object Localization
      :star:code
    • Weakly Supervised Object Localization through Inter-class Feature Similarity and Intra-Class Appearance Consistency
    • Weakly Supervised Object Localization via Transformer with Implicit Spatial Calibration
      :star:code
• 单阶目标检测
  • Unsupervised Domain Adaptation for One-stage Object Detector using Offsets to Bounding Box
• 目标计数
  • Few-shot Object Counting and Detection
    :star:code
  • Class-Agnostic Object Counting Robust to Intraclass Diversity
    :star:code
• OOD
  • Out-of-Distribution Detection with Semantic Mismatch under Masking
    :star:code
  • Out-of-Distribution Detection with Boundary Aware Learning
  • DICE: Leveraging Sparsification for Out-of-Distribution Detection
    :star:code
  • Class Is Invariant to Context and Vice Versa: On Learning Invariance for Out-of-Distribution Generalization
    :star:code
  • Data Invariants to Understand Unsupervised Out-of-Distribution Detection
• VOD
  • PTSEFormer: Progressive Temporal-Spatial Enhanced TransFormer towards Video Object Detection
    :star:code
  • SALISA: Saliency-Based Input Sampling for Efficient Video Object Detection
  • Bridging Images and Videos: A Simple Learning Framework for Large Vocabulary Video Object Detection
  • Efficient One-Stage Video Object Detection by Exploiting Temporal Consistency
    :star:code
• 小目标检测
  • RFLA: Gaussian Receptive Field based Label Assignment for Tiny Object Detection
    :star:code
• 图像检测
  • Discovering Transferable Forensic Features for CNN-generated Images Detection
    :open_mouth:oral:star:code:house:project
• 目标发现
  • Object Discovery and Representation Networks
• 变化检测
  • Objects Can Move: 3D Change Detection by Geometric Transformation Consistency
    :star:code

5.Image/Video Retrieval(图像/视频检索)

• Text-Based Temporal Localization of Novel Events
• 跨域检索
  • Feature Representation Learning for Unsupervised Cross-domain Image Retrieval
    :star:code
• 图像检索
  • Hierarchical Average Precision Training for Pertinent Image Retrieval
    :star:code
  • Adaptive Fine-Grained Sketch-Based Image Retrieval
    :star:code
  • A Sketch Is Worth a Thousand Words: Image Retrieval with Text and Sketch
    :star:code:house:project
  • Granularity-aware Adaptation for Image Retrieval over Multiple Tasks
  • Reliability-Aware Prediction via Uncertainty Learning for Person Image Retrieval
    :star:code
  • StyleBabel: Artistic Style Tagging and Captioning
  • Deep Hash Distillation for Image Retrieval
    :star:code
  • Conditional Stroke Recovery for Fine-Grained Sketch-Based Image Retrieval
  • Fine-Grained Fashion Representation Learning by Online Deep Clustering
• 视频检索
  • LocVTP: Video-Text Pre-training for Temporal Localization
    :star:code
  • Dual-Stream Knowledge-Preserving Hashing for Unsupervised Video Retrieval
  • Multi-Query Video Retrieval
    :star:code
  • Learning Audio-Video Modalities from Image Captions
    :house:project
  • Audio-Visual Mismatch-Aware Video Retrieval via Association and Adjustment
  • ECLIPSE: Efficient Long-Range Video Retrieval Using Sight and Sound
    :star:code
  • Video Geo-localization(检索)
    • GAMa: Cross-view Video Geo-localization
      :star:code
• 文本-视频检索
  • TS2-Net: Token Shift and Selection Transformer for Text-Video Retrieval
    :star:code
  • Lightweight Attentional Feature Fusion: A New Baseline for Text-to-Video Retrieval
    :star:code
  • Learning Linguistic Association towards Efficient Text-Video Retrieval
    :star:code
• 图像-文本检索
  • CODER: Coupled Diversity-Sensitive Momentum Contrastive Learning for Image-Text Retrieval
• 细粒度图像检索
  • SEMICON: A Learning-to-hash Solution for Large-scale Fine-grained Image Retrieval
    :star:code
• 视频时刻检索
  • Selective Query-guided Debiasing Network for Video Corpus Moment Retrieval
• 视频-文本检索
  • VTC: Improving Video-Text Retrieval with User Comments
    :star:code:house:project
  • MILES: Visual BERT Pre-training with Injected Language Semantics for Video-Text Retrieval
• 最近邻搜索
  • Connecting Compression Spaces with Transformer for Approximate Nearest Neighbor Search
    :star:code

4.Video/Image Captioning(视频/图像字幕)

• D3Net: A Unified Speaker-Listener Architecture for 3D Dense Captioning and Visual Grounding
  :house:project
• 图像字幕
  • GRIT: Faster and Better Image captioning Transformer Using Dual Visual Features
    :star:code
  • Explicit Image Caption Editing
    :star:code
  • ECCV Caption: Correcting False Negatives by Collecting Machine-and-Human-Verified Image-Caption Associations for MS-COCO
    :star:code
  • GEB+: A Benchmark for Generic Event Boundary Captioning, Grounding and Retrieval
    :star:code
  • Object-Centric Unsupervised Image Captioning
    :star:code
  • Unifying Event Detection and Captioning as Sequence Generation via Pre-training
    :star:code

3.Image Progress(图像处理)

• 图像质量评估
  • Shift-tolerant Perceptual Similarity Metric
    :star:code
• 图像修补(image retouching)
  • Neural Color Operators for Sequential Image Retouching
    :star:code
• 图像变形(Image Warping)
  • Learning Local Implicit Fourier Representation for Image Warping
    :star:code
• 图像恢复
  • D2HNet: Joint Denoising and Deblurring with Hierarchical Network for Robust Night Image Restoration
    :star:code
  • Simple Baselines for Image Restoration
    :star:code
  • Improving Image Restoration by Revisiting Global Information Aggregation
    :star:code
  • Seeing through a Black Box: Toward High-Quality Terahertz Imaging via Subspace-and-Attention Guided Restoration
  • JPEG Artifacts Removal via Contrastive Representation Learning
    :star:code
  • TAPE: Task-Agnostic Prior Embedding for Image Restoration
  • Spectrum-Aware and Transferable Architecture Search for Hyperspectral Image Restoration
  • DRCNet: Dynamic Image Restoration Contrastive Network
• 图像修复
  • Learning Prior Feature and Attention Enhanced Image Inpainting
    :star:code
  • Inpainting at Modern Camera Resolution by Guided PatchMatch with Auto-Curation
    :star:code
  • High-Fidelity Image Inpainting with GAN Inversion
  • Unbiased Multi-Modality Guidance for Image Inpainting
  • Image Inpainting with Cascaded Modulation GAN and Object-Aware Training
    :star:code
  • Perceptual Artifacts Localization for Inpainting
    :star:code
  • Hourglass Attention Network for Image Inpainting
    :star:code
  • Diverse Image Inpainting with Normalizing Flow
• 图像增强
  • SepLUT: Separable Image-adaptive Lookup Tables for Real-time Image Enhancement
  • Uncertainty Inspired Underwater Image Enhancement
    :star:code
  • Unsupervised Night Image Enhancement: When Layer Decomposition Meets Light-Effects Suppression
    :star:code
  • LEDNet: Joint Low-Light Enhancement and Deblurring in the Dark
    :star:code:house:project
  • NEST: Neural Event Stack for Event-Based Image Enhancement
    :star:code
  • Seeing Far in the Dark with Patterned Flash
    :star:code
  • Local Color Distributions Prior for Image Enhancement
    :house:project
  • SemAug: Semantically Meaningful Image Augmentations for Object Detection through Language Grounding
• 图像和谐化
  • DCCF: Deep Comprehensible Color Filter Learning Framework for High-Resolution Image Harmonization
    :open_mouth:oral:star:code
  • Spatial-Separated Curve Rendering Network for Efficient and High-Resolution Image Harmonization
    :star:code
  • Semantic-Guided Multi-Mask Image Harmonization
    :star:code
• 图像去卷积
  • Learning Discriminative Shrinkage Deep Networks for Image Deconvolution
• 去雾
  • Boosting Supervised Dehazing Methods via Bi-Level Patch Reweighting
  • Unpaired Deep Image Dehazing Using Contrastive Disentanglement Learning
  • Perceiving and Modeling Density for Image Dehazing
    :star:code
  • Frequency and Spatial Dual Guidance for Image Dehazing
    :star:code
• 去噪
  • Deep Semantic Statistics Matching (D2SM) Denoising Network
    :star:code:house:project
  • Optimizing Image Compression via Joint Learning with Denoising
    :star:code
  • Fast and High Quality Image Denoising via Malleable Convolution
    :house:project
  • Unidirectional Video Denoising by Mimicking Backward Recurrent Modules with Look-Ahead Forward Ones
    :star:code
  • TempFormer: Temporally Consistent Transformer for Video Denoising
• 去雪
  • SLiDE: Self-supervised LiDAR De-snowing through Reconstruction Difficulty
• 去雨
  • Not Just Streaks: Towards Ground Truth for Single Image Deraining
    :house:project
  • Blind Image Decomposition
    :star:code
  • ART-SS: An Adaptive Rejection Technique for Semi-Supervised Restoration for Adverse Weather-Affected Images
    :star:code
  • Rethinking Video Rain Streak Removal: A New Synthesis Model and a Deraining Network with Video Rain Prior
    :star:code
• 去模糊
  • Animation from Blur: Multi-modal Blur Decomposition with Motion Guidance
    :star:code
  • United Defocus Blur Detection and Deblurring via Adversarial Promoting Learning
    :star:code
  • Learning Degradation Representations for Image Deblurring
    :star:code
  • Learning Deep Non-Blind Image Deconvolution without Ground Truths
  • DeMFI: Deep Joint Deblurring and Multi-Frame Interpolation with Flow-Guided Attentive Correlation and Recursive Boosting
    :star:code
  • Realistic Blur Synthesis for Learning Image Deblurring
    :house:project
  • Stripformer: Strip Transformer for Fast Image Deblurring
    :star:code
  • Event-Based Fusion for Motion Deblurring with Cross-Modal Attention
    :house:project
  • ERDN: Equivalent Receptive Field Deformable Network for Video Deblurring
    :star:code
  • Event-Guided Deblurring of Unknown Exposure Time Videos
    :house:project
• 去摩尔纹
  • Towards Efficient and Scale-Robust Ultra-High-Definition Image Demoireing
    :star:code:house:project
• 去反射
  • Zero-Shot Learning for Reflection Removal of Single 360-Degree Image
• 去阴影
  • Style-Guided Shadow Removal
    :star:code
• 语义图像编辑
  • Context-Consistent Semantic Image Editing with Style-Preserved Modulation
    :star:code
• 图像着色
  • PalGAN: Image Colorization with Palette Generative Adversarial Networks
    :star:code
  • Semantic-Sparse Colorization Network for Deep Exemplar-Based Colorization
  • CT2: Colorization Transformer via Color Tokens
  • BigColor: Colorization Using a Generative Color Prior for Natural Images
  • Colorization for In Situ Marine Plankton Images
  • ColorFormer: Image Colorization via Color Memory Assisted Hybrid-Attention Transformer
    :star:code
  • Bridging the Domain Gap towards Generalization in Automatic Colorization
    :star:code
  • L-CoDer: Language-Based Colorization with Color-Object Decoupling Transformer
• 图像裁剪
  • Human-Centric Image Cropping with Partition-Aware and Content-Preserving Features
    :star:code
• 图像融合
  • Neural Image Representations for Multi-Image Fusion and Layer Separation
    :house:project
• Rolling shutter(果冻效应)
  • Bringing Rolling Shutter Images Alive with Dual Reversed Distortion
    :star:code

2.Image Segmentation(图像分割)

• PseudoClick: Interactive Image Segmentation with Click Imitation
• GitNet: Geometric Prior-Based Transformation for Birds-Eye-View Segmentation
• Pixel-Wise Energy-Biased Abstention Learning for Anomaly Segmentation on Complex Urban Driving Scenes
  :open_mouth:oral:star:code
• Highly Accurate Dichotomous Image Segmentation
  :house:project
• Graph-Constrained Contrastive Regularization for Semi-Weakly Volumetric Segmentation
• Slim Scissors: Segmenting Thin Object from Synthetic Background
  :star:code
• RankSeg: Adaptive Pixel Classification with Image Category Ranking for Segmentation
  :star:code
• Unsupervised Segmentation in Real-World Images via Spelke Object Inference
• Learning Instance-Specific Adaptation for Cross-Domain Segmentation
  :house:project
• Scaling Open-Vocabulary Image Segmentation with Image-Level Labels
• Uncertainty-Aware Multi-modal Learning via Cross-Modal Random Network Prediction
• Fashionformer: A Simple, Effective and Unified Baseline for Human Fashion Segmentation and Recognition
  :star:code
• 语义分割
  • Multi-Exit Semantic Segmentation Networks
  • Language-Grounded Indoor 3D Semantic Segmentation in the Wild
    :house:project
  • Where in the World Is This Image? Transformer-Based Geo-Localization in the Wild
  • Open-World Semantic Segmentation for LIDAR Point Clouds
    :star:code
  • SiamDoGe: Domain Generalizable Semantic Segmentation Using Siamese Network
    :star:code
  • TACS: Taxonomy Adaptive Cross-Domain Semantic Segmentation
    :star:code
  • Prototypical Contrast Adaptation for Domain Adaptive Semantic Segmentation
    :star:code
  • RBC: Rectifying the Biased Context in Continual Semantic Segmentation
    :star:code
  • ESS: Learning Event-Based Semantic Segmentation from Still Images
    :house:project
  • Learning Implicit Feature Alignment Function for Semantic Segmentation
    :star:code
  • Data Efficient 3D Learner via Knowledge Transferred from 2D Model
    :star:code
  • Multi-Scale and Cross-Scale Contrastive Learning for Semantic Segmentation
    :star:code
  • 2DPASS: 2D Priors Assisted Semantic Segmentation on LiDAR Point Clouds
    :star:code
  • Open-world Semantic Segmentation via Contrasting and Clustering Vision-Language Embedding
  • ML-BPM: Multi-teacher Learning with Bidirectional Photometric Mixing for Open Compound Domain Adaptation in Semantic Segmentation
  • Union-Set Multi-source Model Adaptation for Semantic Segmentation
  • Continual Semantic Segmentation via Structure Preserving and Projected Feature Alignment
  • Multi-Granularity Distillation Scheme Towards Lightweight Semi-Supervised Semantic Segmentation
    :star:code
  • LiDAL: Inter-Frame Uncertainty Based Active Learning for 3D LiDAR Semantic Segmentation
    :star:code
  • DODA: Data-Oriented Sim-to-Real Domain Adaptation for 3D Semantic Segmentation
    :star:code
  • SQN: Weakly-Supervised Semantic Segmentation of Large-Scale 3D Point Clouds
    :star:code
  • Learning Semantic Segmentation from Multiple Datasets with Label Shifts
    :house:project
  • CAR: Class-Aware Regularizations for Semantic Segmentation
    :star:code
  • Style-Hallucinated Dual Consistency Learning for Domain Generalized Semantic Segmentation
    :star:code
  • A Transformer-Based Decoder for Semantic Segmentation with Multi-level Context Mining
    :star:code
  • Extract Free Dense Labels from CLIP
  • A Simple Baseline for Open-Vocabulary Semantic Segmentation with Pre-trained Vision-Language Model
    :star:code
  • UCTNet: Uncertainty-Aware Cross-Modal Transformer Network for Indoor RGB-D Semantic Segmentation
  • CP2: Copy-Paste Contrastive Pretraining for Semantic Segmentation
    :star:code
  • 域适应语义分割
    • DecoupleNet: Decoupled Network for Domain Adaptive Semantic Segmentation
      :star:code
    • HRDA: Context-Aware High-Resolution Domain-Adaptive Semantic Segmentation
      :star:code
    • D2ADA: Dynamic Density-Aware Active Domain Adaptation for Semantic Segmentation
    • Online Domain Adaptation for Semantic Segmentation in Ever-Changing Conditions
      :star:code
    • Bi-directional Contrastive Learning for Domain Adaptive Semantic Segmentation
      :star:code:house:project
  • 小样本语义分割
    • Self-Support Few-Shot Semantic Segmentation
      :star:code
    • Cross-Domain Few-Shot Semantic Segmentation
      :star:code
  • 弱监督语义分割
    • Max Pooling with Vision Transformers reconciles class and shape in weakly supervised semantic segmentation
      :star:code
    • Adaptive Spatial-BCE Loss for Weakly Supervised Semantic Segmentation
      :star:code
  • Adversarial Erasing Framework via Triplet with Gated Pyramid Pooling Layer for Weakly Supervised Semantic Segmentation
    :star:code
  • 无监督语义分割
    • TransFGU: A Top-down Approach to Fine-Grained Unsupervised Semantic Segmentation
      :star:code
• 实例分割
  • OSFormer: One-Stage Camouflaged Instance Segmentation with Transformers
    :star:code
  • Geodesic-Former: A Geodesic-Guided Few-Shot 3D Point Cloud Instance Segmenter
    :star:code
  • Learning Regional Purity for Instance Segmentation on 3D Point Clouds
  • 3D Instances as 1D Kernels
    :star:code
  • 2D Amodal Instance Segmentation Guided by 3D Shape Prior
  • Box-supervised Instance Segmentation with Level Set Evolution
    :star:code
  • Long-tailed Instance Segmentation using Gumbel Optimized Loss
    :star:code
  • Active Pointly-Supervised Instance Segmentation
  • Trapped in Texture Bias? A Large Scale Comparison of Deep Instance Segmentation
    :star:code
  • Learning with Free Object Segments for Long-Tailed Instance Segmentation
    :star:code
  • A Simple Single-Scale Vision Transformer for Object Detection and Instance Segmentation
    :star:code
  • Box2Mask: Weakly Supervised 3D Semantic Instance Segmentation Using Bounding Boxes
    :house:project
  • Learning to Detect Every Thing in an Open World
    :house:project
• 全景分割
  • Pointly-Supervised Panoptic Segmentation
    :star:code
• 运动分割
  • Fast Two-View Motion Segmentation Using Christoffel Polynomials
  • Quantum Motion Segmentation
• 小样本分割
  • Dense Cross-Query-and-Support Attention Weighted Mask Aggregation for Few-Shot Segmentation
    :star:code
  • Cost Aggregation with 4D Convolutional Swin Transformer for Few-Shot Segmentation
    :star:code:house:project
  • Doubly Deformable Aggregation of Covariance Matrices for Few-shot Segmentation
    :star:code
  • Interclass Prototype Relation for Few-Shot Segmentation
  • HM: Hybrid Masking for Few-Shot Segmentation
    :star:code
  • Adaptive Agent Transformer for Few-Shot Segmentation
  • Dense Gaussian Processes for Few-Shot Segmentation
    :star:code
• 抠图
  • TransMatting: Enhancing Transparent Objects Matting with Transformers
    :star:code
• 3D分割
  • MvDeCor: Multi-view Dense Correspondence Learning for Fine-grained 3D Segmentation
    :star:code:house:project
  • PointInst3D: Segmenting 3D Instances by Points
• 手分割
  • Generative Adversarial Network for Future Hand Segmentation from Egocentric Video
    :house:project
• 零件分割
  • Improving Few-Shot Part Segmentation Using Coarse Supervision
• 场景分割
  • Master of All: Simultaneous Generalization of Urban-Scene Segmentation to All Adverse Weather Conditions
    :house:project

1.其它

• Generative Meta-Adversarial Network for Unseen Object Navigation
  :star:code
• Housekeep: Tidying Virtual Households Using Commonsense Reasoning
  :house:project
• OPD: Single-View 3D Openable Part Detection
  :house:project
• Referring Object Manipulation of Natural Images with Conditional Classifier-Free Guidance
• Webly Supervised Concept Expansion for General Purpose Vision Models
  :house:project
• PACS: A Dataset for Physical Audiovisual Commonsense Reasoning
• Fabric Material Recovery from Video Using Multi-Scale Geometric Auto-Encoder
• MegBA: A GPU-Based Distributed Library for Large-Scale Bundle Adjustment
  :star:code
• Bandwidth-Aware Adaptive Codec for DNN Inference Offloading in IoT
• Efficient Deep Visual and Inertial Odometry with Adaptive Visual Modality Selection
  :star:code
• Flow Graph to Video Grounding for Weakly-Supervised Multi-step Localization
  :star:code
• Bayesian Tracking of Video Graphs Using Joint Kalman Smoothing and Registration
• SeqTR: A Simple Yet Universal Network for Visual Grounding
  :star:code
• Spatial and Visual Perspective-Taking via View Rotation and Relation Reasoning for Embodied Reference Understanding
  :star:code
• Switch-BERT: Learning to Model Multimodal Interactions by Switching Attention and Input
• Fine-Grained Visual Entailment
  :star:code
• FindIt: Generalized Localization with Natural Language Queries
  :house:project
• Decomposing the Tangent of Occluding Boundaries according to Curvatures and Torsions
• Real-Time Neural Character Rendering with Pose-Guided Multiplane Images
• TensoRF: Tensorial Radiance Fields
  :house:project
• TAVA: Template-Free Animatable Volumetric Actors
  :star:code
• Relative Pose from SIFT Features
  :star:code
• Solution Space Analysis of Essential Matrix Based on Algebraic Error Minimization
• CoVisPose: Co-Visibility Pose Transformer for Wide-Baseline Relative Pose Estimation in 360° Indoor Panoramas
• Space-Partitioning RANSAC
  :star:code
• Correspondence Reweighted Translation Averaging
• Beyond Periodicity: Towards a Unifying Framework for Activations in Coordinate-MLPs
  :star:code
• GigaDepth: Learning Depth from Structured Light with Branching Neural Networks
• Visual Prompt Tuning
  :star:code
• Cross-Modal Knowledge Transfer without Task-Relevant Source Data
• PACTran: PAC-Bayesian Metrics for Estimating the Transferability of Pretrained Models to Classification Tasks
• CYBORGS: Contrastively Bootstrapping Object Representations by Grounding in Segmentation
  :star:code
• SALVe: Semantic Alignment Verification for Floorplan Reconstruction from Sparse Panoramas
• MVP: Multimodality-Guided Visual Pre-training
• Self-Filtering: A Noise-Aware Sample Selection for Label Noise with Confidence Penalization
• Learning to Learn with Smooth Regularization
  :star:code
• Ensemble Learning Priors Driven Deep Unfolding for Scalable Video Snapshot Compressive Imaging
  :star:code
• Approximate Discrete Optimal Transport Plan with Auxiliary Measure Method
• A Comparative Study of Graph Matching Algorithms in Computer Vision
• Semidefinite Relaxations of Truncated Least-Squares in Robust Rotation Search: Tight or Not
• Dynamic Metric Learning with Cross-Level Concept Distillation
  :star:code
• MENet: A Memory-Based Network with Dual-Branch for Efficient Event Stream Processing
• Improving Robustness by Enhancing Weak Subnets
• Learning from Multiple Annotator Noisy Labels via Sample-Wise Label Fusion
  :star:code
• Unbiased Manifold Augmentation for Coarse Class Subdivision
  :star:code
• OccamNets: Mitigating Dataset Bias by Favoring Simpler Hypotheses
  :star:code
• ERA: Enhanced Rational Activations
  :star:code
• Active Label Correction Using Robust Parameter Update and Entropy Propagation
• Revisiting Batch Norm Initialization
  :star:code
• Differentiable Rendering for Synthetic Aperture Radar Imagery
• Batch-efficient EigenDecomposition for Small and Medium Matrices
  :star:code
• Accelerating Score-based Generative Models with Preconditioned Diffusion Sampling
• Improving Covariance Conditioning of the SVD Meta-layer by Orthogonality
  :star:code
• Contrastive Deep Supervision
  :star:code
• Organic Priors in Non-Rigid Structure from Motion
  :open_mouth:oral
• Bootstrapped Masked Autoencoders for Vision BERT Pretraining
  :star:code
• Lipschitz Continuity Retained Binary Neural Network
  :star:code
• NeFSAC: Neurally Filtered Minimal Samples
  :star:code
• Towards Understanding The Semidefinite Relaxations of Truncated Least-Squares in Robust Rotation Search
• Latency-Aware Collaborative Perception
  :star:code
• MHR-Net: Multiple-Hypothesis Reconstruction of Non-Rigid Shapes from 2D Views
• SelectionConv: Convolutional Neural Networks for Non-rectilinear Image Data
• Overcoming Shortcut Learning in a Target Domain by Generalizing Basic Visual Factors from a Source Domain
  :star:code
• Discrete-Constrained Regression for Local Counting Models
• Streamable Neural Fields
  :star:code
• Contributions of Shape, Texture, and Color in Visual Recognition
  :star:code
• Single Frame Atmospheric Turbulence Mitigation: A Benchmark Study and A New Physics-Inspired Transformer Model
• Latent Discriminant deterministic Uncertainty
  :star:code
• SPIN: An Empirical Evaluation on Sharing Parameters of Isotropic Networks
  :star:code
• UFO: Unified Feature Optimization
  :star:code
• POP: Mining POtential Performance of new fashion products via webly cross-modal query expansion
  :star:code
• My View is the Best View: Procedure Learning from Egocentric Videos
  :star:code:house:project
• Equivariance and Invariance Inductive Bias for Learning from Insufficient Data
  :star:code
• Contrastive Monotonic Pixel-Level Modulation
  :open_mouth:oral:star:code
• Neural-Sim: Learning to Generate Training Data with NeRF
  :star:code
• Learning Hierarchy Aware Features for Reducing Mistake Severity
  :star:code
• Translating a Visual LEGO Manual to a Machine-Executable Plan
  :star:code:house:project
• Hardly Perceptible Trojan Attack against Neural Networks with Bit Flips
  :star:code
• LGV: Boosting Adversarial Example Transferability from Large Geometric Vicinity
• MonteBoxFinder: Detecting and Filtering Primitives to Fit a Noisy Point Cloud
  :star:code:house:project
• Neural Strands: Learning Hair Geometry and Appearance from Multi-View Images
  :house:project
• A Repulsive Force Unit for Garment Collision Handling in Neural Networks
  :house:project
• Minimal Neural Atlas: Parameterizing Complex Surfaces with Minimal Charts and Distortion
  :star:code
• Revisiting the Critical Factors of Augmentation-Invariant Representation Learning
  :star:code
• Fast Two-step Blind Optical Aberration Correction
  :star:code
• Transformers as Meta-Learners for Implicit Neural Representations
  :star:code:house:project
• Neighborhood Collective Estimation for Noisy Label Identification and Correction
  :star:code
• Rethinking Robust Representation Learning Under Fine-grained Noisy Faces
• Contrast-Phys: Unsupervised Video-based Remote Physiological Measurement via Spatiotemporal Contrast
  :star:code
• RelPose: Predicting Probabilistic Relative Rotation for Single Objects in the Wild
  :house:project
• PRIF: Primary Ray-based Implicit Function
  :house:project
• Context-Aware Streaming Perception in Dynamic Environments
  :star:code
• AdaBin: Improving Binary Neural Networks with Adaptive Binary Sets
• TRoVE: Transforming Road Scene Datasets into Photorealistic Virtual Environments
  :star:code
• L3: Accelerator-Friendly Lossless Image Format for High-Resolution, High-Throughput DNN Training
  :star:code
• GCISG: Guided Causal Invariant Learning for Improved Syn-to-real Generalization
• Learning Continuous Implicit Representation for Near-Periodic Patterns
  :star:code:house:project
• A Deep Moving-camera Background Model
  :star:code
• NashAE: Disentangling Representations through Adversarial Covariance Minimization
  :star:code
• FusionVAE: A Deep Hierarchical Variational Autoencoder for RGB Image Fusion
• Diversified Dynamic Routing for Vision Tasks
• Fast-ParC: Position Aware Global Kernel for ConvNets and ViTs
• Improving the Reliability for Confidence Estimation
• Attaining Class-level Forgetting in Pretrained Model using Few Samples
• Overexposure Mask Fusion: Generalizable Reverse ISP Multi-Step Refinement
• Photo-realistic Neural Domain Randomization
• Editable indoor lighting estimation
  :house:project
• A Kendall Shape Space Approach to 3D Shape Estimation from 2D Landmarks
• DeepShadow: Neural Shape from Shadow
  :star:code:house:project
• Intrinsic Neural Fields: Learning Functions on Manifolds
• Unsupervised Pose-Aware Part Decomposition for Man-Made Articulated Objects
• MeshUDF: Fast and Differentiable Meshing of Unsigned Distance Field Networks
• S2N: Suppression-Strengthen Network for Event-Based Recognition under Variant Illuminations
  :star:code
• A Spectral View of Randomized Smoothing under Common Corruptions: Benchmarking and Improving Certified Robustness
• Transform Your Smartphone into a DSLR Camera: Learning the ISP in the Wild
  :star:code
• Data Association between Event Streams and Intensity Frames under Diverse Baselines
• Instance Contour Adjustment via Structure-Driven CNN
• 3D Scene Inference from Transient Histograms
• Neural Space-Filling Curves
  :house:project
• LWGNet – Learned Wirtinger Gradients for Fourier Ptychographic Phase Retrieval
  :star:code
• PANDORA: Polarization-Aided Neural Decomposition of Radiance
• Benchmarking Omni-Vision Representation through the Lens of Visual Realms
  :house:project
• When Deep Classifiers Agree: Analyzing Correlations between Learning Order and Image Statistics
• MUGEN: A Playground for Video-Audio-Text Multimodal Understanding and GENeration
  :house:project
• The Missing Link: Finding Label Relations across Datasets
• Domain Adaptive Hand Keypoint and Pixel Localization in the Wild
  :house:project
• DFNet: Enhance Absolute Pose Regression with Direct Feature Matching
  :house:project
• GTCaR: Graph Transformer for Camera Re-Localization
• Is Geometry Enough for Matching in Visual Localization?
  :star:code
• Reducing Information Loss for Spiking Neural Networks
• Deep Partial Updating: Towards Communication Efficient Updating for On-Device Inference
• SP-Net: Slowly Progressing Dynamic Inference Networks
• Meta-GF: Training Dynamic-Depth Neural Networks Harmoniously
  :star:code
• You Already Have It: A Generator-Free Low-Precision DNN Training Framework Using Stochastic Rounding
• Real Spike: Learning Real-Valued Spikes for Spiking Neural Networks
• Exploring Lottery Ticket Hypothesis in Spiking Neural Networks
  :star:code
• On the Angular Update and Hyperparameter Tuning of a Scale-Invariant Network
• LANA: Latency Aware Network Acceleration
• Understanding the Dynamics of DNNs Using Graph Modularity
  :star:code
• MIME: Minority Inclusion for Majority Group Enhancement of AI Performance
  :house:project
• Trust, but Verify: Using Self-Supervised Probing to Improve Trustworthiness
  :star:code
• Learning to Censor by Noisy Sampling
• Anti-Neuron Watermarking: Protecting Personal Data against Unauthorized Neural Networks
• Recover Fair Deep Classification Models via Altering Pre-trained Structure
• Decouple-and-Sample: Protecting Sensitive Information in Task Agnostic Data Release
  :star:code
• Latent Space Smoothing for Individually Fair Representations
• Parameterized Temperature Scaling for Boosting the Expressive Power in Post-Hoc Uncertainty Calibration
  :star:code
• Image-Based CLIP-Guided Essence Transfer
  :star:code
• End-to-End Visual Editing with a Generatively Pre-trained Artist
  :house:project
• Sobolev Training for Implicit Neural Representations with Approximated Image Derivatives
  :star:code
• L-Tracing: Fast Light Visibility Estimation on Neural Surfaces by Sphere Tracing
• Temporal-MPI: Enabling Multi-Plane Images for Dynamic Scene Modelling via Temporal Basis Learning
• 3D-Aware Semantic-Guided Generative Model for Human Synthesis
  :house:project
• Unified Implicit Neural Stylization
  :house:project
• Deep Portrait Delighting
  :house:project
• Free-Viewpoint RGB-D Human Performance Capture and Rendering
  :house:project
• Multiview Regenerative Morphing with Dual Flows
  :star:code
• NeRF for Outdoor Scene Relighting
  :house:project
• Intelli-Paint: Towards Developing More Human-Intelligible Painting Agents
• Motion Transformer for Unsupervised Image Animation
• NÜWA: Visual Synthesis Pre-training for Neural visUal World creAtion
• Implicit Neural Representations for Variable Length Human Motion Generation
  :star:code
• Learning Object Placement via Dual-Path Graph Completion
• Compositional Visual Generation with Composable Diffusion Models
  :house:project
• Spatial-Frequency Domain Information Integration for Pan-Sharpening
• ReCoNet: Recurrent Correction Network for Fast and Efficient Multi-Modality Image Fusion
  :star:code
• Rethinking Generic Camera Models for Deep Single Image Camera Calibration to Recover Rotation and Fisheye Distortion
• Modeling Mask Uncertainty in Hyperspectral Image Reconstruction
  :star:code
• Deep Fourier-Based Exposure Correction Network with Spatial-Frequency Interaction
  :star:code
• Towards Real-World HDRTV Reconstruction: A Data Synthesis-Based Approach
• Attention-Aware Learning for Hyperparameter Prediction in Image Processing Pipelines
• Memory-Augmented Model-Driven Network for Pansharpening
  :star:code
• All You Need Is RAW: Defending against Adversarial Attacks with Camera Image Pipelines
• GRIT-VLP: Grouped Mini-Batch Sampling for Efficient Vision and Language Pre-training
  :star:code
• Transformer with Implicit Edges for Particle-Based Physics Simulation
  :star:code
• LA3: Efficient Label-Aware AutoAugment
• BA-Net: Bridge Attention for Deep Convolutional Neural Networks
  :star:code
• SAU: Smooth Activation Function Using Convolution with Approximate Identities
• Almost-Orthogonal Layers for Efficient General-Purpose Lipschitz Networks
  :star:code
• DLME: Deep Local-Flatness Manifold Embedding
• Accurate Detection of Proteins in Cryo-Electron Tomograms from Sparse Labels
• Social ODE: Multi-agent Trajectory Forecasting with Neural Ordinary Differential Equations
• Entropy-Driven Sampling and Training Scheme for Conditional Diffusion Generation
  :star:code
• Geometry-Guided Progressive NeRF for Generalizable and Efficient Neural Human Rendering
• Controllable Shadow Generation Using Pixel Height Maps
• Subspace Diffusion Generative Models
  :star:code
• MINER: Multiscale Implicit Neural Representation
  :house:project
• An Embedded Feature Whitening Approach to Deep Neural Network Optimization
  :star:code
• Q-FW: A Hybrid Classical-Quantum Frank-Wolfe for Quadratic Binary Optimization
• Scalable Learning to Optimize: A Learned Optimizer Can Train Big Models
  :star:code
• QISTA-ImageNet: A Deep Compressive Image Sensing Framework Solving ℓq-Norm Optimization Problem
• Rethinking Confidence Calibration for Failure Prediction
  :star:code
• PRIME: A Few Primitives Can Boost Robustness to Common Corruptions
  :star:code
• Learning with Noisy Labels by Efficient Transition Matrix Estimation to Combat Label Miscorrection
• Learning to Drive by Watching YouTube Videos: Action-Conditioned Contrastive Policy Pretraining
  :house:project
• Balancing between Forgetting and Acquisition in Incremental Subpopulation Learning
  :star:code
• Sound Localization by Self-Supervised Time Delay Estimation
  :house:project
• X-Learner: Learning Cross Sources and Tasks for Universal Visual Representation
• A Contrastive Objective for Learning Disentangled Representations
  :star:code
• A Gyrovector Space Approach for Symmetric Positive Semi-Definite Matrix Learning
• Trading Positional Complexity vs Deepness in Coordinate Networks
  :house:project
• TO-Scene: A Large-Scale Dataset for Understanding 3D Tabletop Scenes
  :star:code
• Primitive-Based Shape Abstraction via Nonparametric Bayesian Inference
• S2Net: Stochastic Sequential Pointcloud Forecasting
• LaLaLoc++: Global Floor Plan Comprehension for Layout Localisation in Unvisited Environments
• Variance-Aware Weight Initialization for Point Convolutional Neural Networks
• AdaAfford: Learning to Adapt Manipulation Affordance for 3D Articulated Objects via Few-Shot Interactions
  :house:project
• human relighting
  • Geometry-aware Single-image Full-body Human Relighting
• 奇异值检测(Novelty Detection)
  • Semantic Novelty Detection via Relational Reasoning
    :star:code
• Multi-attribute Learning
  • Label2Label: A Language Modeling Framework for Multi-Attribute Learning
    :star:code
• 偏见识别
  • Discover and Mitigate Unknown Biases with Debiasing Alternate Networks
    :star:code
• 新类别发现(Novel Class Discovery)
  • Novel Class Discovery without Forgetting
• 密集预测
  • FADE: Fusing the Assets of Decoder and Encoder for Task-Agnostic Upsampling
    :star:code
• 变分自动编码器(VAEs)
  • Continual Variational Autoencoder Learning via Online Cooperative Memorization
    :star:code
• 开集识别
  • Towards Accurate Open-Set Recognition via Background-Class Regularization
• 草图
  • Abstracting Sketches through Simple Primitives
    :star:code
  • FS-COCO: Towards Understanding of Freehand Sketches of Common Objects in Context
    :house:project
  • DoodleFormer: Creative Sketch Drawing with Transformers
    :star:code
• 聚类
  • A Data-Centric Approach for Improving Ambiguous Labels with Combined Semi-Supervised Classification and Clustering
    :star:code
• Visual Grounding
  • SiRi: A Simple Selective Retraining Mechanism for Transformer-based Visual Grounding
    :star:code
• 互动结构理解
  • Break and Make: Interactive Structural Understanding Using LEGO Bricks
    :star:code
• HDR全景图生成
  • StyleLight: HDR Panorama Generation for Lighting Estimation and Editing
    :star:code:house:project
• 手语识别
  • Automatic dense annotation of large-vocabulary sign language videos
    :house:project
  • Deep Radial Embedding for Visual Sequence Learning
  • Temporal Lift Pooling for Continuous Sign Language Recognition
    :star:code
• 读唇术
  • Speaker-adaptive Lip Reading with User-dependent Padding
• BNN
  • Recurrent Bilinear Optimization for Binary Neural Networks
    :open_mouth:oral:star:code
  • Towards Accurate Binary Neural Networks via Modeling Contextual Dependencies
    :star:code
• 图像取证
  • Totems: Physical Objects for Verifying Visual Integrity
    :house:project
• 图像对齐
  • Learning Depth from Focus in the Wild
    :star:code
• visual hand pressure estimation
  • PressureVision: Estimating Hand Pressure from a Single RGB Image
    :star:code
• 光亮估计
  • Estimating Spatially-Varying Lighting in Urban Scenes with Disentangled Representation
• 室内场景照明编辑
  • Physically-Based Editing of Indoor Scene Lighting from a Single Image
• HDR
  • Exposure-Aware Dynamic Weighted Learning for Single-Shot HDR Imaging
    :star:code
• 关键点定位
  • Multi-Domain Multi-Definition Landmark Localization for Small Datasets
• XAI
  • STEEX: Steering Counterfactual Explanations with Semantics
    :star:code
  • Making Heads or Tails: Towards Semantically Consistent Visual Counterfactuals
    :star:code
  • HIVE: Evaluating the Human Interpretability of Visual Explanations
    :house:project
  • Shap-CAM: Visual Explanations for Convolutional Neural Networks Based on Shapley Value
• 掌纹识别
  • BézierPalm: A Free Lunch for Palmprint Recognition
    :house:project
• 视线估计
  • Look Both Ways: Self-Supervising Driver Gaze Estimation and Road Scene Saliency
    :open_mouth:oral:star:code
• 运动迁移
  • Motion and Appearance Adaptation for Cross-Domain Motion Transfer
• 远程呼吸监测
  • Remote Respiration Monitoring of Moving Person Using Radio Signals
• 图像-图形生成
  • Relationformer: A Unified Framework for Image-to-Graph Generation
    :star:code

扫码CV君微信（注明：CVPR）入微信交流群：