官网链接：https://wacv2023.thecvf.com/home

会议日期：2023年1月3日-1月7日

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

2023 年论文分类汇总戳这里

↘️CVPR-2023-Papers ↘️WACV-2023-Papers

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

:exclamation::exclamation::exclamation::star2::star2::star2:WACV 2023收录论文已全部公布，下载可在【我爱计算机视觉】后台回复“paper”，即可收到。共计 638 篇。

目录

:dog:	:mouse:	:hamster:	:tiger:
65.Open Set Recognition(开集识别)	66.Scene Flow Estimation(场景流估计)	67.Sketches(草图识别)
61.geo-localization(城市地理定位)	62.Dense Prediction(密集预测)	63.Place Recognition(位置识别)	64.Visual Odometry(视觉里程计)
57.Federated Learning(联邦学习)	58.HOI(人物交互)	59.Meta learning(元学习)	60.Image-to-Image Translation(图像-图像翻译)
53.Gaze Estimation(视线估计)	54.Optical Flow(光流)	55.Clustering(聚类)	56.Vision-Language(视觉语言)
49.Neural Radiance(渲染)	50.Contrastive Learning(对比学习)	51.SGG(场景图生成)	52.Human Motion Prediction(人类运动预测)
45.Class-Incremental Learning(类增量学习)	46.Metric Learning(度量学习)	47.Data Augmentation(数据增强)	48.Light Fields(光场)
41.Action Generation(动作生成)	42.Landmark Detection(关键点检测)	43.Active Learning(主动学习)	44.Multi-Task Learning(多任务学习)
37.OT(目标跟踪)	38.Sound(音频处理)	39.Style Transfer(风格迁移)	40.AD(异常检测)
33.View Synthesis(视图合成)	34.SLAM\Robots	35.VQA(视觉问答)	36.Soft Biometrics(软生物技术)
29.Image Classification(图像分类)	30.RL(强化学习)	31.Deepfake Detection(假象检测)	32.Continual Learning(持续学习)
25.Image Captioning(图像字幕)	26.Dataset(数据集)	27.Person ReID(人员重识别)	28.OPE(物体姿态估计)
21.PC(点云)	22.HAR(人体动作识别与检测)	23.AD(智能驾驶)	24.Image Retrieval(图像检索)
17.OCR(文本检测)	18.NAS(神经架构搜索)	19.MC\KD\Pruning(模型压缩\知识蒸馏\剪枝)	20.Transformer
13.Image Segmentation(图像分割)	14.SSL(半监督学习)	15.Image Synthesis(图像合成)	16.SR(超分辨率)
9.RS\Satellite Image(遥感\卫星图像)	10.AL(对抗学习)	11.Face(人脸)	12.FSL or DA\G(小样本学习 or 域适应\泛化)
5.OD(目标检测)	6.Video(视频相关)	7.Pose(人体姿态)	8.Image Processing(图像处理)
1.其它	2.Medical Image(医学影像)	3.3D(三维视觉)	4.GAN(生成对抗网络)

67.Sketches(草图识别)

• WHFL: Wavelet-Domain High Frequency Loss for Sketch-to-Image Translation

66.Scene Flow Estimation(场景流估计)

• M-FUSE: Multi-frame Fusion for Scene Flow Estimation
  :star:code

65.Open Set Recognition(开集识别)

• Ancestor Search: Generalized Open Set Recognition via Hyperbolic Side Information Learning

64.Visual Odometry(视觉里程计)

• Pixel-Wise Prediction Based Visual Odometry via Uncertainty Estimation

63.Place Recognition(位置识别)

• ETR: An Efficient Transformer for Re-ranking in Visual Place Recognition
• MixVPR: Feature Mixing for Visual Place Recognition

62.Dense Prediction(密集预测)

• Dense Prediction With Attentive Feature Aggregation

61.geo-localization(城市地理定位)

• TransVLAD: Multi-Scale Attention-Based Global Descriptors for Visual Geo-Localization
  :star:code

60.Image-to-Image Translation(图像-图像翻译)

• Panoptic-Aware Image-to-Image Translation
• 图像翻译
  • RIFT: Disentangled Unsupervised Image Translation via Restricted Information Flow
• 域到域翻译
  • Learning Style Subspaces for Controllable Unpaired Domain Translation
    :star:code

59.Meta learning(元学习)

• Meta-OLE: Meta-learned Orthogonal Low-Rank Embedding

58.Human Object Interaction(人物交互)

• Skew-Robust Human-Object Interactions in Videos
• 手物交互
  • Fine-grained Affordance Annotation for Egocentric Hand-Object Interaction Videos

57.Federated Learning(联邦学习)

• Learning Across Domains and Devices: Style-Driven Source-Free Domain Adaptation in Clustered Federated Learning
  :star:code
• Federated Learning for Commercial Image Sources

56.Vision-Language(视觉语言)

• VL-Taboo: An Analysis of Attribute-based Zero-shot Capabilities of Vision-Language Models
  :star:code
• Learning by Hallucinating: Vision-Language Pre-training with Weak Supervision
• Perceiver-VL: Efficient Vision-and-Language Modeling with Iterative Latent Attention
  :star:code
• GAFNet: A Global Fourier Self Attention Based Novel Network for multi-modal ownstream tasks
• VLN
  • Structure-Encoding Auxiliary Tasks for Improved Visual Representation in Vision-and-Language Navigatio
  • A Priority Map for Vision-and-Language Navigation with Trajectory Plans and Feature-Location Cues
    :star:code

55.Clustering(聚类)

• Self-Supervised Clustering based on Manifold Learning and Graph Convolutional Networks
  :star:code

54.Optical Flow(光流)

• Weakly-Supervised Optical Flow Estimation for Time-of-Flight
• Rebalancing Gradient To Improve Self-Supervised Co-Training of Depth, Odometry and Optical Flow Predictions
  :star:code
• DCVNet: Dilated Cost Volume Networks for Fast Optical Flow
• MFCFlow : A Motion Feature Compensated Multi-Frame Recurrent Network for Optical Flow Estimation
• BrightFlow: Brightness-Change-Aware Unsupervised Learning of Optical Flow
  :star:code
• Towards Equivariant Optical Flow Estimation with Deep Learning(https://github.com/stsavian/equivariant_of_estimation)
• Learning Lightweight Neural Networks via Channel-Split Recurrent Convolution
• Meta-Learning for Adaptation of Deep Optical Flow Networks

53.Gaze Estimation(视线估计)

• Searching Efficient Neural Architecture with Multi-resolution Fusion Transformer for Appearance-based Gaze Estimation
• iris localization(虹膜定位)
  • Segmentation-free Direct Iris Localization Networks
• 视线跟随
  • Patch-level Gaze Distribution Prediction for Gaze Following
• 视线重定向
  • Fine Gaze Redirection Learning with Gaze Hardness-aware Transformation
    :star:code
  • CUDA-GHR: Controllable Unsupervised Domain Adaptation for Gaze and Head Redirection
    :star:code

52.Human Motion Prediction(人类运动预测)

• Multi-view Tracking Using Weakly Supervised Human Motion Prediction
  :star:code
• Anticipative Feature Fusion Transformer for Multi-Modal Action Anticipation
  :star:code
• GliTr: Glimpse Transformers with Spatiotemporal Consistency for Online Action Prediction
• Back to MLP: A Simple Baseline for Human Motion Prediction
  :star:code
• Intention-Conditioned Long-Term Human Egocentric Action Anticipation
• 行人轨迹预测
  • Online Adaptive Temporal Memory with Certainty Estimation for Human Trajectory Prediction

51.Scene Graph Generation(场景图生成)

• Grounding Scene Graphs on Natural Images via Visio-Lingual Message Passing
  :star:code:house:project
• Improving Predicate Representation in Scene Graph Generation by Self-Supervised Learning
• More Knowledge, Less Bias: Unbiasing Scene Graph Generation with Explicit Ontological Adjustment
  :star:code
• Composite Relationship Fields with Transformers for Scene Graph Generation
  :star:code

50.Contrastive Learning(对比学习)

• Similarity Contrastive Estimation for Self-Supervised Soft Contrastive Learning
  :star:code
• Representation Disentanglement in Generative Models with Contrastive Learning
• Addressing Feature Suppression in Unsupervised Visual Representations
• Ego-Vehicle Action Recognition based on Semi-Supervised Contrastive Learning

49.Neural Radiance(渲染)

• Ev-NeRF: Event Based Neural Radiance Field
• DDNeRF: Depth Distribution Neural Radiance Fields
• X-NeRF: Explicit Neural Radiance Field for Multi-Scene 360deg Insufficient RGB-D Views
  :star:code
• Fast Differentiable Transient Rendering for Non-Line-of-Sight Reconstruction
  :star:code
• Compressing Explicit Voxel Grid Representations: fast NeRFs become also small
• Control-NeRF: Editable Feature Volumes for Scene Rendering and Manipulation
  :house:project
• Beyond RGB: Scene-Property Synthesis with Neural Radiance Fields
  :star:code

48.Light Fields(光场)

• 光场
  • I See-Through You: A Framework for Removing Foreground Occlusion in Both Sparse and Dense Light Field Images
• 相机
  • Event-based RGB sensing with structured light
    :star:code
  • Burst Vision Using Single-Photon Cameras
  • TVCalib: Camera Calibration for Sports Field Registration in Soccer
    :house:project
• 兴趣点检测
  • EventPoint: Self-Supervised Interest Point Detection and Description for Event-Based Camera

47.Data Augmentation(数据增强)

• Rethinking Rotation in Self-Supervised Contrastive Learning: Adaptive Positive or Negative Data Augmentation
  :star:code

46.Metric Learning(度量学习)

• InDiReCT: Language-Guided Zero-Shot Deep Metric Learning for Images
  :star:code

45.Class-Incremental Learning(类增量学习)

• AdvisIL - A Class-Incremental Learning Advisor
  :star:code
• FeTrIL: Feature Translation for Exemplar-Free Class-Incremental Learning
  :star:code
• 增量学习
  • Neural Weight Search for Scalable Task Incremental Learning
    :star:code

44.Multi-Task Learning(多任务学习)

• Cross-task Attention Mechanism for Dense Multi-task Learning
  :star:code

43.Active Learning(主动学习)

• Randomness is the Root of All Evil:More Reliable Evaluation of Deep Active Learning
  :house:project

42.Landmark Detection(关键点检测)

• CoKe: Contrastive Learning for Robust Keypoint Detection

41.Action Generation(动作生成)

• 全身运动合成
  • DSAG: A Scalable Deep Framework for Action-Conditioned Multi-Actor Full Body Motion Synthesis

40.Anomaly Detection(异常检测)

• Asymmetric Student-Teacher Networks for Industrial Anomaly Detection
  :star:code
• Zero-Shot Versus Many-Shot: Unsupervised Texture Anomaly Detection
  :star:code
• No Shifted Augmentations (NSA): compact distributions for robust self-supervised Anomaly Detection
• GLAD: A Global-to-Local Anomaly Detector
• 道路异常检测
  • Image-Consistent Detection of Road Anomalies as Unpredictable Patches
• 异常聚类
  • Anomaly Clustering: Grouping Images into Coherent Clusters of Anomaly Types

39.Style Transfer(风格迁移)

• Line Search-Based Feature Transformation for Fast, Stable, and Tunable Content-Style Control in Photorealistic Style Transfer
  :star:code
• RAST: Restorable Arbitrary Style Transfer via Multi-Restoration
• Dance Style Transfer with Cross-modal Transformer
  :tv:video
• Is Bigger Always Better? An Empirical Study on Efficient Architectures for Style Transfer and Beyond

38.Sound(音频处理)

• AudioViewer: Learning to Visualize Sounds
  :house:project
• Audio Visual Event Localization视听事件定位
  • AVE-CLIP: AudioCLIP-based Multi-window Temporal Transformer for Audio Visual Event Localization
• 音频去噪
  • BirdSoundsDenoising: Deep Visual Audio Denoising for Bird Sounds
• 视听分割
  • Unsupervised Audio-Visual Lecture Segmentation
    :house:project
• 生源定位
  • Hear The Flow: Optical Flow-Based Self-Supervised Visual Sound Source Localization
    :star:code
  • Exploiting Visual Context Semantics for Sound Source Localization
• 语音识别
  • Audio-Visual Efficient Conformer for Robust Speech Recognition
    :star:code
• 音频分离
  • SeCo: Separating Unknown Musical Visual Sounds with Consistency Guidance

37.Object Tracking(目标跟踪)

• Efficient Visual Tracking with Exemplar Transformers
  :star:code
• Hard to Track Objects with Irregular Motions and Similar Appearances?Make It Easier by Buffering the Matching Space
• HOOT: Heavy Occlusions in Object Tracking Benchmark
• VirtualHome Action Genome: A Simulated Spatio-Temporal Scene Graph Dataset With Consistent Relationship Labels
• Tracking Growth and Decay of Plant Roots in Minirhizotron Images
  :star:code
• Planar Object Tracking via Weighted Optical Flow
• Multi-Frame Attention with Feature-Level Warping for Drone Crowd Tracking
• 多目标跟踪
  • AttTrack: Online Deep Attention Transfer for Multi-object Tracking
  • Detection Recovery in Online Multi-Object Tracking With Sparse Graph Tracker
    :star:code
  • MMPTRACK: Large-scale Densely Annotated Multi-camera Multiple People Tracking Benchmark
  • TransMOT: Spatial-Temporal Graph Transformer for Multiple Object Tracking

36.Soft Biometrics(软生物技术)

• 手指静脉识别
  • Analysis of Master Vein Attacks on Finger Vein Recognition Systems
• 隐形眼镜虹膜PAD算法的错误分类
  • Misclassifications of Contact Lens Iris PAD Algorithms: Is it Gender Bias or Environmental Conditions
• 生物信息识别
  • Can Shadows Reveal Biometric Information?
• 虹膜
  • DeformIrisNet: An Identity-Preserving Model of Iris Texture Deformation
    :star:code

35.VQA(视觉问答)

• DRAMA: Joint Risk Localization and Captioning in Driving
• VLC-BERT: Visual Question Answering with Contextualized Commonsense Knowledge
  :star:code
• Barlow constrained optimization for Visual Question Answering
  :star:code
• How To Practice VQA on a Resource-Limited Target Domain
  :house:project
• Guiding Visual Question Answering With Attention Priors
• VideoQA
  • Dense but Efficient VideoQA for Intricate Compositional Reasoning
  • Text-Guided Object Detector for Multi-modal Video Question Answering
  • Watching the News: Towards VideoQA Models that can Read
    :house:project
• 视觉问题生成
  • K-VQG: Knowledge-Aware Visual Question Generation for Common-Sense Acquisition
    :house:project

34.SLAM\Robots

• SLAM
  • Probabilistic Volumetric Fusion for Dense Monocular SLAM
• AR
  • Heightfields for Efficient Scene Reconstruction for AR
    :star:code

33.View Synthesis(视图合成)

• Vision Transformer for NeRF-Based View Synthesis From a Single Input Image
• Self-improving Multiplane-to-layer Images for Novel View Synthesis
  :house:project

32.Continual Learning(持续学习)

• Continual Learning with Dependency Preserving Hypernetworks
• Do Pre-trained Models Benefit Equally in Continual Learning
  :star:code
• Saliency Guided Experience Packing for Replay in Continual Learning

31.Deepfake Detection(假象检测)

• TI2Net: Temporal Identity Inconsistency Network for Deepfake Detection
• 图像伪造
  • CFL-Net: Image Forgery Localization Using Contrastive Learning
    :star:code

30.Reinforcement Learning(强化学习)

• Switching to Discriminative Image Captioning by Relieving a Bottleneck of Reinforcement Learning
  :star:code

29.Image Classification(图像分类)

• Wavelength-Aware 2D Convolutions for Hyperspectral Imaging
  :star:code
• ML-Decoder: Scalable and Versatile Classification Head
• CNN2Graph: Building Graphs for Image Classification
• Token Pooling in Vision Transformers for Image Classification
• Augmentation by Counterfactual Explanation -Fixing an Overconfident Classifier
• Treatment Learning Causal Transformer for Noisy Image Classification
  :star:code
• 长尾识别
  • Difficulty-Net: Learning to Predict Difficulty for Long-Tailed Recognition
  • Mutual Learning for Long-Tailed Recognition
• pen-Set Classification
  • Large-Scale Open-Set Classification Protocols for ImageNet
• 细粒度分类
  • SSFE-Net: Self-Supervised Feature Enhancement for Ultra-Fine-Grained Few-Shot Class Incremental Learning
• 多标签分类
  • Spatial Consistency Loss for Training Multi-Label Classifiers From Single-Label Annotations
• 小样本分类
  • Contrastive Knowledge-Augmented Meta-Learning for Few-Shot Classification
    :star:code
  • CORL: Compositional Representation Learning for Few-Shot Classification

28.Pose Estimation(姿态估计)

• COPE: End-to-end trainable Constant Runtime Object Pose Estimation
• 6D
  • CRT-6D: Fast 6D Object Pose Estimation with Cascaded Refinement Transformers
    :star:code
  • SD-Pose: Structural Discrepancy Aware Category-Level 6D Object Pose Estimation
• 物体计数
  • Few-shot Object Counting with Similarity-Aware Feature Enhancement
    :star:code
• 物体重识别
  • Bent & Broken Bicycles: Leveraging synthetic data for damaged object re-identification
    :house:project

27.Person ReID(人员重识别)

• 行人分析
  • Towards A Framework for Privacy-Preserving Pedestrian Analysis
• 行人搜索
  • Gallery Filter Network for Person Search
    :star:code
  • UPAR: Unified Pedestrian Attribute Recognition and Person Retrieval
    :star:code
  • SAT: Scale-Augmented Transformer for Person Search
• Re-id
  • Camera Alignment and Weighted Contrastive Learning for Domain Adaptation in Video Person ReID
    :star:code
  • MEVID: Multi-view Extended Videos with Identities for Video Person Re-Identification
    :star:code
  • Feature Disentanglement Learning with Switching and Aggregation for Video-based Person Re-Identification
  • Graph-Based Self-Learning for Robust Person Re-Identification
  • Body Part-Based Representation Learning for Occluded Person Re-Identification
    :star:code
• 步态识别
  • Gait Recognition Using 3-D Human Body Shape Inference
• 步态迁移
  • CTrGAN: Cycle Transformers GAN for Gait Transfer
• 嫌疑人识别
  • A Suspect Identification Framework using Contrastive Relevance Feedback
• 人群计数
  • Dynamic Mixture of Counter Network for Location-Agnostic Crowd Counting

26.Dataset\Benchmark(数据集\基准)

• OpenEarthMap: A Benchmark Dataset for Global High-Resolution Land Cover Mapping
  :sunflower:dataset
• A Continual Deepfake Detection Benchmark: Dataset, Methods, and Essentials
  :sunflower:dataset
• The CropAndWeed Dataset: a Multi-Modal Learning Approach for Efficient Crop and Weed Manipulation
  :sunflower:dataset
• IDD-3D: A Dataset for Driving in Unstructured Road Scenes
  :sunflower:dataset
• Vis2Rec: A Large-Scale Visual Dataset for Visit Recommendation
  :sunflower:dataset
• Creating a Forensic Database of Shoeprints from Online Shoe-Tread Photos
  :sunflower:dataset
• 目标检测、分割、跟踪
  • BURST: A Benchmark for Unifying Object Recognition, Segmentation and Tracking in Video
    :star:code

25.Image Captioning(图像字幕)

• 人体图像分析
  • Split To Learn: Gradient Split for Multi-Task Human Image Analysis
• 图像字幕
  • Expert-defined Keywords Improve Interpretability of Retinal Image Captioning
    :star:code
• 视频字幕
  • Lightweight Video Denoising Using Aggregated Shifted Window Attention
    :star:code

24.Image Retrieval(图像检索)

• Boosting vision transformers for image retrieval
  :star:code
• Certified Defense for Content Based Image Retrieval
• Fashion Image Retrieval with Text Feedback by Additive Attention Compositional Learning
• Content-Based Music-Image Retrieval Using Self- and Cross-Modal Feature Embedding Memory
• 图像-句子检索
  • Cross-modal Semantic Enhanced Interaction for Image-Sentence Retrieval
• 图像-文本检索
  • Dissecting Deep Metric Learning Losses for Image-Text Retrieval
    :star:code
  • NAPReg: Nouns As Proxies Regularization for Semantically Aware Cross-Modal Embeddings
    :star:code
• 跨域检索
  • Contrastive Learning of Semantic Concepts for Open-set Cross-domain Retrieval
• 图像-文本匹配
  • More Than Just Attention: Improving Cross-Modal Attentions with Contrastive Constraints for Image-Text Matching

23.Autonomous Driving(智能驾驶)

• IDD-3D: Indian Driving Dataset for 3D Unstructured Road Scenes
  :star:code
• PP4AV: A Benchmarking Dataset for Privacy-Preserving Autonomous Driving
  :star:code
• Benchmarking Visual Localization for Autonomous Navigation
  :star:code
• 车辆重识别
  • Relation Preserving Triplet Mining for Stabilising the Triplet Loss In re-Identification Systems
    :star:code
• 车道线检测
  • Learning to Detect 3D Lanes by Shape Matching and Embedding
  • 3D-SpLineNet: 3D Traffic Line Detection using Parametric Spline Representations
    :house:project
• 轨迹预测
  • Robustness of Trajectory Prediction Models Under Map-Based Attacks

22.Human Action Recognition(人体动作识别与检测)

• 动作识别
  • Modality Mixer for Multi-modal Action Recognition
  • STAR-Transformer: A Spatio-temporal Cross Attention Transformer for Human Action Recognition
  • Holistic Interaction Transformer Network for Action Detection
    :star:code
  • Reconstructing Humpty Dumpty: Multi-feature Graph Autoencoder for Open Set Action Recognition
    :star:code
  • DA-AIM: Exploiting Instance-based Mixed Sampling via Auxiliary Source Domain Supervision for Domain-adaptive Action Detection
    :star:code
  • Spatio-Temporal Action Detection Under Large Motion
    :star:code
  • Efficient Skeleton-Based Action Recognition via Joint-Mapping Strategies
  • Recur, Attend or Convolve? On Whether Temporal Modeling Matters for Cross-Domain Robustness in Action Recognition
  • Semantics Guided Contrastive Learning of Transformers for Zero-Shot Temporal Activity Detection
  • Adaptive Local-Component-Aware Graph Convolutional Network for One-Shot Skeleton-Based Action Recognition
  • Multi-View Action Recognition using Contrastive Learning
    :star:code
  • Stop or Forward: Dynamic Layer Skipping for Efficient Action Recognition
  • A Simple and Efficient Pipeline to Build an End-to-End Spatial-Temporal Action Detector
    :star:code
• 时序动作定位
  • Temporal Feature Enhancement Dilated Convolution Network for Weakly-Supervised Temporal Action Localization
  • Action-aware Masking Network with Group-based Attention for Temporal Action Localization

21.Point Cloud(点云)

• PointNeuron: 3D Neuron Reconstruction via Geometry and Topology Learning of Point Clouds
• Visualizing Global Explanations of Point Cloud DNNs
  :star:code
• RSF: Optimizing Rigid Scene Flow From 3D Point Clouds Without Labels
• Nearest Neighbors Meet Deep Neural Networks for Point Cloud Analysis
• Explainability-Aware One Point Attack for Point Cloud Neural Networks
  :star:code
• Centroid Distance Keypoint Detector for Colored Point Clouds
  :star:code
• 点云分类
  • Fractual Projection Forest: Fast and Explainable Point Cloud Classifier
    :star:code
  • Cross-Modality Feature Fusion Network for Few-Shot 3D Point Cloud Classification
• 点云分割
  • Sim2real Transfer Learning for Point Cloud Segmentation: An Industrial Application Case on Autonomous Disassembly
  • 点云语义分割
    • GaIA: Graphical Information Gain based Attention Network for Weakly Supervised Point Cloud Semantic Segmentation
• 点云配准
  • Overlap-guided Gaussian Mixture Models for Point Cloud Registration
    :star:code
  • SGPCR: Spherical Gaussian Point Cloud Representation and its Application to Object Registration and Retrieval
• 点云重建
  • PointInverter: Point Cloud Reconstruction and Editing via a Generative Model with Shape Priors
    :star:code
• 3D点云
  • PIDS: Joint Point Interaction-Dimension Search for 3D Point Cloud
  • Anomaly Detection in 3D Point Clouds using Deep Geometric Descriptors

20.Transformer

• EmbryosFormer: Deformable Transformer and Collaborative Encoding-Decoding for Embryos Stage Development Classification
  :star:code
• Delving into Masked Autoencoders for Multi-Label Thorax Disease Classification
  :star:code
• Accumulated Trivial Attention Matters in Vision Transformers on Small Datasets
  :star:code
• Couplformer: Rethinking Vision Transformer With Coupling Attention
• Trans4Map: Revisiting Holistic Bird's-Eye-View Mapping From Egocentric Images to Allocentric Semantics With Vision Transformers
  :star:code
• PatchDropout: Economizing Vision Transformers Using Patch Dropout
  :star:code
• OutfitTransformer: Learning Outfit Representations for Fashion Recommendation
• Discrete Cosin TransFormer: Image Modeling From Frequency Domain
• Orthogonal Transforms For Learning Invariant Representations In Equivariant Neural Networks

19.Model Compression\Knowledge Distillation\Pruning(模型压缩\知识蒸馏\剪枝)

• 剪枝
  • Pushing the Efficiency Limit Using Structured Sparse Convolutions
    :star:code
  • Calibrating Deep Neural Networks using Explicit Regularisation and Dynamic Data Pruning
• 知识蒸馏
  • Collaborative Multi-Teacher Knowledge Distillation for Learning Low Bit-width Deep Neural Networks
  • Understanding the Role of Mixup in Knowledge Distillation: \An Empirical Study
    :star:code
  • Understanding the Role of Mixup in Knowledge Distillation:An Empirical Study
    :star:code
  • TinyHD: Efficient Video Saliency Prediction with Heterogeneous Decoders using Hierarchical Maps Distillation
  • [Online Knowledge Distillation for Multi-task Learning]](https://openaccess.thecvf.com/content/WACV2023/papers/Jacob_Online_Knowledge_Distillation_for_Multi-Task_Learning_WACV_2023_paper.pdf)
  • Adversarial local distribution regularization for knowledge distillation
• 自我蒸馏
  • SSSD: Self-Supervised Self Distillation
• DC
  • SD-Conv: Towards the Parameter-Efficiency of Dynamic Convolution
• 量化
  • HyperBlock Floating Point: Generalised Quantization Scheme for Gradient and Inference Computation
  • SPIQ: Data-Free Per-Channel Static Input Quantization
  • On Quantizing Implicit Neural Representations
  • Hyperspherical Quantization: Toward Smaller and More Accurate Models
• 轻量级
  • Learning Lightweight Neural Networks via Channel-Split Recurrent Convolution
    :star:code

18.NAS(神经架构搜索)

• Revisiting Training-free NAS Metrics: An Efficient Training-based Method
  :star:code
• SVD-NAS: Coupling Low-Rank Approximation and Neural Architecture Search
  :star:code
• FreeREA: Training-Free Evolution-based Architecture Search
  :star:code
• Toward Edge-Efficient Dense Predictions with Synergistic Multi-Task Neural Architecture Search

17.OCR(文本检测)

• OCR-VQGAN: Taming Text-within-Image Generation
  :star:code
• Efficient few-shot learning for pixel-precise handwritten document layout analysis
• D-Extract: Extracting Dimensional Attributes From Product Images
  :star:code
• 文本识别
  • Seq-UPS: Sequential Uncertainty-aware Pseudo-label Selection for Semi-Supervised Text Recognition
• 表格检测
  • LayerDoc: Layer-Wise Extraction of Spatial Hierarchical Structure in Visually-Rich Documents
• LOGO检测
  • Image-Text Pre-Training for Logo Recognition
• 文档检测
  • One-Shot Doc Snippet Detection: Powering Search in Document Beyond Text
• 文档理解
  • Multi-scale Cell-based Layout Representation for Document Understanding
    :star:code
• 文本擦除
  • Modeling Stroke Mask for End-to-End Text Erasing

16.Super-Resolution(超分辨率)

• Single Image Super-Resolution via a Dual Interactive Implicit Neural Network
• HIME: Efficient Headshot Image Super-Resolution with Multiple Exemplars
• Deep Model-Based Super-Resolution With Non-Uniform Blur
  :star:code
• Kernel-Aware Burst Blind Super-Resolution
• Enriched CNN-Transformer Feature Aggregation Networks for Super-Resolution
  :star:code
• Joint Video Rolling Shutter Correction and Super-Resolution
• 视频超分辨率
  • Cross-Resolution Flow Propagation for Foveated Video Super-Resolution
    :star:code
  • Fast Online Video Super-Resolution With Deformable Attention Pyramid
  • Efficient Reference-based Video Super-Resolution (ERVSR):Single Reference Image Is All You Need(https://github.com/haewonc/ERVSR)

15.Image Synthesis(图像合成)

• One-Shot Synthesis of Images and Segmentation Masks
  :star:code
• Style-Guided Inference of Transformer for High-resolution Image Synthesis
• Evaluating Generative Networks Using Gaussian Mixtures of Image Features
• More Control for Free! Image Synthesis with Semantic Diffusion Guidance
• 图像生成
  • Adaptively-Realistic Image Generation from Stroke and Sketch with Diffusion Model
    :star:code:house:project
  • Spatially Multi-Conditional Image Generation
    :star:code
• 文本-图像合成
  • Arbitrary Style Guidance for Enhanced Diffusion-Based Text-to-Image Generation
    :star:code
• 文字引导的图像操作
  • Interactive Image Manipulation with Complex Text Instructions

14.Un\Self\Semi-Supervised Learning(无\自\半监督学习)

• 自监督
  • Self-Supervised Pyramid Representation Learning for Multi-Label Visual Analysis and Beyond
    :star:code
  • Global-Local Self-Distillation for Visual Representation Learning
    :star:code
  • Accelerating Self-Supervised Learning via Efficient Training Strategies
  • FUSSL: Fuzzy Uncertain Self Supervised Learning
  • Self-Supervised Correspondence Estimation via Multiview Registration
    :house:project
  • Similarity Contrastive Estimation for Image and Video Soft Contrastive Self-Supervised Learning
  • Self-Supervised Relative Pose With Homography Model-Fitting in the Loop
  • Self-Distilled Self-supervised Representation Learning
    :star:code
  • Multi-Level Contrastive Learning for Self-Supervised Vision Transformers
  • Self-Supervised Distilled Learning for Multi-modal Misinformation Identification
  • An Embedding-Dynamic Approach to Self-Supervised Learning
• 半监督
  • Class-Level Confidence Based 3D Semi-Supervised Learning
  • Dynamic Re-Weighting for Long-Tailed Semi-Supervised Learning
  • Unifying Distribution Alignment as a Loss for Imbalanced Semi-supervised Learning
    :star:code
  • Semantics-Depth-Symbiosis: Deeply Coupled Semi-Supervised Learning of Semantics and Depth
  • Semi-Supervised Learning for Sparsely-Labeled Sequential Data:Application to Healthcare Video Processing
• 无监督
  • TTTFlow: Unsupervised Test-Time Training with Normalizing Flow
    :star:code

13.Image Segmentation(图像分割)

• Image Segmentation-based Unsupervised Multiple Objects Discovery
• WSNet: Towards An Effective Method for Wound Image Segmentation
  :star:code
• Autoencoder-based background reconstruction and foreground segmentation with background noise estimation
• Unsupervised multi-object segmentation using attention and soft-argmax
• VOS
  • Unsupervised Video Object Segmentation via Prototype Memory Network
  • Treating Motion as Option To Reduce Motion Dependency in Unsupervised Video Object Segmentation
    :star:code
  • A Simple and Powerful Global Optimization for Unsupervised Video Object Segmentation
    :house:project
• VSS
  • Domain Adaptive Video Semantic Segmentation via Cross-Domain Moving Object Mixing
  • Human-in-the-Loop Video Semantic Segmentation Auto-Annotation
• 语义分割
  • Attribution-aware Weight Transfer: A Warm-Start Initialization for Class-Incremental Semantic Segmentation
    :star:code
  • Full Contextual Attention for Multi-resolution Transformers in Semantic Segmentation
  • Urban Scene Semantic Segmentation with Low-Cost Coarse Annotation
  • LoopDA: Constructing Self-loops to Adapt Nighttime Semantic Segmentation
    :star:code
  • Empirical Generalization Study: Unsupervised Domain Adaptation vs. Domain Generalization Methods for Semantic Segmentation in the Wild
  • Semantic Segmentation with Active Semi-Supervised Learning
  • Self-supervised Learning with Local Contrastive Loss for Detection and Semantic Segmentation
  • Semantic Segmentation of Degraded Images Using Layer-Wise Feature Adjustor
  • Reducing Annotation Effort by Identifying and Labeling Contextually Diverse Classes for Semantic Segmentation Under Domain Shift
    :star:code
  • Cooperative Self-Training for Multi-Target Adaptive Semantic Segmentation
    :star:code
  • Refign: Align and Refine for Adaptation of Semantic Segmentation to Adverse Conditions
    :star:code
  • BEVSegFormer: Bird's Eye View Semantic Segmentation From Arbitrary Camera Rigs
  • ProtoSeg: Interpretable Semantic Segmentation with Prototypical Parts
    :star:code
  • Complementary Bi-directional Feature Compression for Indoor 360° Semantic Segmentation with Self-distillation
  • Automated Detection of Label Errors in Semantic Segmentation Datasets via Deep earning and Uncertainty Quantification
  • 弱监督语义分割
    • ingle Stage Weakly Supervised Semantic Segmentation of Complex Scenes](https://openaccess.thecvf.com/content/WACV2023/papers/Akiva_Single_Stage_Weakly_Supervised_Semantic_Segmentation_of_Complex_Scenes_WACV_2023_paper.pdf)
  • 半监督语义分割
    • Pruning-Guided Curriculum Learning for Semi-Supervised Semantic Segmentation
  • Multi-class part parsing
    • AFPSNet: Multi-Class Part Parsing Based on Scaled Attention and Feature Fusion
• BEV segmentation
  • X-Align: Cross-Modal Cross-View Alignment for Bird's-Eye-View Segmentation
    :star:code
• 全景分割
  • MonoDVPS: A Self-Supervised Monocular Depth Estimation Approach to Depth-aware Video Panoptic Segmentation
  • PRN: Panoptic Refinement Network
  • Intra-Batch Supervision for Panoptic Segmentation on High-Resolution Images
    :house:project
• 实例分割
  • From Forks to Forceps: A New Framework for Instance Segmentation of Surgical Instruments
  • CellTranspose: Few-shot Domain Adaptation for Cellular Instance Segmentation
  • Weakly Supervised Cell-Instance Segmentation With Two Types of Weak Labels by Single Instance Pasting
  • Self-Supervised Learning With Masked Image Modeling for Teeth Numbering, Detection of Dental Restorations, and Instance Segmentation in Dental Panoramic Radiographs
    :star:code
  • Weakly-Supervised Point Cloud Instance Segmentation With Geometric Priors
  • NeuralBF: Neural Bilateral Filtering for Top-down Instance Segmentation on Point Clouds
    :house:project
  • SCTS: Instance Segmentation of Single Cells Using a Transformer-Based Semantic-Aware Model and Space-Filling Augmentation
    :star:code
• 小样本分割
  • Elimination of Non-Novel Segments at Multi-Scale for Few-Shot Segmentation
  • Learning Few-shot Segmentation from Bounding Box Annotations
• 叶子疾病分割
  • AnoLeaf: Unsupervised Leaf Disease Segmentation via Structurally Robust Generative Inpainting
• 细胞分割
  • Knowing What to Label for Few Shot Microscopy Image Cell Segmentation
    :star:code
• 目标分割
  • Unsupervised 4D LiDAR Moving Object Segmentation in Stationary Settings with Multivariate Occupancy Time Series
• 抠图
  • Video Object Matting via Hierarchical Space-Time Semantic Guidance

12.One\Few-Shot Learning or Domain Adaptation\Generalization\Shift(单\小样本学习 or 域适应\泛化\偏移)

• 域适应
  • Self-Distillation for Unsupervised 3D Domain Adaptation
    :house:project
  • CoNMix for Source-free Single and Multi-target Domain Adaptation
    :star:code:house:project
  • Learning Classifiers of Prototypes and Reciprocal Points for Universal Domain Adaptation
  • Select, Label, and Mix: Learning Discriminative Invariant Feature Representations for Partial Domain Adaptation
    :house:project
  • TVT: Transferable Vision Transformer for Unsupervised Domain Adaptation
    :star:code
  • Backprop Induced Feature Weighting for Adversarial Domain Adaptation with Iterative Label Distribution Alignment
• Generative Alignment of Posterior Probabilities for Source-free Domain Adaptation
  • SALAD : Source-free Active Label-Agnostic Domain Adaptation for Classification, Segmentation and Detection
  • 半监督域适应
    • Semi-Supervised Domain Adaptation with Auto-Encoder via Simultaneous Learning
• 域泛化
  • Intra-Source Style Augmentation for Improved Domain Generalization
  • Center-aware Adversarial Augmentation for Single Domain Generalization
  • FFM: Injecting Out-of-Domain Knowledge via Factorized Frequency Modification
  • Improving Diversity with Adversarially Learned Transformations for Domain Generalization
    :star:code
• 零样本
  • Learning Attention Propagation for Compositional Zero-Shot Learning
• 小样本
  • Aggregating Bilateral Attention for Few-Shot Instance Localization
  • HyperShot: Few-Shot Learning by Kernel HyperNetworks
    :star:code
  • Few-Shot Learning of Compact Models via Task-Specific Meta Distillation
  • Semantic Guided Latent Parts Embedding for Few-Shot Learning
    :star:code
  • Self-Attention Message Passing for Contrastive Few-Shot Learning
    :star:code

11.Face(人脸)

• My Face My Choice: Privacy Enhancing Deepfakes for Social Media Anonymization
• Improving Deep Facial Phenotyping for Ultra-rare Disorder Verification Using Model Ensembles
  :star:code
• 读唇术
  • Towards MOOCs for Lip Reading: Using Synthetic Talking Heads to Train Humans in Lipreading at Scale
• 3D人脸
  • Controllable 3D Generative Adversarial Face Model via Disentangling Shape and Appearance
    :house:project
  • 3DMM-RF: Convolutional Radiance Fields for 3D Face Modeling
• 人脸识别
  • DigiFace-1M: 1 Million Digital Face Images for Face Recognition
    :star:code
  • CAST: Conditional Attribute Subsampling Toolkit for Fine-Grained Evaluation
    :star:code
  • CYBORG: Blending Human Saliency Into the Loss Improves Deep Learning-Based Synthetic Face Detection
  • Unifying Margin-Based Softmax Losses in Face Recognition
  • Harnessing Unrecognizable Faces for Improving Face Recognition
  • QMagFace: Simple and Accurate Quality-Aware Face Recognition
    :star:code
  • A Quality Aware Sample-to-Sample Comparison for Face Recognition
• 人脸修复/恢复
  • Nested Deformable Multi-head Attention for Facial Image Inpainting
    :star:code
  • AT-DDPM: Restoring Faces degraded by Atmospheric Turbulence using Denoising Diffusion Probabilistic Models
• 人脸交换
  • FaceOff: A Video-to-Video Face Swapping System
  • FaceDancer: Pose- and Occlusion-Aware High Fidelity Face Swapping
  • FastSwap: A Lightweight One-Stage Framework for Real-Time Face Swapping
    :star:code
• 人脸表情识别
  • Uncertainty-aware Label Distribution Learning for Facial Expression Recognition
    :star:code
  • 微表情识别
    • RNAS-MER: A Refined Neural Architecture Search With Hybrid Spatiotemporal Operations for Micro-Expression Recognition
• 人脸重现
  • Audio-Visual Face Reenactment
    :house:project
• 人脸命名
  • Weakly Supervised Face Naming with Symmetry-Enhanced Contrastive Loss
• 人脸重建
  • ReEnFP: Detail-Preserving Face Reconstruction by Encoding Facial Priors
• 人脸合成
  • Scaling Neural Face Synthesis to High FPS and Low Latency by Neural Caching
  • CG-NeRF: Conditional Generative Neural Radiance Fields for 3D-aware Image Synthesis
• Deepfake
  • Proactive Deepfake Defence via Identity Watermarking
• Facial Action Unit Detection
  • FAN-Trans: Online Knowledge Distillation for Facial Action Unit Detection
• 人脸质量评估
  • IFQA: Interpretable Face Quality Assessment
    :star:code
• 活体检测
• Domain Invariant Vision Transformer Learning for Face Anti-Spoofing
• 基于表情的脸部皱纹合成
  • Mesh-Tension Driven Expression-Based Wrinkles for Synthetic Faces
• 文字和图像引导的3D头像生成
  • Text and Image Guided 3D Avatar Generation and Manipulation
• 说话人脸
  • Towards Generating Ultra-High Resolution Talking-Face Videos with Lip synchronization
• 唇语阅读
  • Towards MOOCs for Lipreading: Using Synthetic Talking Heads to Train Humans in Lipreading at Scale

10.Adversarial Learning(对抗学习)

• Leveraging Local Patch Differences in Multi-Object Scenes for Generative Adversarial Attacks
• Inducing Data Amplification Using Auxiliary Datasets in Adversarial Training
  :star:code
• Interpreting Disparate Privacy-Utility Tradeoff in Adversarial Learning via Attribute Correlation
• FLOAT: Fast Learnable Once-for-All Adversarial Training for Tunable Trade-off between Accuracy and Robustness
• Adversarial robustness in discontinuous spaces via alternating sampling & descent
• PatchZero: Defending against Adversarial Patch Attacks by Detecting and Zeroing the Patch
• Avoiding Lingering in Learning Active Recognition by Adversarial Disturbance
• 对抗样本
  • Closer Look at the Transferability of Adversarial Examples:How They Fool Different Models Differently
• 主动攻击
  • Do Adaptive Active Attacks Pose Greater Risk Than Static Attacks?

9.Remote Sensing\Satellite Image(遥感\卫星图像)

• RS
  • Handling Image and Label Resolution Mismatch in Remote Sensing
  • GAF-Net: Improving the Performance of Remote Sensing Image Fusion Using Novel Global Self and Cross Attention Learning
• 变化检测
  • Self-Pair: Synthesizing Changes from Single Source for Object Change Detection in Remote Sensing Imagery
    :star:code
• 航空图像检测
  • ARUBA: An Architecture-Agnostic Balanced Loss for Aerial Object Detection
  • Transformers For Recognition In Overhead Imagery: A Reality Check
• 航空图像分割
  • Semantic Segmentation in Aerial Imagery Using Multi-level Contrastive Learning with Local Consistency
• 国际边界检测
  • Computer Vision for International Border Legibility

8.Image Processing(图像处理)

• 图像质量评估
  • No Reference Opinion Unaware Quality Assessment of Authentically Distorted Images
    :star:code
• 图像恢复
  • Large-to-small Image Resolution Asymmetry in Deep Metric Learning
    :star:code
  • DSTrans: Dual-Stream Transformer for Hyperspectral Image Restoration
    :star:code
  • Semi-Supervised Learning for Low-light Image Restoration through Quality Assisted Pseudo-Labeling
    :star:code
  • Real-Time Restoration of Dark Stereo Images
    :house:project
• 图像修复
  • GeoFill: Reference-Based Image Inpainting With Better Geometric Understanding
  • Keys to Better Image Inpainting: Structure and Texture Go Hand in Hand
    :house:project
• 图像增强
  • Perceptual Image Enhancement for Smartphone Real-Time Applications
    :star:code
  • Robust Real-World Image Enhancement Based on Multi-Exposure LDR Images
  • End-to-End Single-Frame Image Signal Processing for High Dynamic Range Scenes
  • PSENet: Progressive Self-Enhancement Network for Unsupervised Extreme-Light Image Enhancement
    :star:code
• 图像着色
  • Guiding Users to Where to Give Color Hints for Efficient Interactive Sketch Colorization via Unsupervised Region Prioritization
  • Generative Colorization of Structured Mobile Web Pages
    :star:code
  • iColoriT: Towards Propagating Local Hints to the Right Region in Interactive Colorization by Leveraging Vision Transformer
    :house:project
  • Pik-Fix: Restoring and Colorizing Old Photos
    :star:code
• 图像补全
  • An Unified Framework for Language Guided Image Completion
• 图像重新缩放
  • Effective Invertible Arbitrary Image Rescaling
• HDR重构
  • Single-Image HDR Reconstruction by Multi-Exposure Generation
    :star:code
• 去噪
  • On the Importance of Denoising When Learning To Compress Images
    :star:code
  • Self Supervised Low Dose Computed Tomography Image Denoising Using Invertible Network Exploiting Inter Slice Congruence
  • Video joint denoising and demosaicing with recurrent CNNs
    :house:project
• 去雾
  • Aerial Image Dehazing With Attentive Deformable Transformers
    :star:code
• 去反射
  • Burst Reflection Removal using Reflection Motion Aggregation Cues
• De-fencing
  • Efficient Flow-Guided Multi-frame De-fencing
• Deconvolution
  • Wiener Guided DIP for Unsupervised Blind Image Deconvolution
• 阴影消除
  • Fine-Context Shadow Detection Using Shadow Removal
  • LRA&LDRA: Rethinking Residual Predictions for Efficient Shadow Detection and Removal
  • SHARDS: Efficient SHAdow Removal using Dual Stage Network for High-Resolution Images

7.Human Pose(人体姿态)

• Kinematic-aware Hierarchical Attention Network for Human Pose Estimation in Videos
  :star:code
• HuPR: A Benchmark for Human Pose Estimation Using Millimeter Wave Radar
  :star:code
• Computer Vision to the Rescue: Infant Postural Symmetry Estimation from Incongruent Annotations
  :star:code
• 多人姿态估计
  • SoMoFormer: Multi-Person Pose Forecasting with Transformers
    :house:project
• 三维人体
  • Placing Human Animations into 3D Scenes by Learning Interaction- and Geometry-Driven Keyframes
  • Uplift and Upsample: Efficient 3D Human Pose Estimation with Uplifting Transformers
    :star:code
  • Learning 3D Human Pose Estimation from Dozens of Datasets using a Geometry-Aware Autoencoder to Bridge Between Skeleton Formats
    :house:project
  • GarSim: Particle Based Neural Garment Simulator
  • Learnable Human Mesh Triangulation for 3D Human Pose and Shape Estimation
  • ElliPose: Stereoscopic 3D Human Pose Estimation by Fitting Ellipsoids
  • Rethinking the Data Annotation Process for Multi-view 3D Pose Estimation with Active Learning and Self-Training
  • CameraPose: Weakly-Supervised Monocular 3D Human Pose Estimation by Leveraging In-the-wild 2D Annotations
• 手部姿势
  • 3D手
    • Marker-Removal Networks To Collect Precise 3D Hand Data for RGB-Based Estimation and Its Application in Piano
    • HandGCNFormer: A Novel Topology-Aware Transformer Network for 3D Hand Pose Estimation
    • Image-free Domain Generalization via CLIP for 3D Hand Pose Estimation
  • 手部重建
    • THOR-Net: End-to-end Graformer-based Realistic Two Hands and Object Reconstruction with Self-supervision
      :star:code
  • 手-物体姿势估计
    • Interacting Hand-Object Pose Estimation via Dense Mutual Attention
      :star:code

6.Video(视频相关)

• A Deep Neural Framework to Detect Individual Advertisement (Ad) from Videos
• TCAM: Temporal Class Activation Maps for Object Localization in Weakly-Labeled Unconstrained Videos
  :star:code
• Recipe2Video: Synthesizing Personalized Videos from Recipe Texts
• 视频增强
  • Fast and Accurate: Video Enhancement Using Sparse Depth
• 视频理解
  • Event-Specific Audio-Visual Fusion Layers:A Simple and New Perspective on Video Understanding
  • ViewCLR: Learning Self-supervised Video Representation for Unseen Viewpoints
  • 通用事件边界检测
    • Motion Aware Self-Supervision for Generic Event Boundary Detection
      :star:code
• 视频摘要
  • Contrastive Losses Are Natural Criteria for Unsupervised Video Summarization
    :star:code
  • Progressive Video Summarization via Multimodal Self-Supervised Learning
• 多人检测
  • Two-level Data Augmentation for Calibrated Multi-view Detection
    :star:code
• 场景识别
  • MovieCLIP: Visual Scene Recognition in Movies
    :house:project
• Video Grounding
  • Language-free Training for Zero-shot Video Grounding
• 视频异常检测(VAD)
  • DyAnNet: A Scene Dynamicity Guided Self-Trained Video Anomaly Detection Network
  • Cross-Domain Video Anomaly Detection without Target Domain Adaptation
  • Bi-Directional Frame Interpolation for Unsupervised Video Anomaly Detection
  • Towards Interpretable Video Anomaly Detection
  • Normality Guided Multiple Instance Learning for Weakly Supervised Video Anomaly Detection
• 图像视频编解码
  • Universal Deep Image Compression via Content-Adaptive Optimization with Adapters
    :star:code
  • A neural video codec with spatial rate-distortion control
  • Boosting Neural Video Codecs by Exploiting Hierarchical Redundancy
  • Neural Distributed Image Compression with Cross-Attention Feature Alignment
    :star:code
  • Lossy Image Compression with Quantized Hierarchical VAEs
• 视频人像合成
  • Dynamic Neural Portraits
• 视频帧插值
  • Splatting-based Synthesis for Video Frame Interpolation
    :house:project
  • FLAVR: Flow-Agnostic Video Representations for Fast Frame Interpolation
    :house:project
  • Enhanced Bi-directional Motion Estimation for Video Frame Interpolation
    :star:code
• 视频运动重定位
  • Cross-Identity Video Motion Retargeting With Joint Transformation and Synthesis
    :star:code
• 视频运动放大
  • Lightweight Network For Video Motion Magnification
    :star:code
• 视频稳定
  • GlobalFlowNet: Video Stabilization using Deep Distilled Global Motion Estimates
    :star:code
  • Sim2RealVS: A New Benchmark for Video Stabilization with a Strong Baseline
• 视频分类
  • MASTAF: A Model-Agnostic Spatio-Temporal Attention Fusion Network for Few-shot Video Classification
    :house:project
  • Visually explaining 3D-CNN predictions for video classification with an adaptive occlusion sensitivity analysis
    :star:code
• 视频分割
  • LiveSeg: Unsupervised Multimodal Temporal Segmentation of Long Livestream Videos
• 视频伪造检测
  • Watch Those Words:Video Falsification Detection Using Word-Conditioned Facial Motion
• 视频跟踪
  • PreViTS: Contrastive Pretraining with Video Tracking Supervision

5.Object Detection(目标检测)

• ConfMix: Unsupervised Domain Adaptation for Object Detection via Confidence-based Mixing
  :star:code
• The Box Size Confidence Bias Harms Your Object Detector
  :star:code
• Resolving Class Imbalance for LiDAR-based Object Detector by Dynamic Weight Average and Contextual Ground Truth Sampling
• Is Your Noise Correction Noisy? PLS: Robustness To Label Noise With Two Stage Detection
  :star:code
• Phantom Sponges: Exploiting Non-Maximum Suppression to Attack Deep Object Detectors
• Domain Adaptive Object Detection for Autonomous Driving under Foggy Weather
  :star:code
• ROMA: Run-Time Object Detection To Maximize Real-Time Accuracy
• Towards Online Domain Adaptive Object Detection
  :star:code
• MT-DETR: Robust End-to-end Multimodal Detection with Confidence Fusion
  :star:code
• Towards Few-Annotation Learning for Object Detection:Are Transformer-based Models More Efficient ?
• Scaling Novel Object Detection with Weakly Supervised Detection Transformers
• Mobile Robot Manipulation using Pure Object Detection
  :star:code
• Domain Adaptation Using Self-Training With Mixup for One-Stage Object Detection
• Gradient-Based Quantification of Epistemic Uncertainty for Deep Object Detectors
  :star:code
• 小样本目标检测
  • Few-shot Object Detection via Improved Classification Features
• 弱监督目标检测
  • D2F2WOD: Learning Object Proposals for Weakly-Supervised Object Detection via Progressive Domain Adaptation
  • Complementary Cues from Audio Help Combat Noise in Weakly-Supervised Object Detection
    :house:project
• 3D目标检测
  • TransPillars: Coarse-To-Fine Aggregation for Multi-Frame 3D Object Detection
  • Adaptive Feature Fusion for Cooperative Perception Using LiDAR Point Clouds
  • ImpDet: Exploring Implicit Fields for 3D Object Detection
  • Li3DeTr: A LiDAR based 3D Detection Transformer
  • Far3Det: Towards Far-Field 3D Detection
  • Dense Voxel Fusion for 3D Object Detection
  • MonoEdge: Monocular 3D Object Detection Using Local Perspectives
  • Multivariate Probabilistic Monocular 3D Object Detection
    :star:code
  • SAILOR: Scaling Anchors via Insights into Latent Object Representation
• VOD
  • BoxMask: Revisiting Bounding Box Supervision for Video Object Detection
• OOD
  • Out-of-distribution Detection via Frequency-regularized Generative Models
    :star:code
  • Heatmap-based Out-of-Distribution Detection
    :star:code
  • Out-of-Distribution Detection with Reconstruction Error and Typicality-based Penalty
  • Mixture Outlier Exposure: Towards Out-of-Distribution Detection in Fine-grained Environments
    :star:code
  • Hyperdimensional Feature Fusion for Out-of-Distribution Detection
    :star:code
  • Task Agnostic and Post-hoc Unseen Distribution Detection
• WSOD
  • D2DF2WOD: Learning Object Proposals for Weakly-Supervised Object Detection via Progressive Domain Adaptation
• 伪装目标检测
  • MFFN: Multi-view Feature Fusion Network for Camouflaged Object Detection
    :star:code
• 目标发现
  • Foreground Guidance and Multi-Layer Feature Fusion for Unsupervised Object Discovery with Transformers
    :star:code
• 变化检测
  • The Change You Want to See
• 用于穿行式安检系统的三维雷达图像的实时隐蔽武器检测
  • Real-Time Concealed Weapon Detection on 3D Radar Images for Walk-Through Screening System
• 图像识别
  • Federated Domain Generalization for Image Recognition via Cross-Client Style Transfer
    :house:project
• 入侵物种检测
  • Deep Learning Methodology for Early Detection and Outbreak Prediction of Invasive Species Growth
• 用于红外图像中的海洋涡流检测
  • Computer Vision for Ocean Eddy Detection in Infrared Imagery

4.GAN(生成对抗网络)

• HoechstGAN: Virtual Lymphocyte Staining Using Generative Adversarial Networks
• Image Completion with Heterogeneously Filtered Spectral Hints
  :star:code
• Indirect Adversarial Losses via an Intermediate Distribution for Training GANs
• SLI-pSp: Injecting Multi-Scale Spatial Layout in pSp
• Multi-scale Contrastive Learning for Complex Scene Generation
• Realistic Full-Body Anonymization with Surface-Guided GANs
• Fantastic Style Channels and Where to Find Them:A Submodular Framework for Discovering Diverse Directions in GANs
• 3D GAN Inversion with Pose Optimization
  :house:project
• UVCGAN: UNet Vision Transformer cycle-consistent GAN for unpaired image-to-image translation
  :star:code
• SketchInverter: Multi-Class Sketch-Based Image Generation via GAN Inversion
• 风格编辑
  • DyStyle: Dynamic Neural Network for Multi-Attribute-Conditioned Style Editings
• fashion attribute editing(时尚属性编辑)
  • Leveraging Off-the-shelf Diffusion Model for Multi-attribute Fashion Image Manipulation
• 匿名化
  • DeepPrivacy2: Towards Realistic Full-Body Anonymization
    :star:code
• 指纹生成
  • Synthetic Latent Fingerprint Generator
    :house:project
• 开集识别
  • MORGAN: Meta-Learning-based Few-Shot Open-Set Recognition via Generative Adversarial Network
    :star:code

3.3D(三维视觉)

• Generative Range Imaging for Learning Scene Priors of 3D LiDAR Data
  :star:code
• Seg&Struct: The Interplay Between Part Segmentation and Structure Inference for 3D Shape Parsing
• Surface normal estimation from optimized and distributed light sources using DNN-based photometric stereo
• Meta-Auxiliary Learning for Future Depth Prediction in Videos
• 3D Neural Sculpting (3DNS): Editing Neural Signed Distance Functions
• Improving the Robustness of Point Convolution on k-Nearest Neighbor Neighborhoods with a Viewpoint-Invariant Coordinate Transform
• CountNet3D: A 3D Computer Vision Approach to Infer Counts of Occluded Objects
• Learning Graph Variational Autoencoders with Constraints and Structured Priors for Conditional Indoor 3D Scene Generation
• 三维重建
  • Automated Line Labelling: Dataset for Contour Detection and 3D Reconstruction
• 表面重建
  • Recovering Fine Details for Neural Implicit Surface Reconstruction
    :star:code
• 深度估计
  • Frequency-Aware Self-Supervised Monocular Depth Estimation
    :star:code
  • Attention Attention Everywhere: Monocular Depth Prediction with Skip Attention
    :star:code
  • High-Resolution Depth Estimation for 360-degree Panoramas through Perspective and Panoramic Depth Images Registration
  • Improving Pixel-Level Contrastive Learning by Leveraging Exogenous Depth Information
  • Temporally Consistent Online Depth Estimation in Dynamic Scenes
    :star:code
  • Self-Supervised Monocular Depth Estimation: Solving the Edge-Fattening Problem
    :star:code
  • High-Resolution Depth Estimation for 360◦ Panoramas through Perspective and Panoramic Depth Images Registration
  • Self-supervised Monocular Depth Estimation from Thermal Images via Adversarial Multi-spectral Adaptation
• 深度补全
  • Sparsity Agnostic Depth Completion
  • SIUNet: Sparsity Invariant U-Net for Edge-Aware Depth Completion
  • Uncertainty-Aware Interactive LiDAR Sampling for Deep Depth Completion
• MVS
  • Multi-View Photometric Stereo Revisited
  • DELS-MVS: Deep Epipolar Line Search for Multi-View Stereo
  • nLMVS-Net: Deep Non-Lambertian Multi-View Stereo
    :house:project
  • 360MVSNet: Deep Multi-view Stereo Network with 360◦ Images for Indoor Scene Reconstruction
  • Improving the Pair Selection and the Model Fusion Steps of Satellite Multi-View Stereo Pipelines
• RGB-D重建
  • High-Quality RGB-D Reconstruction via Multi-View Uncalibrated Photometric Stereo and Gradient-SDF
    :star:code
• Stereo Matching
  • Expansion of Visual Hints for Improved Generalization in Stereo Matching
• 神经辐射场
  • ScanNeRF: a Scalable Benchmark for Neural Radiance Fields
    :house:project
• 三维定位
  • 3D Change Localization and Captioning From Dynamic Scans of Indoor Scenes
• CAD
  • Automatically Annotating Indoor Images with CAD Models via RGB-D Scans

2.Medical Image(医学影像)

• DBCE: A Saliency Method for Medical Deep Learning Through Anatomically-Consistent Free-Form Deformations
• Representation Recovering for Self-Supervised Pre-training on Medical Images
• Magnification Prior: A Self-Supervised Method for Learning Representations on Breast Cancer Histopathological Images
  :star:code
• A Morphology Focused Diffusion Probabilistic Model for Synthesis of Histopathology Images
• 3D医学影像分析
  • Masked Image Modeling Advances 3D Medical Image Analysis
    :star:code
• 胸部X光分类
  • Probabilistic Integration of Object Level Annotations in Chest X-ray Classification
• CT图像融合
  • Self-Supervised 2D/3D Registration for X-Ray to CT Image Fusion
• 医学图像定位
  • Attend Who is Weak: Pruning-assisted Medical Image Localization under Sophisticated and Implicit Imbalances
• 医学图像分割
  • Few-shot Medical Image Segmentation with Cycle-resemblance Attention
  • Medical Image Segmentation via Cascaded Attention Decoding
  • HiFormer: Hierarchical Multi-scale Representations Using Transformers for Medical Image Segmentation
  • Training Auxiliary Prototypical Classifiers for Explainable Anomaly Detection in Medical Image Segmentation
  • The Fully Convolutional Transformer for Medical Image Segmentation
    :star:code
  • 病变分割
    • Cut-Paste Consistency Learning for Semi-Supervised Lesion Segmentation
      :star:code
• 医学图像分类
  • ScoreNet: Learning Non-Uniform Attention and Augmentation for Transformer-Based Histopathological Image Classification
    :star:code
• 医学图像超分辨率
  • Multimodal Multi-Head Convolutional Attention With Various Kernel Sizes for Medical Image Super-Resolution
    :star:code
• 心血管检测
  • Performer: A Novel PPG-to-ECG Reconstruction Transformer for a Digital Biomarker of Cardiovascular Disease Detection
• 远程心率估计
  • ALPINE: Improving Remote Heart Rate Estimation using Contrastive Learning
• CT重建
  • PINER: Prior-informed Implicit Neural Representation Learning for Test-time Adaptation in Sparse-view CT Reconstruction
    :star:code
• MRI
  • MRI Imputation based on Fused Index- and Intensity-Registration
  • DSFormer: A Dual-domain Self-supervised Transformer for Accelerated Multi-contrast MRI Reconstruction
• 黑色素细胞检测
  • VSGD-Net: Virtual Staining Guided Melanocyte Detection on Histopathological Images
    :star:code

1.其它

• Instance-Dependent Noisy Label Learning via Graphical Modelling
• Color Recommendation for Vector Graphic Documents based on Multi-Palette Representation
• TeST: Test-time Self-Training under Distribution Shift
• Simultaneous Acquisition of High Quality RGB Image and Polarization Information using a Sparse Polarization Sensor
  :star:code
• Exemplar Guided Deep Neural Network for Spatial Transcriptomics Analysis of Gene Expression Prediction
• Enabling ISP-less Low-Power Computer Vision
• AdaNorm: Adaptive Gradient Norm Correction based Optimizer for CNNs
  :star:code
• Composite Learning for Robust and Effective Dense Predictions
• SAILOR: Scaling Anchors via Insights into Latent Object
  :star:code
• Modeling the Lighting in Scenes as Style for Auto White-Balance Correction
  :star:code
• DE-CROP: Data-efficient Certified Robustness for Pretrained Classifiers
  :house:project
• Anisotropic Multi-Scale Graph Convolutional Network for Dense Shape Correspondence
• ATCON: Attention Consistency for Vision Models
  :star:code
• LAVA: Label-efficient Visual Learning and Adaptation
• Interpolated SelectionConv for Spherical Images and Surfaces
• Augmentation by Counterfactual Explanation -- Fixing an Overconfident Classifier
• Weakly Supervised Annotations for Multi-modal Greeting Cards Dataset
• Multimodal Vision Transformers with Forced Attention for Behavior Analysis
  :star:code
• Compact and Optimal Deep Learning with Recurrent Parameter Generators
• Motif Mining: Finding and Summarizing Remixed Image Content
• LINEEX: Data Extraction from Scientific Line Charts
  :star:code
• Neural Implicit Representations for Physical Parameter Inference From a Single Video
  :house:project
• Physically Plausible Animation of Human Upper Body from a Single Image
• Partially Calibrated Semi-Generalized Pose From Hybrid Point Correspondences
• Learning How to MIMIC: Using Model Explanations To Guide Deep Learning Training
  :star:code
• Robust and Efficient Alignment of Calcium Imaging Data through Simultaneous Low Rank and Sparse Decomposition
• Improving Multi-Fidelity Optimization With a Recurring Learning Rate for Hyperparameter Tuning
• What can we Learn by Predicting Accuracy?
• Enabling ISPless Low-Power Computer Vision
  :star:code
• Jointly Learning Band Selection and Filter Array Design for Hyperspectral Imaging
• LCS: Learning Compressible Subspaces for Efficient, Adaptive, Real-Time Network Compression at Inference Time
  :star:code
• Self-Attentive Pooling for Efficient Deep Learning
  :star:code
• Fine-Grained Activities of People Worldwide
  :house:project
• Relaxing Contrastiveness in Multimodal Representation Learning
• Spike-Based Anytime Perception
• Towards Disturbance-Free Visual Mobile Manipulation
  :house:project
• SERF: Towards Better Training of Deep Neural Networks Using Log-Softplus ERror Activation Function
• RADIANT: Better rPPG Estimation Using Signal Embeddings and Transformer
  :star:code
• Dataset Condensation With Distribution Matching
• HyperPosePDF - Hypernetworks Predicting the Probability Distribution on SO(3)
• RANCER: Non-Axis Aligned Anisotropic Certification with Randomized Smoothing
• Match Cutting: Finding Cuts with Smooth Visual Transitions
• SIRA: Relightable Avatars from a Single Image
• Are Straight-Through gradients and Soft-Thresholding all you need for Sparse Training?
  :star:code
• Patch-based Privacy Preserving Neural Network for Vision Tasks
• Adaptive Sample Selection for Robust Learning under Label Noise
  :star:code
• Concept Correlation and Its Effects on Concept-Based Models
• Improving Saliency Models' Predictions of the Next Fixation With Humans' Intrinsic Cost of Gaze Shifts
• Mapping DNN Embedding Manifolds for Network Generalization Prediction
• GEMS: Generating Efficient Meta-Subnets
• Learning incoherent light emission steering from metasurfaces using generative models
  :star:code
• EfficientPhys: Enabling Simple, Fast and Accurate Camera-Based Cardiac Measurement
• Performance comparison of DVS data spatial downscaling methods using Spiking Neural Networks
  :star:code
• Encouraging Disentangled and Convex Representation with Controllable Interpolation Regularization
• A Protocol for Evaluating Model Interpretation Methods from Visual Explanations
• Learning Latent Structural Relations with Message Passing Prior
• Bootstrapping the Relationship Between Images and Their Clean and Noisy Labels
  :star:code
• ImPosing: Implicit Pose Encoding for Efficient Visual Localization
• GEMS: Scene Expansion using Generative Models of Graphs
• SONGs: Self-Organizing Neural Graphs
  :star:code
• Context-empowered Visual Attention Prediction in Pedestrian Scenarios
• Exploiting Long-Term Dependencies for Generating Dynamic Scene Graphs
• BNN
  • LAB: Learnable Activation Binarizer for Binary Neural Networks
    :star:code
  • Searching for Robust Binary Neural Networks via Bimodal Parameter Perturbation
• 图像配准
  • Diffeomorphic Image Registration with Neural Velocity Field
• 视觉重建
  • SimGlim: Simplifying glimpse based active visual reconstruction