https://github.com/52cv/cvpr-2022-papers
https://github.com/52cv/cvpr-2022-papers
Last synced: 5 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/52cv/cvpr-2022-papers
- Owner: 52CV
- Created: 2022-01-19T02:39:36.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2022-07-28T06:14:49.000Z (almost 4 years ago)
- Last Synced: 2025-02-24T05:14:39.052Z (over 1 year ago)
- Size: 3.73 MB
- Stars: 633
- Watchers: 21
- Forks: 58
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# CVPR-2022-Papers
官网链接：https://cvpr2022.thecvf.com/
开会时间：2022年6月19日-6月24日
### ❣❣❣近日，[CVPR 2022 接收论文公布！ 总计2067篇！](https://mp.weixin.qq.com/s/WfzbGK34z3gIk1E9su8moA)，全部论文已发布，多多关注!!
### ❣❣❣另外打包下载所有论文，可在[【我爱计算机视觉】微信公众号](https://user-images.githubusercontent.com/62801906/163739684-175f0b8a-871e-4a41-b310-b549625fdcb1.png)后台回复“paper”。
## 历年综述论文分类汇总戳这里↘️[CV-Surveys](https://github.com/52CV/CV-Surveys)施工中~~~~~~~~~~
## 2022 年论文分类汇总戳这里
↘️[CVPR-2022-Papers](https://github.com/52CV/CVPR-2022-Papers)
↘️[WACV-2022-Papers](https://github.com/52CV/WACV-2022-Papers)
## 2021年论文分类汇总戳这里
↘️[ICCV-2021-Papers](https://github.com/52CV/ICCV-2021-Papers)
↘️[CVPR-2021-Papers](https://github.com/52CV/CVPR-2021-Papers)
## 2020 年论文分类汇总戳这里
↘️[CVPR-2020-Papers](https://github.com/52CV/CVPR-2020-Papers)
↘️[ECCV-2020-Papers](https://github.com/52CV/ECCV-2020-Papers)
## 目录
|:cat:|:dog:|:tiger:|:wolf:|
|------|------|------|------|
|[1.其它](#1)|[2.Image Segmentation(图像分割)](#2)|[3.Image Progress(图像处理)](#4)|[4.Image Captioning(图像字幕)](#)|
|[5.Object Detection(目标检测)](#5)|[6.Object Tracking(目标跟踪)](#6)|[7.Point Cloud(点云)](#7)|[8.Action Detection(人体动作检测与识别)](#8)|
|[9.Human Pose Estimation(人体姿态估计)](#9)|[10.3D(三维视觉)](#10)|[11.Face](#11)|[12.Image-to-Image Translation(图像到图像翻译)](#12)|
|[13.GAN](#13)|[14.Video](#14)|[15.Transformer](#15)|[16.Semi/self-supervised learning(半/自监督)](#16)|
|[17.Medical Image(医学影像)](#17)|[18.Person Re-Identification(人员重识别)](#18)|[19.Neural Architecture Search(神经架构搜索)](#19)|[20.Autonomous vehicles(自动驾驶)](#20)|
|[21.UAV/Remote Sensing/Satellite Image(无人机/遥感/卫星图像)](#21)|[22.Image Synthesis/Generation(图像合成)](#22)|[23.Image Retrieval(图像检索)](#23)|[24.Super-Resolution(超分辨率)](#24)|
|[25.Fine-Grained/Image Classification(细粒度/图像分类)](#25)|[26.GCN/GNN](#26)|[27.Pose Estimation(物体姿势估计)](#27)|[28.Style Transfer(风格迁移)](#28)|
|[29.Augmented Reality/Virtual Reality/Robotics(增强/虚拟现实/机器人)](#29)|[30.Visual Answer Questions(视觉问答)](#30)|[31.Vision-Language(视觉语言)](#31)|[32.Data Augmentation(数据增强)](#32)|
|[33.Human-Object Interaction(人物交互)](#33)|[34.Model Compression/Knowledge Distillation/Pruning(模型压缩/知识蒸馏/剪枝)](#34)|[35.OCR](#35)|[36.Optical Flow(光流估计)](#36)|
|[37.Contrastive Learning(对比学习)](#37)|[38.Meta-Learning(元学习)](#38)|[39.Continual Learning(持续学习)](#39)|[40.Adversarial Learning(对抗学习)](#40)|
|[41.Incremental Learning(增量学习)](#41)|[42.Metric Learning(度量学习)](#42)|[43.Multi-Task Learning(多任务学习)](#43)|[44.Federated Learning(联邦学习)](#44)|
|[45.Dense Prediction(密集预测)](#45)|[46.Scene Graph Generation(场景图生成)](#46)|[47.Few/Zero-Shot Learning/Domain Generalization/Adaptation(小/零样本/域泛化/适应)](#47)|[48.Visual Grounding](#48)|
|[49.Image Geo-localization(图像地理定位)](#49)|[50.Anomaly Detection(异常检测)](#50)|[51.光学、几何、光场成像](#51)|[52.Human Motion Forecasting(人体运动预测)](#52)|
|[53.Sign Language Translation(手语翻译)](#53)|[54.Dataset(数据集)](#54)|[55.Novel View Synthesis(视图合成)](#55)|[56.Sound](#56)|
|[57.Gaze Estimation(视线估计)](#57)|[58.Neural rendering(神经渲染)](#58)|[59.动画](#59)|[60.Visual Emotion Analysis(视觉情感分析)](#60)|
* 聚类
* [DeepDPM: Deep Clustering With an Unknown Number of Clusters](https://arxiv.org/abs/2203.14309)
:star:[code](https://github.com/BGU-CS-VIL/DeepDPM)
* 场景流
* [Exploiting Rigidity Constraints for LiDAR Scene Flow Estimation](https://openaccess.thecvf.com/content/CVPR2022/papers/Dong_Exploiting_Rigidity_Constraints_for_LiDAR_Scene_Flow_Estimation_CVPR_2022_paper.pdf)
:star:[code](https://github.com/gtdong-ustc/LiDARSceneFlow)
* 图识别
* [Improving Subgraph Recognition With Variational Graph Information Bottleneck](https://arxiv.org/abs/2112.09899)
:star:[code](https://github.com/Samyu0304/Improving-Subgraph-Recognition-with-Variation-Graph-Information-Bottleneck-VGIB-)
* 运动模糊
* [Motion-From-Blur: 3D Shape and Motion Estimation of Motion-Blurred Objects in Videos](https://openaccess.thecvf.com/content/CVPR2022/papers/Rozumnyi_Motion-From-Blur_3D_Shape_and_Motion_Estimation_of_Motion-Blurred_Objects_in_CVPR_2022_paper.pdf)
* 人像眼镜和阴影消除
* [Portrait Eyeglasses and Shadow Removal by Leveraging 3D Synthetic Data](https://arxiv.org/abs/2203.10474)
:star:[code](https://github.com/StoryMY/take-off-eyeglasses)
* 识别唇语
* [Sub-Word Level Lip Reading With Visual Attention](https://arxiv.org/abs/2110.07603)
* 模拟时钟读数
* [It's About Time: Analog Clock Reading in the Wild](https://openaccess.thecvf.com/content/CVPR2022/papers/Yang_Its_About_Time_Analog_Clock_Reading_in_the_Wild_CVPR_2022_paper.pdf)
:star:[code](https://github.com/charigyang/itsabouttime):house:[project](https://www.robots.ox.ac.uk/~vgg/research/time/)
* 指纹识别
* [Fingerprinting Deep Neural Networks Globally via Universal Adversarial Perturbations](https://arxiv.org/abs/2202.08602)
:open_mouth:oral
* 基于草图的图像操作
* [SketchEdit: Mask-Free Local Image Manipulation with Partial Sketches](https://arxiv.org/abs/2111.15078)
:star:[code](https://github.com/zengxianyu/sketchedit):house:[project](https://zengxianyu.github.io/sketchedit/)
* 草图识别
* [Finding Badly Drawn Bunnies](https://openaccess.thecvf.com/content/CVPR2022/papers/Yang_Finding_Badly_Drawn_Bunnies_CVPR_2022_paper.pdf)
* 去偏移
* [Debiased Learning From Naturally Imbalanced Pseudo-Labels](https://arxiv.org/abs/2201.01490)
:star:[code](https://github.com/frank-xwang/debiased-pseudo-labeling)
* 线段分类
* [Transformer Based Line Segment Classifier With Image Context for Real-Time Vanishing Point Detection in Manhattan World](https://openaccess.thecvf.com/content/CVPR2022/papers/Tong_Transformer_Based_Line_Segment_Classifier_With_Image_Context_for_Real-Time_CVPR_2022_paper.pdf)
* Interactive object understanding
* [Human Hands as Probes for Interactive Object Understanding](https://arxiv.org/abs/2112.09120)
:star:[code](https://github.com/uiuc-robovision/hands-as-probes):house:[project](https://s-gupta.github.io/hands-as-probes/)
* 数字人类
* [GOAL: Generating 4D Whole-Body Motion for Hand-Object Grasping](https://arxiv.org/abs/2112.11454)
:star:[code](https://github.com/otaheri/GOAL):house:[project](https://goal.is.tue.mpg.de/)
* 强化学习
* [DECORE: Deep Compression With Reinforcement Learning](https://arxiv.org/abs/2106.06091)
* 视觉关系检测
* [A Probabilistic Graphical Model Based on Neural-symbolic Reasoning for Visual Relationship Detection](https://openaccess.thecvf.com/content/CVPR2022/papers/Yu_A_Probabilistic_Graphical_Model_Based_on_Neural-Symbolic_Reasoning_for_Visual_CVPR_2022_paper.pdf)
* 裂缝识别
* [Geometry-Aware Guided Loss for Deep Crack Recognition](https://openaccess.thecvf.com/content/CVPR2022/papers/Chen_Geometry-Aware_Guided_Loss_for_Deep_Crack_Recognition_CVPR_2022_paper.pdf)
* 眼球认证
* [EyePAD++: A Distillation-Based Approach for Joint Eye Authentication and Presentation Attack Detection Using Periocular Images](https://openaccess.thecvf.com/content/CVPR2022/papers/Dhar_EyePAD_A_Distillation-Based_Approach_for_Joint_Eye_Authentication_and_Presentation_CVPR_2022_paper.pdf)
* 视听事件定位
* [Cross-Modal Background Suppression for Audio-Visual Event Localization](https://openaccess.thecvf.com/content/CVPR2022/papers/Xia_Cross-Modal_Background_Suppression_for_Audio-Visual_Event_Localization_CVPR_2022_paper.pdf)
:star:[code](https://github.com/marmot-xy/CMBS)
* 无偏见学习
* [A Conservative Approach for Unbiased Learning on Unknown Biases](https://openaccess.thecvf.com/content/CVPR2022/papers/Jeon_A_Conservative_Approach_for_Unbiased_Learning_on_Unknown_Biases_CVPR_2022_paper.pdf)
:star:[code](https://github.com/aandyjeon/UBNet)
* Object Proposal Generation
* [ProposalCLIP: Unsupervised Open-Category Object Proposal Generation via Exploiting CLIP Cues](https://arxiv.org/abs/2201.06696)
* 读唇术
* [Multi-Grained Spatio-Temporal Features Perceived Network for Event-Based Lip-Reading](https://openaccess.thecvf.com/content/CVPR2022/papers/Tan_Multi-Grained_Spatio-Temporal_Features_Perceived_Network_for_Event-Based_Lip-Reading_CVPR_2022_paper.pdf)
:star:[code](https://github.com/tgc1997/event-based-lip-reading):house:[project](https://sites.google.com/view/event-based-lipreading)
* 对应学习
* [MS2DG-Net: Progressive Correspondence Learning via Multiple Sparse Semantics Dynamic Graph](https://openaccess.thecvf.com/content/CVPR2022/papers/Dai_MS2DG-Net_Progressive_Correspondence_Learning_via_Multiple_Sparse_Semantics_Dynamic_Graph_CVPR_2022_paper.pdf)
:star:[code](https://github.com/changcaiyang/MS2DG-Net)
* 视觉定位
* [Zero Experience Required: Plug & Play Modular Transfer Learning for Semantic Visual Navigation](https://openaccess.thecvf.com/content/CVPR2022/papers/Al-Halah_Zero_Experience_Required_Plug__Play_Modular_Transfer_Learning_for_CVPR_2022_paper.pdf)
:star:[code](https://github.com/ziadalh/zero_experience_required):house:[project](https://vision.cs.utexas.edu/projects/zsel/)
* 视觉识别
* [Causal Transportability for Visual Recognition](https://arxiv.org/abs/2204.12363)
:star:[code](https://github.com/cvlab-columbia/CT4Recognition)
* [A Simple Episodic Linear Probe Improves Visual Recognition in the Wild](https://openaccess.thecvf.com/content/CVPR2022/papers/Liang_A_Simple_Episodic_Linear_Probe_Improves_Visual_Recognition_in_the_CVPR_2022_paper.pdf)
* [Contextual Debiasing for Visual Recognition With Causal Mechanisms](https://openaccess.thecvf.com/content/CVPR2022/papers/Liu_Contextual_Debiasing_for_Visual_Recognition_With_Causal_Mechanisms_CVPR_2022_paper.pdf)
* Long-term action quality assessment
* [Likert Scoring With Grade Decoupling for Long-Term Action Assessment](https://openaccess.thecvf.com/content/CVPR2022/papers/Xu_Likert_Scoring_With_Grade_Decoupling_for_Long-Term_Action_Assessment_CVPR_2022_paper.pdf)
* 运动识别
* [Decoupling and Recoupling Spatiotemporal Representation for RGB-D-Based Motion Recognition](https://openaccess.thecvf.com/content/CVPR2022/papers/Zhou_Decoupling_and_Recoupling_Spatiotemporal_Representation_for_RGB-D-Based_Motion_Recognition_CVPR_2022_paper.pdf)
:star:[code](https://github.com/damo-cv/MotionRGBD)
* CNN
* [An Image Patch Is a Wave: Phase-Aware Vision MLP](https://arxiv.org/abs/2111.12294)
:star:[code](https://github.com/huawei-noah/Efficient-AI-Backbones/tree/master/wavemlp_pytorch)
* Volume Rendering
* [DIVeR: Real-Time and Accurate Neural Radiance Fields With Deterministic Integration for Volume Rendering](https://arxiv.org/abs/2111.10427)
:star:[code](https://github.com/lwwu2/diver-rt)
* virtual correspondences
* [Virtual Correspondence: Humans as a Cue for Extreme-View Geometry](https://openaccess.thecvf.com/content/CVPR2022/papers/Ma_Virtual_Correspondence_Humans_as_a_Cue_for_Extreme-View_Geometry_CVPR_2022_paper.pdf)
:house:[project](https://virtual-correspondence.github.io/)
* 红外测量
* [Shape From Thermal Radiation: Passive Ranging Using Multi-Spectral LWIR Measurements](https://openaccess.thecvf.com/content/CVPR2022/papers/Nagase_Shape_From_Thermal_Radiation_Passive_Ranging_Using_Multi-Spectral_LWIR_Measurements_CVPR_2022_paper.pdf)
* 4D场景捕捉
* [HSC4D: Human-Centered 4D Scene Capture in Large-Scale Indoor-Outdoor Space Using Wearable IMUs and LiDAR](https://arxiv.org/abs/2203.09215)
:star:[code](https://github.com/climbingdaily/HSC4D):house:[project](http://www.lidarhumanmotion.net/hsc4d/)
* 可变形头像
* [I M Avatar: Implicit Morphable Head Avatars From Videos](https://openaccess.thecvf.com/content/CVPR2022/papers/Zheng_I_M_Avatar_Implicit_Morphable_Head_Avatars_From_Videos_CVPR_2022_paper.pdf)
:star:[code](https://github.com/zhengyuf/IMavatar):house:[project](https://ait.ethz.ch/projects/2022/IMavatar/)
* 活动预测
* [A Hybrid Egocentric Activity Anticipation Framework via Memory-Augmented Recurrent and One-Shot Representation Forecasting](https://openaccess.thecvf.com/content/CVPR2022/papers/Liu_A_Hybrid_Egocentric_Activity_Anticipation_Framework_via_Memory-Augmented_Recurrent_and_CVPR_2022_paper.pdf)
* Mirror Detection
* [Learning Semantic Associations for Mirror Detection](https://openaccess.thecvf.com/content/CVPR2022/papers/Guan_Learning_Semantic_Associations_for_Mirror_Detection_CVPR_2022_paper.pdf)
:star:[code](https://github.com/guanhuankang/Learning-Semantic-Associations-for-Mirror-Detection)
* 双手重建
* [Interacting Attention Graph for Single Image Two-Hand Reconstruction](https://arxiv.org/abs/2203.09364)
:star:[code](https://github.com/Dw1010/IntagHand)
* Image Vectorization
* [Towards Layer-wise Image Vectorization](https://openaccess.thecvf.com/content/CVPR2022/papers/Ma_Towards_Layer-Wise_Image_Vectorization_CVPR_2022_paper.pdf)
:star:[code](https://github.com/Picsart-AI-Research/LIVE-Layerwise-Image-Vectorization)
* 行动学习
* [Set-Supervised Action Learning in Procedural Task Videos via Pairwise Order Consistency](https://openaccess.thecvf.com/content/CVPR2022/papers/Lu_Set-Supervised_Action_Learning_in_Procedural_Task_Videos_via_Pairwise_Order_CVPR_2022_paper.pdf)
:star:[code](https://github.com/ZijiaLewisLu/CVPR22-POC)
* BNN
* [PokeBNN: A Binary Pursuit of Lightweight Accuracy](https://arxiv.org/abs/2112.00133)
:star:[code](https://github.com/google/aqt)
* CNN
* [Condensing CNNs With Partial Differential Equations](https://openaccess.thecvf.com/content/CVPR2022/papers/Kag_Condensing_CNNs_With_Partial_Differential_Equations_CVPR_2022_paper.pdf)
:star:[code](https://github.com/anilkagak2/PDE_GlobalLayer)
* Place Recognition
* [TransVPR: Transformer-based place recognition with multi-level attention aggregation](https://arxiv.org/abs/2201.02001)
:open_mouth:oral
* 物体识别
* [AirObject: A Temporally Evolving Graph Embedding for Object Identification](https://arxiv.org/abs/2111.15150)
:star:[code](https://github.com/Nik-V9/AirObject)
* 边缘检测
* [EDTER: Edge Detection with Transformer](https://arxiv.org/abs/2203.08566)
:star:[code](https://github.com/MengyangPu/EDTER)
* 缺陷检测
* [Semiconductor Defect Detection by Hybrid Classical-Quantum Deep Learning](https://openaccess.thecvf.com/content/CVPR2022/papers/Yang_Semiconductor_Defect_Detection_by_Hybrid_Classical-Quantum_Deep_Learning_CVPR_2022_paper.pdf)
## Open-Set Recognition(开集识别)
* [Task-Adaptive Negative Envision for Few-Shot Open-Set Recognition](https://openaccess.thecvf.com/content/CVPR2022/papers/Huang_Task-Adaptive_Negative_Envision_for_Few-Shot_Open-Set_Recognition_CVPR_2022_paper.pdf)
:star:[code](https://github.com/shiyuanh/TANE)
## Active Learning(主动学习)
* [Active Learning for Open-Set Annotation](https://arxiv.org/abs/2201.06758)
* [Active Learning by Feature Mixing](https://arxiv.org/abs/2203.07034)
* [Towards Robust and Reproducible Active Learning Using Neural Networks](https://arxiv.org/abs/2002.09564)
:star:[code](https://github.com/PrateekMunjal/TorchAL)
## Backdoor Attacks(后门攻击)
* [DEFEAT: Deep Hidden Feature Backdoor Attacks by Imperceptible Perturbation and Latent Representation Constraints](https://openaccess.thecvf.com/content/CVPR2022/papers/Zhao_DEFEAT_Deep_Hidden_Feature_Backdoor_Attacks_by_Imperceptible_Perturbation_and_CVPR_2022_paper.pdf)
* [Better Trigger Inversion Optimization in Backdoor Scanning](https://openaccess.thecvf.com/content/CVPR2022/papers/Tao_Better_Trigger_Inversion_Optimization_in_Backdoor_Scanning_CVPR_2022_paper.pdf)
* [Towards Practical Deployment-Stage Backdoor Attack on Deep Neural Networks](https://arxiv.org/abs/2111.12965)
:open_mouth:oral:star:[code](https://github.com/Unispac/Subnet-Replacement-Attack)
## Multi-view Clustering(多视图聚类)
* [Highly-efficient Incomplete Large-scale Multi-view Clustering with Consensus Bipartite Graph](https://openaccess.thecvf.com/content/CVPR2022/papers/Wang_Highly-Efficient_Incomplete_Large-Scale_Multi-View_Clustering_With_Consensus_Bipartite_Graph_CVPR_2022_paper.pdf)
:star:[code](https://github.com/wangsiwei2010/CVPR22-IMVC-CBG)
* [Multi-Level Feature Learning for Contrastive Multi-View Clustering](https://arxiv.org/abs/2106.11193)
:star:[code](https://github.com/SubmissionsIn/MFLVC)
* [Deep Safe Multi-View Clustering: Reducing the Risk of Clustering Performance Degradation Caused by View Increase](https://openaccess.thecvf.com/content/CVPR2022/papers/Tang_Deep_Safe_Multi-View_Clustering_Reducing_the_Risk_of_Clustering_Performance_CVPR_2022_paper.pdf)
* [MPC: Multi-View Probabilistic Clustering](https://openaccess.thecvf.com/content/CVPR2022/papers/Liu_MPC_Multi-View_Probabilistic_Clustering_CVPR_2022_paper.pdf)
## Machine Translation(机器翻译)
* [VALHALLA: Visual Hallucination for Machine Translation](https://arxiv.org/abs/2206.00100)
:house:[project](http://www.svcl.ucsd.edu/projects/valhalla/)
## Object Counting(目标计数)
* [Rethinking Spatial Invariance of Convolutional Networks for Object Counting](https://arxiv.org/abs/2206.05253)
:star:[code](https://github.com/zhiqic/Rethinking-Counting)
:newspaper:[解读](https://zhuanlan.zhihu.com/p/528028523)
* [Represent, Compare, and Learn: A Similarity-Aware Framework for Class-Agnostic Counting](https://arxiv.org/abs/2203.08354)
:star:[code](https://github.com/flyinglynx/Bilinear-Matching-Network)
## computer-aided design (CAD)
* [Neural Face Identification in a 2D Wireframe Projection of a Manifold Object](https://arxiv.org/abs/2203.04229)
:star:[code](https://github.com/manycore-research/faceformer)
* [JoinABLe: Learning Bottom-up Assembly of Parametric CAD Joints](https://arxiv.org/abs/2111.12772)
:star:[code](https://github.com/AutodeskAILab/JoinABLe)
* [ROCA: Robust CAD Model Retrieval and Alignment from a Single Image](https://openaccess.thecvf.com/content/CVPR2022/papers/Gumeli_ROCA_Robust_CAD_Model_Retrieval_and_Alignment_From_a_Single_CVPR_2022_paper.pdf)
:star:[code](https://github.com/cangumeli/ROCA)
* [CADTransformer: Panoptic Symbol Spotting Transformer for CAD Drawings](https://openaccess.thecvf.com/content/CVPR2022/papers/Fan_CADTransformer_Panoptic_Symbol_Spotting_Transformer_for_CAD_Drawings_CVPR_2022_paper.pdf)
:star:[code](https://github.com/VITA-Group/CADTransformer)
* [GAT-CADNet: Graph Attention Network for Panoptic Symbol Spotting in CAD Drawings](https://openaccess.thecvf.com/content/CVPR2022/papers/Zheng_GAT-CADNet_Graph_Attention_Network_for_Panoptic_Symbol_Spotting_in_CAD_CVPR_2022_paper.pdf)
## Transfer Learning(迁移学习)
* [Revisiting Learnable Affines for Batch Norm in Few-Shot Transfer Learning](https://openaccess.thecvf.com/content/CVPR2022/papers/Yazdanpanah_Revisiting_Learnable_Affines_for_Batch_Norm_in_Few-Shot_Transfer_Learning_CVPR_2022_paper.pdf)
## Graph Matching(图匹配)
* [Graph-Context Attention Networks for Size-Varied Deep Graph Matching](https://openaccess.thecvf.com/content/CVPR2022/papers/Jiang_Graph-Context_Attention_Networks_for_Size-Varied_Deep_Graph_Matching_CVPR_2022_paper.pdf)
:star:[code](https://github.com/ZhehengJiang/GCAN)
* [Appearance and Structure Aware Robust Deep Visual Graph Matching: Attack, Defense and Beyond](https://openaccess.thecvf.com/content/CVPR2022/papers/Ren_Appearance_and_Structure_Aware_Robust_Deep_Visual_Graph_Matching_Attack_CVPR_2022_paper.pdf)
:star:[code](https://github.com/Thinklab-SJTU/RobustMatch)
## Noise Modeling(图像噪声建模)
* [Noise2NoiseFlow: Realistic Camera Noise Modeling Without Clean Images](https://arxiv.org/abs/2206.01103)
:house:[project](https://yorkucvil.github.io/Noise2NoiseFlow/)
## 60.Visual Emotion Analysis(视觉情感分析)
* [MDAN: Multi-level Dependent Attention Network for Visual Emotion Analysis](https://arxiv.org/abs/2203.13443)
## 59.动画
* [APES: Articulated Part Extraction From Sprite Sheets](https://openaccess.thecvf.com/content/CVPR2022/papers/Xu_APES_Articulated_Part_Extraction_From_Sprite_Sheets_CVPR_2022_paper.pdf)
:house:[project](https://zhan-xu.github.io/parts/)
* [BANMo: Building Animatable 3D Neural Models From Many Casual Videos](https://arxiv.org/abs/2112.12761)
:open_mouth:oral:house:[project](https://banmo-www.github.io/)
* [Neural Head Avatars From Monocular RGB Videos](https://arxiv.org/abs/2112.01554)
:star:[code](https://github.com/philgras/neural-head-avatars):house:[project](https://philgras.github.io/neural_head_avatars/neural_head_avatars.html)
* [FLAG: Flow-Based 3D Avatar Generation From Sparse Observations](https://arxiv.org/abs/2203.05789)
:house:[project](https://microsoft.github.io/flag/)
* 图像动画
* [Thin-Plate Spline Motion Model for Image Animation](https://arxiv.org/abs/2203.14367)
:star:[code](https://github.com/yoyo-nb/Thin-Plate-Spline-Motion-Model)
* 人物动画
* [Structured Local Radiance Fields for Human Avatar Modeling](https://arxiv.org/abs/2203.14478)
* 3D character animation(三维角色动画)
* 皮肤预测
* [SkinningNet: Two-Stream Graph Convolutional Neural Network for Skinning Prediction of Synthetic Characters](https://arxiv.org/abs/2203.04746)
:house:[project](https://imatge-upc.github.io/skinningnet/)
* 3D 舞蹈生成
* [Bailando: 3D Dance Generation by Actor-Critic GPT with Choreographic Memory](https://arxiv.org/abs/2203.13055)
:star:[code](https://github.com/lisiyao21/Bailando/)
* [A Brand New Dance Partner: Music-Conditioned Pluralistic Dancing Controlled by Multiple Dance Genres](https://openaccess.thecvf.com/content/CVPR2022/papers/Kim_A_Brand_New_Dance_Partner_Music-Conditioned_Pluralistic_Dancing_Controlled_by_CVPR_2022_paper.pdf)
* 静止图像到动画
* [Controllable Animation of Fluid Elements in Still Images](https://arxiv.org/abs/2112.03051)
:house:[project](https://controllable-cinemagraphs.github.io/)
* 3D human avatars
* [gDNA: Towards Generative Detailed Neural Avatars](https://arxiv.org/abs/2201.04123)
:star:[code](https://github.com/xuchen-ethz/gdna):house:[project](https://xuchen-ethz.github.io/gdna/)
## 58.Neural rendering(神经渲染)
* [Learning Motion-Dependent Appearance for High-Fidelity Rendering of Dynamic Humans from a Single Camera](https://arxiv.org/abs/2203.12780)
* [IRON: Inverse Rendering by Optimizing Neural SDFs and Materials from Photometric Images](https://arxiv.org/abs/2204.02232)
:open_mouth:oral:house:[project](https://kai-46.github.io/IRON-website/)
* [SqueezeNeRF: Further factorized FastNeRF for memory-efficient inference](https://arxiv.org/abs/2204.02585)
* [Direct Voxel Grid Optimization: Super-fast Convergence for Radiance Fields Reconstruction](https://arxiv.org/abs/2111.11215)
:star:[code](https://github.com/sunset1995/DirectVoxGO)
* [Modeling Indirect Illumination for Inverse Rendering](https://arxiv.org/abs/2204.06837)
:star:[code](https://github.com/zju3dv/invrender):house:[project](https://zju3dv.github.io/invrender/)
* [GenDR: A Generalized Differentiable Renderer](https://arxiv.org/abs/2204.13845)
:star:[code](https://github.com/Felix-Petersen/gendr)
泛化可微渲染器
* [CLIP-NeRF: Text-and-Image Driven Manipulation of Neural Radiance Fields](https://arxiv.org/abs/2112.05139)
:star:[code](https://github.com/cassiePython/CLIPNeRF):house:[project](https://cassiepython.github.io/clipnerf/)
* [NeRF-Editing: Geometry Editing of Neural Radiance Fields](https://arxiv.org/abs/2205.04978)
* [AR-NeRF: Unsupervised Learning of Depth and Defocus Effects from Natural Images with Aperture Rendering Neural Radiance Fields](https://arxiv.org/abs/2206.06100)
:house:[project](https://www.kecl.ntt.co.jp/people/kaneko.takuhiro/projects/ar-nerf/)
* [Neural Rays for Occlusion-Aware Image-Based Rendering](https://arxiv.org/abs/2107.13421)
:star:[code](https://github.com/liuyuan-pal/NeuRay):house:[project](https://liuyuan-pal.github.io/NeuRay/)
* [EfficientNeRF Efficient Neural Radiance Fields](https://arxiv.org/abs/2206.00878)
:star:[code](https://github.com/dvlab-research/EfficientNeRF)
* [CoNeRF: Controllable Neural Radiance Fields](https://arxiv.org/abs/2112.01983)
:star:[code](https://github.com/kacperkan/conerf):house:[project](https://conerf.github.io/)
* [Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields](https://arxiv.org/abs/2111.12077)
:house:[project](https://jonbarron.info/mipnerf360/)
* [Hallucinated Neural Radiance Fields in the Wild](https://arxiv.org/abs/2111.15246)
:star:[code](https://github.com/rover-xingyu/Ha-NeRF):house:[project](https://rover-xingyu.github.io/Ha-NeRF/)
* [HumanNeRF: Free-viewpoint Rendering of Moving People from Monocular Video](https://arxiv.org/abs/2201.04127)
:open_mouth:oral:star:[code](https://github.com/chungyiweng/humannerf):house:[project](https://grail.cs.washington.edu/projects/humannerf/):tv:[video](https://youtu.be/GM-RoZEymmw)
* [Ref-NeRF: Structured View-Dependent Appearance for Neural Radiance Fields](https://openaccess.thecvf.com/content/CVPR2022/papers/Verbin_Ref-NeRF_Structured_View-Dependent_Appearance_for_Neural_Radiance_Fields_CVPR_2022_paper.pdf)
* [Deblur-NeRF: Neural Radiance Fields From Blurry Images](https://openaccess.thecvf.com/content/CVPR2022/papers/Ma_Deblur-NeRF_Neural_Radiance_Fields_From_Blurry_Images_CVPR_2022_paper.pdf)
:star:[code](https://github.com/limacv/Deblur-NeRF):house:[project](https://limacv.github.io/deblurnerf/)
* [NeRFReN: Neural Radiance Fields With Reflections](https://arxiv.org/abs/2111.15234)
:house:[project](https://bennyguo.github.io/nerfren/)
* [Depth-Supervised NeRF: Fewer Views and Faster Training for Free](https://arxiv.org/abs/2107.02791)
:star:[code](https://github.com/dunbar12138/DSNeRF):house:[project](http://www.cs.cmu.edu/~dsnerf/)
* [Dense Depth Priors for Neural Radiance Fields From Sparse Input Views](https://arxiv.org/abs/2112.03288)
:star:[code](https://github.com/barbararoessle/dense_depth_priors_nerf):house:[project](https://barbararoessle.github.io/dense_depth_priors_nerf/):tv:[video](https://www.youtube.com/watch?v=zzkvvdcvksc)
* [Light Field Neural Rendering](https://arxiv.org/abs/2112.09687)
:star:[code](https://github.com/google-research/google-research/tree/master/light_field_neural_rendering):house:[project](https://light-field-neural-rendering.github.io/)
* [InfoNeRF: Ray Entropy Minimization for Few-Shot Neural Volume Rendering](https://arxiv.org/abs/2112.15399)
:star:[code](https://github.com/mjmjeong/InfoNeRF):house:[project](https://cv.snu.ac.kr/research/InfoNeRF/)
* [BokehMe: When Neural Rendering Meets Classical Rendering](https://openaccess.thecvf.com/content/CVPR2022/papers/Peng_BokehMe_When_Neural_Rendering_Meets_Classical_Rendering_CVPR_2022_paper.pdf)
:open_mouth:oral:star:[code](https://github.com/JuewenPeng/BokehMe)
* [Plenoxels: Radiance Fields Without Neural Networks](https://openaccess.thecvf.com/content/CVPR2022/papers/Fridovich-Keil_Plenoxels_Radiance_Fields_Without_Neural_Networks_CVPR_2022_paper.pdf)
:star:[code](https://github.com/sxyu/svox2):house:[project](https://alexyu.net/plenoxels/)
* [HDR-NeRF: High Dynamic Range Neural Radiance Fields](https://openaccess.thecvf.com/content/CVPR2022/papers/Huang_HDR-NeRF_High_Dynamic_Range_Neural_Radiance_Fields_CVPR_2022_paper.pdf)
* [Urban Radiance Fields](https://arxiv.org/abs/2111.14643)
:house:[project](https://urban-radiance-fields.github.io/)
* [Aug-NeRF: Training Stronger Neural Radiance Fields With Triple-Level Physically-Grounded Augmentations](https://openaccess.thecvf.com/content/CVPR2022/papers/Chen_Aug-NeRF_Training_Stronger_Neural_Radiance_Fields_With_Triple-Level_Physically-Grounded_Augmentations_CVPR_2022_paper.pdf)
:star:[code](https://github.com/VITA-Group/Aug-NeRF)
* [Fourier PlenOctrees for Dynamic Radiance Field Rendering in Real-Time](https://arxiv.org/abs/2202.08614)
:star:[code](https://github.com/aoliao12138/FPO):house:[project](https://aoliao12138.github.io/FPO/)
* [Point-NeRF: Point-Based Neural Radiance Fields](https://openaccess.thecvf.com/content/CVPR2022/papers/Xu_Point-NeRF_Point-Based_Neural_Radiance_Fields_CVPR_2022_paper.pdf)
* [HumanNeRF: Efficiently Generated Human Radiance Field From Sparse Inputs](https://arxiv.org/abs/2112.02789)
:house:[project](https://zhaofuq.github.io/humannerf/)
* [Ray Priors through Reprojection: Improving Neural Radiance Fields for Novel View Extrapolation](https://arxiv.org/abs/2205.05922)
## 57.Gaze Estimation(视线估计)
* [GazeOnce: Real-Time Multi-Person Gaze Estimation](https://arxiv.org/abs/2204.09480)
* [Contrastive Regression for Domain Adaptation on Gaze Estimation](https://openaccess.thecvf.com/content/CVPR2022/papers/Wang_Contrastive_Regression_for_Domain_Adaptation_on_Gaze_Estimation_CVPR_2022_paper.pdf)
* [Generalizing Gaze Estimation With Rotation Consistency](https://openaccess.thecvf.com/content/CVPR2022/papers/Bao_Generalizing_Gaze_Estimation_With_Rotation_Consistency_CVPR_2022_paper.pdf)
* [GaTector: A Unified Framework for Gaze Object Prediction](https://arxiv.org/abs/2112.03549)
* [Dynamic 3D Gaze From Afar: Deep Gaze Estimation From Temporal Eye-Head-Body Coordination](https://openaccess.thecvf.com/content/CVPR2022/papers/Nonaka_Dynamic_3D_Gaze_From_Afar_Deep_Gaze_Estimation_From_Temporal_CVPR_2022_paper.pdf)
:house:[project](https://vision.ist.i.kyoto-u.ac.jp/)
## 56.Sound
* [Finding Fallen Objects via Asynchronous Audio-Visual Integration](https://openaccess.thecvf.com/content/CVPR2022/papers/Gan_Finding_Fallen_Objects_via_Asynchronous_Audio-Visual_Integration_CVPR_2022_paper.pdf)
:house:[project](http://fallen-object.csail.mit.edu/)
* [Weakly Paired Associative Learning for Sound and Image Representations via Bimodal Associative Memory](https://openaccess.thecvf.com/content/CVPR2022/papers/Lee_Weakly_Paired_Associative_Learning_for_Sound_and_Image_Representations_via_CVPR_2022_paper.pdf)
* [MERLOT Reserve: Neural Script Knowledge Through Vision and Language and Sound](https://arxiv.org/abs/2201.02639)
:star:[code](https://github.com/rowanz/merlot_reserve):house:[project](https://rowanzellers.com/merlotreserve/)
* [Visual Acoustic Matching](https://arxiv.org/abs/2202.06875)
:open_mouth:oral:house:[project](https://vision.cs.utexas.edu/projects/visual-acoustic-matching/)
* 声源定位
* [Self-Supervised Predictive Learning: A Negative-Free Method for Sound Source Localization in Visual Scenes](https://arxiv.org/abs/2203.13412)
:star:[code](https://github.com/zjsong/SSPL)
* [Mix and Localize: Localizing Sound Sources in Mixtures](https://openaccess.thecvf.com/content/CVPR2022/papers/Hu_Mix_and_Localize_Localizing_Sound_Sources_in_Mixtures_CVPR_2022_paper.pdf)
* [A Proposal-Based Paradigm for Self-Supervised Sound Source Localization in Videos](https://openaccess.thecvf.com/content/CVPR2022/papers/Xuan_A_Proposal-Based_Paradigm_for_Self-Supervised_Sound_Source_Localization_in_Videos_CVPR_2022_paper.pdf)
* 音频配对
* [It's Time for Artistic Correspondence in Music and Video](https://openaccess.thecvf.com/content/CVPR2022/papers/Suris_Its_Time_for_Artistic_Correspondence_in_Music_and_Video_CVPR_2022_paper.pdf)
:house:[project](https://musicforvideo.cs.columbia.edu/)
* 语音克隆
* [V2C: Visual Voice Cloning](https://arxiv.org/abs/2111.12890)
:star:[code](https://github.com/chenqi008/V2C)
* 视听语音增强
* [Audio-Visual Speech Codecs: Rethinking Audio-Visual Speech Enhancement by Re-Synthesis](https://openaccess.thecvf.com/content/CVPR2022/papers/Yang_Audio-Visual_Speech_Codecs_Rethinking_Audio-Visual_Speech_Enhancement_by_Re-Synthesis_CVPR_2022_paper.pdf)
:tv:[video](https://github.com/facebookresearch/facestar/releases/download/paper_materials/video.mp4)
* 文本转语音
* [More Than Words: In-the-Wild Visually-Driven Prosody for Text-to-Speech](https://arxiv.org/abs/2111.10139)
:star:[code](https://google-research.github.io/lingvo-lab/vdtts/)
* 语音转人脸图像
* [Cross-Modal Perceptionist: Can Face Geometry be Gleaned from Voices?](https://arxiv.org/abs/2203.09824)
:star:[code](https://github.com/choyingw/Cross-Modal-Perceptionist):house:[project](https://choyingw.github.io/works/Voice2Mesh/index.html)
* 语音分离
* [Reading To Listen at the Cocktail Party: Multi-Modal Speech Separation](https://openaccess.thecvf.com/content/CVPR2022/papers/Rahimi_Reading_To_Listen_at_the_Cocktail_Party_Multi-Modal_Speech_Separation_CVPR_2022_paper.pdf)
:house:[project](https://www.robots.ox.ac.uk/~vgg/research/voiceformer/)
* 语音手势生成
* [Low-Resource Adaptation for Personalized Co-Speech Gesture Generation](https://openaccess.thecvf.com/content/CVPR2022/papers/Ahuja_Low-Resource_Adaptation_for_Personalized_Co-Speech_Gesture_Generation_CVPR_2022_paper.pdf)
:house:[project](https://chahuja.com/diffgan/)
* 扬声器定位
* [Egocentric Deep Multi-Channel Audio-Visual Active Speaker Localization](https://arxiv.org/abs/2201.01928)
* 语音手势生成
* [SEEG: Semantic Energized Co-Speech Gesture Generation](https://openaccess.thecvf.com/content/CVPR2022/papers/Liang_SEEG_Semantic_Energized_Co-Speech_Gesture_Generation_CVPR_2022_paper.pdf)
:star:[code](https://github.com/akira-l/SEEG)
## 55.Novel View Synthesis(视图合成)
* [NPBG++: Accelerating Neural Point-Based Graphics](https://arxiv.org/abs/2203.)
:house:[project](htt.io/npbgpp/)
* [Scene Representation Transformer: Geometry-Free Novel View Synthesis Through Set-Latent Scene Representations](https://arxiv.org/abs/2111.13152)
:house:[project](https://srt-paper.github.io/)
* [AutoRF: Learning 3D Object Radiance Fields from Single View Observations](https://arxiv.org/abs/2204.03593)
:house:[project](https://sirwyver.github.io/AutoRF/)
* [NeurMiPs: Neural Mixture of Planar Experts for View Synthesis](https://arxiv.org/abs/2204.13696)
:star:[code](https://github.com/zhihao-lin/neurmips):house:[project](https://zhihao-lin.github.io/neurmips/):tv:[video](https://youtu.be/PV1dCTWL5Oo):newspaper:[解读](https://zhuanlan.zhihu.com/p/507053208)
* [GeoNeRF: Generalizing NeRF with Geometry Priors](https://arxiv.org/abs/2111.13539)
:star:[code](https://www.idiap.ch/paper/geonerf):house:[project](https://www.idiap.ch/paper/geonerf/):tv:[video](https://www.youtube.com/watch?v=-jNBsG3IP54)
* [FWD: R eal-Time Novel View Synthesis With Forward Warping and Depth](https://openaccess.thecvf.com/content/CVPR2022/papers/Cao_FWD_Real-Time_Novel_View_Synthesis_With_Forward_Warping_and_Depth_CVPR_2022_paper.pdf)
:star:[code](https://github.com/Caoang327/fwd_code)
* [Block-NeRF: Scalable Large Scene Neural View Synthesis](https://openaccess.thecvf.com/content/CVPR2022/papers/Tancik_Block-NeRF_Scalable_Large_Scene_Neural_View_Synthesis_CVPR_2022_paper.pdf)
* [Boosting View Synthesis With Residual Transfer](https://openaccess.thecvf.com/content/CVPR2022/papers/Rong_Boosting_View_Synthesis_With_Residual_Transfer_CVPR_2022_paper.pdf)
:star:[code](https://github.com/facebookresearch/boosting_view_synth):house:[project](https://boosting-view-synth.github.io/)
* [NeRF in the Dark: High Dynamic Range View Synthesis From Noisy Raw Images](https://openaccess.thecvf.com/content/CVPR2022/papers/Mildenhall_NeRF_in_the_Dark_High_Dynamic_Range_View_Synthesis_From_CVPR_2022_paper.pdf)
* [RegNeRF: Regularizing Neural Radiance Fields for View Synthesis from Sparse Inputs](https://arxiv.org/abs/2112.00724)
:open_mouth:oral:star:[code](https://github.com/google-research/google-research/tree/master/regnerf):house:[project](https://m-niemeyer.github.io/regnerf/index.html):tv:[video](https://youtu.be/QyyyvA4-Kwc)
* 视图连接
* [Connecting the Complementary-View Videos: Joint Camera Identification and Subject Association](https://openaccess.thecvf.com/content/CVPR2022/papers/Han_Connecting_the_Complementary-View_Videos_Joint_Camera_Identification_and_Subject_Association_CVPR_2022_paper.pdf)
:star:[code](https://github.com/RuizeHan/DMHA)
## 54.Dataset(数据集)
* [ObjectFolder 2.0: A Multisensory Object Dataset for Sim2Real Transfer](https://arxiv.org/abs/2204.02389)
:star:[code](https://github.com/rhgao/ObjectFolder):house:[project](https://ai.stanford.edu/~rhgao/objectfolder2.0/):newspaper:[粗解](https://zhuanlan.zhihu.com/p/493615566)
* [Assembly101: A Large-Scale Multi-View Video Dataset for Understanding Procedural Activities](https://arxiv.org/abs/2203.14712)
:star:[code](https://github.com/assembly-101?tab=repositories):house:[project](https://assembly-101.github.io/)
* [3MASSIV: Multilingual, Multimodal and Multi-Aspect dataset of Social Media Short Videos](https://arxiv.org/abs/2203.14456)
:sunflower:[dataset](https://sharechat.com/research/3massiv)
* [Hephaestus: A large scale multitask dataset towards InSAR understanding](https://arxiv.org/abs/2204.09435)
* [SmartPortraits: Depth Powered Handheld Smartphone Dataset of Human Portraits for State Estimation, Reconstruction and Synthesis](https://arxiv.org/abs/2204.10211)
:sunflower:[dataset](https://mobileroboticsskoltech.github.io/SmartPortraits/)
* [AKB-48: A Real-World Articulated Object Knowledge Base](https://arxiv.org/abs/2202.08432)
:star:[code](https://liuliu66.github.io/articulationobjects/)
:newspaper:[粗解](https://news.sjtu.edu.cn/jdzh/20220330/169456.html)
* [Primitive3D: 3D Object Dataset Synthesis from Randomly Assembled Primitives](https://arxiv.org/abs/2205.12627)
* [ZeroWaste Dataset: Towards Deformable Object Segmentation in Cluttered Scenes](https://arxiv.org/abs/2106.02740)
:star:[code](https://github.com/dbash/zerowaste/):house:[project](http://ai.bu.edu/zerowaste/)
* [ETHSeg: An Amodel Instance Segmentation Network and a Real-World Dataset for X-Ray Waste Inspection](https://openaccess.thecvf.com/content/CVPR2022/papers/Qiu_ETHSeg_An_Amodel_Instance_Segmentation_Network_and_a_Real-World_Dataset_CVPR_2022_paper.pdf)
一个Amodel实例分割网络和一个用于X射线废物检查的真实数据集
* [MAD: A Scalable Dataset for Language Grounding in Videos From Movie Audio Descriptions](https://openaccess.thecvf.com/content/CVPR2022/papers/Soldan_MAD_A_Scalable_Dataset_for_Language_Grounding_in_Videos_From_CVPR_2022_paper.pdf)
:sunflower:[dataset](https://github.com/Soldelli/MAD)
一个可扩展的数据集，用于从电影音频描述中获得视频的Language Grounding
* [DiLiGenT102: A Photometric Stereo Benchmark Dataset With Controlled Shape and Material Variation](https://openaccess.thecvf.com/content/CVPR2022/papers/Ren_DiLiGenT102_A_Photometric_Stereo_Benchmark_Dataset_With_Controlled_Shape_and_CVPR_2022_paper.pdf)
:sunflower:[dataset](https://photometricstereo.github.io/)
具有受控形状和材料变化的光度测量立体基准数据集
* [DAD-3DHeads: A Large-Scale Dense, Accurate and Diverse Dataset for 3D Head Alignment From a Single Image](https://openaccess.thecvf.com/content/CVPR2022/papers/Martyniuk_DAD-3DHeads_A_Large-Scale_Dense_Accurate_and_Diverse_Dataset_for_3D_CVPR_2022_paper.pdf)
:sunflower:[dataset](https://p.farm/research/dad-3dheads)
一个大规模的密集、准确和多样化的数据集，用于从单一图像中进行三维头部对准
* [Rope3D: The Roadside Perception Dataset for Autonomous Driving and Monocular 3D Object Detection Task](https://openaccess.thecvf.com/content/CVPR2022/papers/Ye_Rope3D_The_Roadside_Perception_Dataset_for_Autonomous_Driving_and_Monocular_CVPR_2022_paper.pdf)
:sunflower:[dataset](https://thudair.baai.ac.cn/rope)
用于自主驾驶和单眼3D物体检测任务的路边感知数据集
* [Ithaca365: Dataset and Driving Perception Under Repeated and Challenging Weather Conditions](https://openaccess.thecvf.com/content/CVPR2022/papers/Diaz-Ruiz_Ithaca365_Dataset_and_Driving_Perception_Under_Repeated_and_Challenging_Weather_CVPR_2022_paper.pdf)
:sunflower:[dataset](https://ithaca365.mae.cornell.edu/)
* [Open Challenges in Deep Stereo: The Booster Dataset](https://openaccess.thecvf.com/content/CVPR2022/papers/Ramirez_Open_Challenges_in_Deep_Stereo_The_Booster_Dataset_CVPR_2022_paper.pdf)
:sunflower:[dataset](https://cvlab-unibo.github.io/booster-web/)
* [RGB-Multispectral Matching: Dataset, Learning Methodology, Evaluation](https://openaccess.thecvf.com/content/CVPR2022/papers/Tosi_RGB-Multispectral_Matching_Dataset_Learning_Methodology_Evaluation_CVPR_2022_paper.pdf)
:house:[project](https://cvlab-unibo.github.io/rgb-ms-web/)
* 卫星数据集
* [DynamicEarthNet: Daily Multi-Spectral Satellite Dataset for Semantic Change Segmentation](https://arxiv.org/abs/2203.12560)
* 动物行为理解数据集
* [Animal Kingdom: A Large and Diverse Dataset for Animal Behavior Understanding](https://arxiv.org/abs/2204.08129)
:open_mouth:oral:house:[project](https://sutdcv.github.io/Animal-Kingdom/):sunflower:[dataset](https://github.com/SUTDCV/Animal-Kingdom)
* 数据集(森林监测)
* [The Auto Arborist Dataset: A Large-Scale Benchmark for Multiview Urban Forest Monitoring Under Domain Shift](https://openaccess.thecvf.com/content/CVPR2022/papers/Beery_The_Auto_Arborist_Dataset_A_Large-Scale_Benchmark_for_Multiview_Urban_CVPR_2022_paper.pdf)
:sunflower:[dataset](https://google.github.io/auto-arborist/)
* 3D目标理解
* [ABO: Dataset and Benchmarks for Real-World 3D Object Understanding](https://arxiv.org/abs/2110.06199)
:sunflower:[dataset](https://amazon-berkeley-objects.s3.amazonaws.com/index.html)
* 数据集(AutoMine)
* [AutoMine: An Unmanned Mine Dataset](https://openaccess.thecvf.com/content/CVPR2022/papers/Li_AutoMine_An_Unmanned_Mine_Dataset_CVPR_2022_paper.pdf)
:sunflower:[dataset](https://automine.cc/)
* 数据集(人脸表情识别)
* [FERV39k: A Large-Scale Multi-Scene Dataset for Facial Expression Recognition in Videos](https://arxiv.org/abs/2203.09463)
:star:[code](https://github.com/wangyanckxx/FERV39k)
* 数据集(手势识别)
* [LD-ConGR: A Large RGB-D Video Dataset for Long-Distance Continuous Gesture Recognition](https://openaccess.thecvf.com/content/CVPR2022/papers/Liu_LD-ConGR_A_Large_RGB-D_Video_Dataset_for_Long-Distance_Continuous_Gesture_CVPR_2022_paper.pdf)
:star:[code](https://github.com/Diananini/LD-ConGR-CVPR2022)
* 数据集(谷物识别)
* [GrainSpace: A Large-Scale Dataset for Fine-Grained and Domain-Adaptive Recognition of Cereal Grains](https://arxiv.org/abs/2203.05306)
:sunflower:[dataset](https://github.com/hellodfan/GrainSpace)
* 数据集(用于空间-时间行动、社会团体和活动检测)
* [JRDB-Act: A Large-Scale Dataset for Spatio-Temporal Action, Social Group and Activity Detection](https://arxiv.org/pdf/2106.08827.pdf))
:sunflower:[dataset](https://jrdb.erc.monash.edu/)
## 53.Sign Language Translation(手语翻译)
* [A Simple Multi-Modality Transfer Learning Baseline for Sign Language Translation](https://arxiv.org/abs/2203.04287)
* [Signing at Scale: Learning to Co-Articulate Signs for Large-Scale Photo-Realistic Sign Language Production](https://arxiv.org/abs/2203.15354)
* [MLSLT: Towards Multilingual Sign Language Translation](https://openaccess.thecvf.com/content/CVPR2022/papers/Yin_MLSLT_Towards_Multilingual_Sign_Language_Translation_CVPR_2022_paper.pdf)
:house:[project](https://mlslt.github.io/)
* 手语识别
* [C2SLR: Consistency-Enhanced Continuous Sign Language Recognition](https://openaccess.thecvf.com/content/CVPR2022/papers/Zuo_C2SLR_Consistency-Enhanced_Continuous_Sign_Language_Recognition_CVPR_2022_paper.pdf)
## 52.Human Motion Forecasting(人体运动预测)
* [Motron: Multimodal Probabilistic Human Motion Forecasting](https://arxiv.org/abs/2203.04132)
:star:[code](https://github.com/TUM-AAS/motron-cvpr22)
* [Progressively Generating Better Initial Guesses Towards Next Stages for High-Quality Human Motion Prediction](https://arxiv.org/abs/2203.16051)
:star:[code](https://github.com/705062791/PGBIG)
* [Spatio-Temporal Gating-Adjacency GCN for Human Motion Prediction](https://openaccess.thecvf.com/content/CVPR2022/papers/Zhong_Spatio-Temporal_Gating-Adjacency_GCN_for_Human_Motion_Prediction_CVPR_2022_paper.pdf)
* [MotionAug: Augmentation With Physical Correction for Human Motion Prediction](https://arxiv.org/abs/2203.09116)
:star:[code](https://github.com/meaten/MotionAug)
* [Future Transformer for Long-term Action Anticipation](https://arxiv.org/abs/2205.14022)
:star:[code](https://github.com/gongda0e/FUTR):house:[project](http://cvlab.postech.ac.kr/research/FUTR/)
* [Weakly-Supervised Action Transition Learning for Stochastic Human Motion Prediction](https://arxiv.org/abs/2205.15608)
:star:[code](https://github.com/wei-mao-2019/WAT)
* [Multi-Objective Diverse Human Motion Prediction With Knowledge Distillation](https://openaccess.thecvf.com/content/CVPR2022/papers/Ma_Multi-Objective_Diverse_Human_Motion_Prediction_With_Knowledge_Distillation_CVPR_2022_paper.pdf)
* [BE-STI: Spatial-Temporal Integrated Network for Class-Agnostic Motion Prediction With Bidirectional Enhancement](https://openaccess.thecvf.com/content/CVPR2022/papers/Wang_BE-STI_Spatial-Temporal_Integrated_Network_for_Class-Agnostic_Motion_Prediction_With_Bidirectional_CVPR_2022_paper.pdf)
:star:[code](https://github.com/be-sti/be-sti)
* [Multi-Person Extreme Motion Prediction](https://openaccess.thecvf.com/content/CVPR2022/papers/Guo_Multi-Person_Extreme_Motion_Prediction_CVPR_2022_paper.pdf)
## 51.光学、几何、光场成像
* [Compressive Single-Photon 3D Cameras](https://openaccess.thecvf.com/content/CVPR2022/papers/Gutierrez-Barragan_Compressive_Single-Photon_3D_Cameras_CVPR_2022_paper.pdf)
* [Fisher Information Guidance for Learned Time-of-Flight Imaging](https://openaccess.thecvf.com/content/CVPR2022/papers/Li_Fisher_Information_Guidance_for_Learned_Time-of-Flight_Imaging_CVPR_2022_paper.pdf)
* Light Field(光场)
* [Occlusion-Aware Cost Constructor for Light Field Depth Estimation](https://arxiv.org/abs/2203.01576)
:star:[code](https://github.com/YingqianWang/OACC-Net):newspaper:[粗解](https://zhuanlan.zhihu.com/p/475067096)
* [Neural Point Light Fields](https://arxiv.org/abs/2112.01473)
:star:[code](https://github.com/princeton-computational-imaging/neural-point-light-fields):house:[project](https://light.princeton.edu/publication/neural-point-light-fields/)
* [Acquiring a Dynamic Light Field Through a Single-Shot Coded Image](https://arxiv.org/abs/2204.12089)
* [Learning Neural Light Fields With Ray-Space Embedding](https://openaccess.thecvf.com/content/CVPR2022/papers/Attal_Learning_Neural_Light_Fields_With_Ray-Space_Embedding_CVPR_2022_paper.pdf)
:star:[code](https://github.com/facebookresearch/neural-light-fields):house:[project](https://neural-light-fields.github.io/)
* 深度重建
* [Deep Hyperspectral-Depth Reconstruction Using Single Color-Dot Projection](https://arxiv.org/abs/2204.03929)
:star:[code](https://github.com/GrayMask/DHD):house:[project](http://www.ok.sc.e.titech.ac.jp/res/DHD/):tv:[video](https://youtu.be/LgGDqDf034g)
* 快门校正
* [Learning Adaptive Warping for Real-World Rolling Shutter Correction](https://arxiv.org/abs/2204.13886)
:star:[code](https://github.com/ljzycmd/BSRSC)
* 热红外成像
* [Infrared Invisible Clothing:Hiding from Infrared Detectors at Multiple Angles in Real World](https://arxiv.org/abs/2205.05909)
:open_mouth:oral
* 相机姿势估计
* [DiffPoseNet: Direct Differentiable Camera Pose Estimation](https://openaccess.thecvf.com/content/CVPR2022/papers/Parameshwara_DiffPoseNet_Direct_Differentiable_Camera_Pose_Estimation_CVPR_2022_paper.pdf)
* 相机重定位
* [SceneSqueezer: Learning to Compress Scene for Camera Relocalization](https://openaccess.thecvf.com/content/CVPR2022/papers/Yang_SceneSqueezer_Learning_To_Compress_Scene_for_Camera_Relocalization_CVPR_2022_paper.pdf)
:open_mouth:oral
* 成像
* [Adaptive Gating for Single-Photon 3D Imaging](https://arxiv.org/abs/2111.15047)
* [All-photon Polarimetric Time-of-Flight Imaging](https://arxiv.org/abs/2112.09278)
* [Computing Wasserstein-p Distance Between Images With Linear Cost](https://openaccess.thecvf.com/content/CVPR2022/papers/Chen_Computing_Wasserstein-p_Distance_Between_Images_With_Linear_Cost_CVPR_2022_paper.pdf)
:star:[code](https://github.com/ucascnic/CudaOT)
* 光学
* [Quantization-aware Deep Optics for Diffractive Snapshot Hyperspectral Imaging](https://openaccess.thecvf.com/content/CVPR2022/papers/Li_Quantization-Aware_Deep_Optics_for_Diffractive_Snapshot_Hyperspectral_Imaging_CVPR_2022_paper.pdf)
:star:[code](https://github.com/wang-lizhi/QuantizationAwareDeepOptics)
* [Dual-Shutter Optical Vibration Sensing](https://openaccess.thecvf.com/content/CVPR2022/papers/Sheinin_Dual-Shutter_Optical_Vibration_Sensing_CVPR_2022_paper.pdf)
* 相机姿势
* [Camera Pose Estimation Using Implicit Distortion Models](https://openaccess.thecvf.com/content/CVPR2022/papers/Pan_Camera_Pose_Estimation_Using_Implicit_Distortion_Models_CVPR_2022_paper.pdf)
* 相机成像
* [Learning to Zoom Inside Camera Imaging Pipeline](https://openaccess.thecvf.com/content/CVPR2022/papers/Tang_Learning_To_Zoom_Inside_Camera_Imaging_Pipeline_CVPR_2022_paper.pdf)
* 相机定位
* [Learning To Detect Scene Landmarks for Camera Localization](https://openaccess.thecvf.com/content/CVPR2022/papers/Do_Learning_To_Detect_Scene_Landmarks_for_Camera_Localization_CVPR_2022_paper.pdf)
:star:[code](https://github.com/microsoft/SceneLandmarkLocalization)
* 孔径成像
* [Synthetic Aperture Imaging With Events and Frames](https://openaccess.thecvf.com/content/CVPR2022/papers/Liao_Synthetic_Aperture_Imaging_With_Events_and_Frames_CVPR_2022_paper.pdf)
:star:[code](https://github.com/smjsc/EF-SAI)
* 高光谱成像
* [Real-Time Hyperspectral Imaging in Hardware via Trained Metasurface Encoders](https://openaccess.thecvf.com/content/CVPR2022/papers/Makarenko_Real-Time_Hyperspectral_Imaging_in_Hardware_via_Trained_Metasurface_Encoders_CVPR_2022_paper.pdf)
:star:[code](https://github.com/makamoa/hyplex)
## 50.Anomaly Detection(异常检测)
* [Catching Both Gray and Black Swans: Open-set Supervised Anomaly Detection](https://arxiv.org/abs/2203.14506)
:star:[code](https://github.com/choubo/DRA)
* [Self-Supervised Predictive Convolutional Attentive Block for Anomaly Detection](https://arxiv.org/abs/2111.09099)
:star:[code](https://github.com/ristea/sspcab)
* [Anomaly Detection via Reverse Distillation From One-Class Embedding](https://arxiv.org/abs/2201.10703)
* [Towards Total Recall in Industrial Anomaly Detection](https://arxiv.org/abs/2106.08265)
:star:[code](https://github.com/amazon-research/patchcore-inspection)
* 离群点检测
* [Robust outlier detection by de-biasing VAE likelihoods](https://arxiv.org/abs/2108.08760)
## 49.Image Geo-localization(图像地理定位)
* [TransGeo: Transformer Is All You Need for Cross-view Image Geo-localization](https://arxiv.org/abs/2204.00097)
:star:[code](https://github.com/Jeff-Zilence/TransGeo2022)
* 视觉地理定位
* [Rethinking Visual Geo-localization for Large-Scale Applications](https://arxiv.org/abs/2204.02287)
:star:[code](https://github.com/gmberton/CosPlace)
* [Deep Visual Geo-localization Benchmark](https://arxiv.org/abs/2204.03444)
:open_mouth:oral:house:[project](https://deep-vg-bench.herokuapp.com/)
* 轨迹重建
* [MonoTrack: Shuttle trajectory reconstruction from monocular badminton video](https://arxiv.org/abs/2204.01899)
## 48.Visual Grounding
* [Multi-View Transformer for 3D Visual Grounding](https://arxiv.org/abs/2204.02174)
:star:[code](https://github.com/sega-hsj/MVT-3DVG)
* [Improving Visual Grounding with Visual-Linguistic Verification and Iterative Reasoning](https://arxiv.org/abs/2205.00272)
:star:[code](https://github.com/yangli18/VLTVG)
视觉定位，通过自然语言定位目标位置 （很有意思的研究）
* [Shifting More Attention to Visual Backbone: Query-Modulated Refinement Networks for End-to-End Visual Grounding](https://arxiv.org/abs/2203.15442)
:star:[code](https://github.com/LukeForeverYoung/QRNet)
* [Pseudo-Q: Generating Pseudo Language Queries for Visual Grounding](https://openaccess.thecvf.com/content/CVPR2022/papers/Jiang_Pseudo-Q_Generating_Pseudo_Language_Queries_for_Visual_Grounding_CVPR_2022_paper.pdf)
:star:[code](https://github.com/LeapLabTHU/Pseudo-Q)
* [Multi-Modal Dynamic Graph Transformer for Visual Grounding](https://openaccess.thecvf.com/content/CVPR2022/papers/Chen_Multi-Modal_Dynamic_Graph_Transformer_for_Visual_Grounding_CVPR_2022_paper.pdf)
:star:[code](https://github.com/iQua/M-DGT)
## 47.Few/Zero-Shot Learning/Domain Generalization/Adaptation(小/零样本/域泛化/适应)
* 小样本
* [Ranking Distance Calibration for Cross-Domain Few-Shot Learning](https://arxiv.org/abs/2112.00260)
* [Few-shot Learning with Noisy Labels](https://arxiv.org/abs/2204.05494)
* [Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a Difference](https://arxiv.org/abs/2204.07305)
:house:[project](https://hushell.github.io/pmf/):tv:[video](https://youtu.be/iEC9lh18laQ)
* [Few-shot Backdoor Defense Using Shapley Estimation](https://arxiv.org/abs/2112.14889)
:newspaper:[解读](https://mp.weixin.qq.com/s/lSY1is6Fmm6A0Db0Jxo4qg)
* [Attribute Surrogates Learning and Spectral Tokens Pooling in Transformers for Few-Shot Learning](https://arxiv.org/abs/2203.09064)
:star:[code](https://github.com/StomachCold/HCTransformers)
* [EASE: Unsupervised Discriminant Subspace Learning for Transductive Few-Shot Learning](https://openaccess.thecvf.com/content/CVPR2022/papers/Zhu_EASE_Unsupervised_Discriminant_Subspace_Learning_for_Transductive_Few-Shot_Learning_CVPR_2022_paper.pdf)
:star:[code](https://github.com/allenhaozhu/EASE)
* [Semi-Supervised Few-Shot Learning via Multi-Factor Clustering](https://openaccess.thecvf.com/content/CVPR2022/papers/Ling_Semi-Supervised_Few-Shot_Learning_via_Multi-Factor_Clustering_CVPR_2022_paper.pdf)
:star:[code](https://gitlab.com/smartllvlab/cluster-fsl)
* [Cross-Domain Few-Shot Learning With Task-Specific Adapters](https://arxiv.org/abs/2107.00358)
:star:[code](https://github.com/VICO-UoE/URL)
* 零样本
* [MSDN: Mutually Semantic Distillation Network for Zero-Shot Learning](https://arxiv.org/abs/2203.03137)
:star:[code](https://github.com/shiming-chen/MSDN):newspaper:[粗解](https://zhuanlan.zhihu.com/p/477624433)
* [Unseen Classes at a Later Time? No Problem](https://arxiv.org/abs/2203.16517)
:star:[code](https://github.com/sumitramalagi/Unseen-classes-at-a-later-time)
* [En-Compactness: Self-Distillation Embedding & Contrastive Generation for Generalized Zero-Shot Learning](https://openaccess.thecvf.com/content/CVPR2022/papers/Kong_En-Compactness_Self-Distillation_Embedding__Contrastive_Generation_for_Generalized_Zero-Shot_Learning_CVPR_2022_paper.pdf)
:newspaper:[解读](https://mp.weixin.qq.com/s/FYn0S46OA6xraH9ofNIctw)
* [Non-Generative Generalized Zero-Shot Learning via Task-Correlated Disentanglement and Controllable Samples Synthesis](https://arxiv.org/abs/2203.05335)
* [Siamese Contrastive Embedding Network for Compositional Zero-Shot Learning](https://openaccess.thecvf.com/content/CVPR2022/papers/Li_Siamese_Contrastive_Embedding_Network_for_Compositional_Zero-Shot_Learning_CVPR_2022_paper.pdf)
:star:[code](https://github.com/XDUxyLi/SCEN-master)
* [KG-SP: Knowledge Guided Simple Primitives for Open World Compositional Zero-Shot Learning](https://arxiv.org/abs/2205.06784)
:star:[code](https://github.com/ExplainableML/KG-SP)
:newspaper:[解读](https://zhuanlan.zhihu.com/p/515190727)
* [Uni-Perceiver: Pre-Training Unified Architecture for Generic Perception for Zero-Shot and Few-Shot Tasks](https://openaccess.thecvf.com/content/CVPR2022/papers/Zhu_Uni-Perceiver_Pre-Training_Unified_Architecture_for_Generic_Perception_for_Zero-Shot_and_CVPR_2022_paper.pdf)
* [Distinguishing Unseen From Seen for Generalized Zero-Shot Learning](https://openaccess.thecvf.com/content/CVPR2022/papers/Su_Distinguishing_Unseen_From_Seen_for_Generalized_Zero-Shot_Learning_CVPR_2022_paper.pdf)
* [VGSE: Visually-Grounded Semantic Embeddings for Zero-Shot Learning](https://arxiv.org/abs/2203.10444)
:star:[code](https://github.com/wenjiaXu/VGSE)
:newspaper:[零样本学习，大幅减少人工标注！马普所和北邮提出富含视觉信息的类别语义嵌入](https://mp.weixin.qq.com/s/VODFh-oW27w2DCnlgLRLBA)
* [Audio-Visual Generalised Zero-Shot Learning With Cross-Modal Attention and Language](https://arxiv.org/abs/2203.03598)
:star:[code](https://github.com/ExplainableML/AVCA-GZSL)
* 域泛化
* [Compound Domain Generalization via Meta-Knowledge Encoding](https://arxiv.org/pdf/2203.13006.pdf)
* [Causality Inspired Representation Learning for Domain Generalization](https://arxiv.org/abs/2203.14237)
:star:[code](https://github.com/BIT-DA/CIRL)
* [Towards Unsupervised Domain Generalization](https://arxiv.org/abs/2107.06219)
:newspaper:[CVPR 2022丨清华大学提出：无监督域泛化 (UDG)](https://mp.weixin.qq.com/s/ifC6GI5oVpE4ncttmJipjQ)
本次任务的主要目标是域泛化（domain generalization(DG)），是首篇将DG推广到unsupervised learning 领域的，并提出一个新的研究领域 unsupervised domain generalization(UDG)。
* [Towards Principled Disentanglement for Domain Generalization](https://arxiv.org/abs/2111.13839)
:open_mouth:oral:star:[code](https://github.com/hlzhang109/DDG)
* [Meta Convolutional Neural Networks for Single Domain Generalization](https://openaccess.thecvf.com/content/CVPR2022/papers/Wan_Meta_Convolutional_Neural_Networks_for_Single_Domain_Generalization_CVPR_2022_paper.pdf)
* [PCL: Proxy-Based Contrastive Learning for Domain Generalization](https://openaccess.thecvf.com/content/CVPR2022/papers/Yao_PCL_Proxy-Based_Contrastive_Learning_for_Domain_Generalization_CVPR_2022_paper.pdf)
* [Localized Adversarial Domain Generalization](https://arxiv.org/abs/2205.04114)
* [Unsupervised Domain Generalization by Learning a Bridge Across Domains](https://arxiv.org/abs/2112.02300)
* [Style Neophile: Constantly Seeking Novel Styles for Domain Generalization](https://openaccess.thecvf.com/content/CVPR2022/papers/Kang_Style_Neophile_Constantly_Seeking_Novel_Styles_for_Domain_Generalization_CVPR_2022_paper.pdf)
* [BoosterNet: Improving Domain Generalization of Deep Neural Nets Using Culpability-Ranked Features](https://openaccess.thecvf.com/content/CVPR2022/papers/Bayasi_BoosterNet_Improving_Domain_Generalization_of_Deep_Neural_Nets_Using_Culpability-Ranked_CVPR_2022_paper.pdf)
* [Failure Modes of Domain Generalization Algorithms](https://arxiv.org/abs/2111.13733)
* [Geometric and Textural Augmentation for Domain Gap Reduction](https://openaccess.thecvf.com/content/CVPR2022/papers/Liu_Geometric_and_Textural_Augmentation_for_Domain_Gap_Reduction_CVPR_2022_paper.pdf)
:star:[code](https://github.com/xch-liu/geom-tex-dg)
* [Revisiting Domain Generalized Stereo Matching Networks From a Feature Consistency Perspective](https://arxiv.org/abs/2203.10887)
:star:[code](https://github.com/jiaw-z/FCStereo)
* 域外泛化
* [The Two Dimensions of Worst-case Training and the Integrated Effect for Out-of-domain Generalization](https://arxiv.org/abs/2204.04384)
* 域适应
* [Continual Test-Time Domain Adaptation](https://arxiv./abs/2203.13591)
:star:[code](https://github.com/qinenergy/cotta)
* [Safe Self-Refinement for Transformer-based Domain Adaptation](https://arxiv.org/abs/2204.07683)
:star:[code](https://github.com/tsun/SSRT):newspaper:[解读](https://zhuanlan.zhihu.com/p/501027339)
* [Source-Free Domain Adaptation via Distribution Estimation](https://arxiv.org/abs/2204.11257)
:newspaper:[解读](https://mp.weixin.qq.com/s/lSY1is6Fmm6A0Db0Jxo4qg)
* [Learning Distinctive Margin toward Active Domain Adaptation](https://arxiv.org/abs/2203.05738)
:star:[code](https://github.com/TencentYoutuResearch/ActiveLearning-SDM)
:newspaper:[解读](https://mp.weixin.qq.com/s/FYn0S46OA6xraH9ofNIctw)
* [DINE: Domain Adaptation from Single and Multiple Black-box Predictors](https://arxiv.org/abs/2104.01539)
:star:[code](https://github.com/tim-learn/DINE/)
* [Exploring Domain-Invariant Parameters for Source Free Domain Adaptation](https://openaccess.thecvf.com/content/CVPR2022/papers/Wang_Exploring_Domain-Invariant_Parameters_for_Source_Free_Domain_Adaptation_CVPR_2022_paper.pdf)
* [Physically Disentangled Intra- and Inter-Domain Adaptation for Varicolored Haze Removal](https://openaccess.thecvf.com/content/CVPR2022/papers/Li_Physically_Disentangled_Intra-_and_Inter-Domain_Adaptation_for_Varicolored_Haze_Removal_CVPR_2022_paper.pdf)
:star:[code](https://github.com/HuaYuuu/PDI2A-CVPR2022)
* [No-Reference Point Cloud Quality Assessment via Domain Adaptation](https://arxiv.org/abs/2112.02851)
:star:[code](https://github.com/Qi-Yangsjtu/IT-PCQA)
* [Slimmable Domain Adaptation](https://openaccess.thecvf.com/content/CVPR2022/papers/Meng_Slimmable_Domain_Adaptation_CVPR_2022_paper.pdf)
:star:[code](https://github.com/HIK-LAB/SlimDA)
* [SHIFT: A Synthetic Driving Dataset for Continuous Multi-Task Domain Adaptation](https://openaccess.thecvf.com/content/CVPR2022/papers/Sun_SHIFT_A_Synthetic_Driving_Dataset_for_Continuous_Multi-Task_Domain_Adaptation_CVPR_2022_paper.pdf)
* [Geometric Anchor Correspondence Mining With Uncertainty Modeling for Universal Domain Adaptation](https://openaccess.thecvf.com/content/CVPR2022/papers/Chen_Geometric_Anchor_Correspondence_Mining_With_Uncertainty_Modeling_for_Universal_Domain_CVPR_2022_paper.pdf)
* 无监督域适应
* [Reusing the Task-specific Classifier as a Discriminator: Discriminator-free Adversarial Domain Adaptation](https://arxiv.org/abs/2204.03838)
:star:[code](https://github.com/xiaoachen98/DALN)
* [Category Contrast for Unsupervised Domain Adaptation in Visual Tasks](https://arxiv.org/abs/2106.02885)
* [The Norm Must Go On: Dynamic Unsupervised Domain Adaptation by Normalization](https://arxiv.org/abs/2112.00463)
:star:[code](https://github.com/jmiemirza/DUA)
* [Spectral Unsupervised Domain Adaptation for Visual Recognition](http://arxiv.org/abs/2204.13983)
## 46.Scene Graph Generation(场景图生成)
* [PPDL: Predicate Probability Distribution Based Loss for Unbiased Scene Graph Generation](https://openaccess.thecvf.com/content/CVPR2022/papers/Li_PPDL_Predicate_Probability_Distribution_Based_Loss_for_Unbiased_Scene_Graph_CVPR_2022_paper.pdf)
* [Fine-Grained Predicates Learning for Scene Graph Generation](https://arxiv.org/abs/2204.02597)
:star:[code](https://github.com/XinyuLyu/FGPL)
* [HL-Net: Heterophily Learning Network for Scene Graph Generatio](https://arxiv.org/abs/2205.01316)
:star:[code](https://github.com/siml3/HL-Net)
场景图生成：异质学习网络
:newspaper:[解读](https://mp.weixin.qq.com/s/lSY1is6Fmm6A0Db0Jxo4qg)
* [RU-Net: Regularized Unrolling Network for Scene Graph Generation](https://arxiv.org/abs/2205.01297)
:star:[code](https://github.com/siml3/RU-Net)
场景图生成：正则展开网络
:newspaper:[解读](https://mp.weixin.qq.com/s/lSY1is6Fmm6A0Db0Jxo4qg)
* [The Devil is in the Labels: Noisy Label Correction for Robust Scene Graph Generation](https://arxiv.org/abs/2206.03014)
:star:[code](https://github.com/muktilin/NICE)
* [Dynamic Scene Graph Generation via Anticipatory Pre-Training](https://openaccess.thecvf.com/content/CVPR2022/papers/Li_Dynamic_Scene_Graph_Generation_via_Anticipatory_Pre-Training_CVPR_2022_paper.pdf)
* [Stacked Hybrid-Attention and Group Collaborative Learning for Unbiased Scene Graph Generation](https://arxiv.org/abs/2203.09811)
:star:[code](https://github.com/dongxingning/SHA-GCL-for-SGG)
* [Structured Sparse R-CNN for Direct Scene Graph Generation](https://openaccess.thecvf.com/content/CVPR2022/papers/Teng_Structured_Sparse_R-CNN_for_Direct_Scene_Graph_Generation_CVPR_2022_paper.pdf)
:star:[code](https://github.com/MCG-NJU/Structured-Sparse-RCNN)
* [HL-Net: Heterophily Learning Network for Scene Graph Generation](https://openaccess.thecvf.com/content/CVPR2022/papers/Lin_HL-Net_Heterophily_Learning_Network_for_Scene_Graph_Generation_CVPR_2022_paper.pdf)
:star:[code](https://github.com/siml3/HL-Net)
* [Not All Relations Are Equal: Mining Informative Labels for Scene Graph Generation](https://arxiv.org/abs/2111.13517)
* [SGTR: End-to-end Scene Graph Generation with Transformer](https://arxiv.org/abs/2112.12970)
:star:[code](https://github.com/Scarecrow0/SGTR)
* 视频场景图生成
* [Classification-Then-Grounding: Reformulating Video Scene Graphs As Temporal Bipartite Graphs](https://openaccess.thecvf.com/content/CVPR2022/papers/Gao_Classification-Then-Grounding_Reformulating_Video_Scene_Graphs_As_Temporal_Bipartite_Graphs_CVPR_2022_paper.pdf)
:star:[code](https://github.com/Dawn-LX/VidSGG-BIG)
## 45.Dense Prediction(密集预测)
* [Does Robustness on ImageNet Transfer to Downstream Tasks?](https://arxiv.org/abs/2204.03934)
* [MPViT: Multi-Path Vision Transformer for Dense Prediction](https://arxiv.org/abs/2112.11010)
:star:[code](https://github.com/youngwanLEE/MPViT)
* [Learning Multiple Dense Prediction Tasks From Partially Annotated Data](https://arxiv.org/abs/2111.14893)
:star:[code](https://github.com/VICO-UoE/MTPSL)
## 44.Federated Learning(联邦学习)
* [CD2-pFed: Cyclic Distillation-guided Channel Decoupling for Model Personalization in Federated Learning](https://arxiv.org/abs/2204.03880)
* [Auditing Privacy Defenses in Federated Learning via Generative Gradient Leakage](https://arxiv.org/pdf/2203.15696.pdf)
:star:[code](https://github.com/zhuohangli/GGL)
* [FedCorr: Multi-Stage Federated Learning for Label Noise Correction](https://arxiv.org/abs/2204.04677)
:star:[code](https://github.com/Xu-Jingyi/FedCorr)
* [Fine-tuning Global Model via Data-Free Knowledge Distillation for Non-IID Federated Learning](https://arxiv.org/abs/2203.09249)
* [Rethinking Architecture Design for Tackling Data Heterogeneity in Federated Learning](https://arxiv.org/abs/2106.06047)
* [Layer-Wised Model Aggregation for Personalized Federated Learning](http://arxiv.org/abs/2205.03993)
* [Federated Learning With Position-Aware Neurons](https://arxiv.org/abs/2203.14666)
* [Local Learning Matters: Rethinking Data Heterogeneity in Federated Learning](https://arxiv.org/abs/2111.14213)
:star:[code](https://github.com/mmendiet/FedAlign)
* [FedDC: Federated Learning With Non-IID Data via Local Drift Decoupling and Correction](https://arxiv.org/abs/2203.11751)
:star:[code](https://github.com/gaoliang13/FedDC)
* [Learn From Others and Be Yourself in Heterogeneous Federated Learning](https://openaccess.thecvf.com/content/CVPR2022/papers/Huang_Learn_From_Others_and_Be_Yourself_in_Heterogeneous_Federated_Learning_CVPR_2022_paper.pdf)
:star:[code](https://github.com/WenkeHuang/FCCL)
* [FedCor: Correlation-Based Active Client Selection Strategy for Heterogeneous Federated Learning](https://arxiv.org/abs/2103.13822)
* [Robust Federated Learning With Noisy and Heterogeneous Clients](https://openaccess.thecvf.com/content/CVPR2022/papers/Fang_Robust_Federated_Learning_With_Noisy_and_Heterogeneous_Clients_CVPR_2022_paper.pdf)
:star:[code](https://github.com/FangXiuwen/Robust_FL)
* [ResSFL: A Resistance Transfer Framework for Defending Model Inversion Attack in Split Federated Learning](https://arxiv.org/abs/2205.04007)
:star:[code](https://github.com/zlijingtao/ResSFL)
## 43.Multi-Task Learning(多任务学习)
* [Controllable Dynamic Multi-Task Architectures](https://arxiv.org/abs/2203.14949)
:house:[project](https://www.nec-labs.com/~mas/DYMU/)
* [Task Adaptive Parameter Sharing for Multi-Task Learning](https://arxiv.org/abs/2203.16708)
* [Raw High-Definition Radar for Multi-Task Learning](https://arxiv.org/abs/2112.10646)
:star:[code](https://github.com/valeoai/RADIal)
## 42.Metric Learning(度量学习)
* [Self-Taught Metric Learning without Labels](https://arxiv.org/abs/2205.01903)
:star:[code](https://github.com/tjddus9597/STML-CVPR22):house:[project](http://cvlab.postech.ac.kr/research/STML/)
* [Enhancing Adversarial Robustness for Deep Metric Learning](https://arxiv.org/abs/2203.01439)
* [Hypergraph-Induced Semantic Tuplet Loss for Deep Metric Learning](https://openaccess.thecvf.com/content/CVPR2022/papers/Lim_Hypergraph-Induced_Semantic_Tuplet_Loss_for_Deep_Metric_Learning_CVPR_2022_paper.pdf)
:star:[code](https://github.com/ljin0429/HIST)
* [Non-Isotropy Regularization for Proxy-Based Deep Metric Learning](https://arxiv.org/abs/2203.08547)
:star:[code](https://github.com/ExplainableML/NonIsotropicProxyDML)
* [Hyperbolic Vision Transformers: Combining Improvements in Metric Learning](https://arxiv.org/abs/2203.10833)
:star:[code](https://github.com/htdt/hyp_metric)
* [Enhancing Adversarial Robustness for Deep Metric Learning](https://arxiv.org/abs/2203.01439)
* [Weakly-Supervised Metric Learning With Cross-Module Communications for the Classification of Anterior Chamber Angle Images](https://openaccess.thecvf.com/content/CVPR2022/papers/Huang_Weakly-Supervised_Metric_Learning_With_Cross-Module_Communications_for_the_Classification_of_CVPR_2022_paper.pdf)
:star:[code](https://github.com/Jingqi-H/GCNet)
* [Integrating Language Guidance Into Vision-Based Deep Metric Learning](https://arxiv.org/abs/2203.08543)
:star:[code](https://github.com/ExplainableML/LanguageGuidance_for_DML)
## 41.Incremental Learning(增量学习)
* 增量学习
* [Energy-based Latent Aligner for Incremental Learning](https://arxiv.org/abs/2203.14952)
:star:[code](https://github.com/JosephKJ/ELI)
* [General Incremental Learning with Domain-aware Categorical Representations](https://arxiv.org/abs/2204.04078)
* [Forward Compatible Few-Shot Class-Incremental Learning](https://arxiv.org/abs/2203.06953)
:star:[code](https://github.com/zhoudw-zdw/CVPR22-Fact)
* [Mimicking the Oracle: An Initial Phase Decorrelation Approach for Class Incremental Learning](https://arxiv.org/abs/2112.04731)
:star:[code](https://github.com/Yujun-Shi/CwD)
* [Few-Shot Incremental Learning for Label-to-Image Translation](https://openaccess.thecvf.com/content/CVPR2022/papers/Chen_Few-Shot_Incremental_Learning_for_Label-to-Image_Translation_CVPR_2022_paper.pdf)
* 类增量学习
* [Doodle It Yourself: Class Incremental Learning by Drawing a Few Sketches](https://arxiv.org/abs/2203.14843)
* [Constrained Few-shot Class-incremental Learning](https://arxiv.org/abs/2203.16588)
:star:[code](https://github.com/IBM/constrained-FSCIL)
* [Class-Incremental Learning with Strong Pre-trained Models](https://arxiv.org/abs/2204.03634)
* [Class-Incremental Learning by Knowledge Distillation With Adaptive Feature Consolidation](https://arxiv.org/abs/2204.00895)
:star:[code](https://github.com/kminsoo/AFC)
* [Bring Evanescent Representations to Life in Lifelong Class Incremental Learning](https://openaccess.thecvf.com/content/CVPR2022/papers/Toldo_Bring_Evanescent_Representations_to_Life_in_Lifelong_Class_Incremental_Learning_CVPR_2022_paper.pdf)
* [Self-Sustaining Representation Expansion for Non-Exemplar Class-Incremental Learning](https://arxiv.org/abs/2203.06359)
* [MetaFSCIL: A Meta-Learning Approach for Few-Shot Class Incremental Learning](https://openaccess.thecvf.com/content/CVPR2022/papers/Chi_MetaFSCIL_A_Meta-Learning_Approach_for_Few-Shot_Class_Incremental_Learning_CVPR_2022_paper.pdf)
* [Federated Class-Incremental Learning](https://arxiv.org/abs/2203.11473)
:star:[code](https://github.com/conditionWang/FCIL)
* [vCLIMB: A Novel Video Class Incremental Learning Benchmark](https://arxiv.org/abs/2201.09381)
:open_mouth:oral:star:[code](https://github.com/ojedaf/vCLIMB_Benchmark):house:[project](https://vclimb.netlify.app/)
## 40.Adversarial Learning(对抗学习)
* [Give Me Your Attention: Dot-Product Attention Considered Harmful for Adversarial Patch Robustness](https://arxiv.org/abs/2203.13639)
* [Masking Adversarial Damage: Finding Adversarial Saliency for Robust and Sparse Network](https://arxiv.org/abs/2204.02738)
* [Towards Practical Certifiable Patch Defense with Vision Transformer](https://arxiv.org/abs/2203.08519)
:newspaper:[解读](https://mp.weixin.qq.com/s/FYn0S46OA6xraH9ofNIctw)
* [Enhancing Adversarial Training with Second-Order Statistics of Weights](https://arxiv.org/abs/2203.06020)
:star:[code](https://github.com/Alexkael/S2O)
* [Practical Evaluation of Adversarial Robustness via Adaptive Auto Attack](https://arxiv.org/abs/2203.05154)
:star:[code](https://github.com/liuye6666/adaptive_auto_attack)
* [Improving Adversarial Transferability via Neuron Attribution-Based Attacks](https://arxiv.org/abs/2204.00008)
:star:[code](https://github.com/jpzhang1810/NAA)
* [Two Coupled Rejection Metrics Can Tell Adversarial Examples Apart](https://arxiv.org/abs/2105.14785)
:star:[code](https://github.com/P2333/Rectified-Rejection)
* [Bounded Adversarial Attack on Deep Content Features](https://openaccess.thecvf.com/content/CVPR2022/papers/Xu_Bounded_Adversarial_Attack_on_Deep_Content_Features_CVPR_2022_paper.pdf)
* [Subspace Adversarial Training](https://arxiv.org/abs/2111.12229)
:star:[code](https://github.com/nblt/Sub-AT)
* [Cross-Modal Transferable Adversarial Attacks From Images to Videos](https://arxiv.org/abs/2112.05379)
:star:[code](https://github.com/zhipeng-wei/Image-to-Video-I2V-attack)
* [Understanding and Increasing Efficiency of Frank-Wolfe Adversarial Training](https://arxiv.org/abs/2012.12368)
:star:[code](https://github.com/TheoT1/FW-AT-Adapt)
* [Quarantine: Sparsity Can Uncover the Trojan Attack Trigger for Free](https://arxiv.org/abs/2205.11819)
:star:[code](https://github.com/VITA-Group/Backdoor-LTH)
* [Robust Combination of Distributed Gradients Under Adversarial Perturbations](https://openaccess.thecvf.com/content/CVPR2022/papers/Kim_Robust_Combination_of_Distributed_Gradients_Under_Adversarial_Perturbations_CVPR_2022_paper.pdf)
* [Adversarial Texture for Fooling Person Detectors in the Physical World](https://arxiv.org/abs/2203.03373)
* [DTA: Physical Camouflage Attacks Using Differentiable Transformation Network](https://arxiv.org/abs/2203.09831)
:house:[project](https://islab-ai.github.io/dta-cvpr2022/)
* [BppAttack: Stealthy and Efficient Trojan Attacks Against Deep Neural Networks via Image Quantization and Contrastive Adversarial Learning](https://arxiv.org/abs/2205.13383)
:star:[code](https://github.com/RU-System-Software-and-Security/BppAttack)
* [Pyramid Adversarial Training Improves ViT Performance](https://arxiv.org/abs/2111.15121)
:house:[project](https://pyramidat.github.io/)
* [NinjaDesc: Content-Concealing Visual Descriptors via Adversarial Learning](https://arxiv.org/abs/2112.12785)
* 对抗样本
* [Label-Only Model Inversion Attacks via Boundary Repulsion](https://arxiv.org/abs/2203.01925)
:star:[code](https://github.com/m-kahla/Label-Only-Model-Inversion-Attacks-via-Boundary-Repulsion)
* [Self-supervised Learning of Adversarial Example: Towards Good Generalizations for Deepfake Detection](https://arxiv.org/abs/2203.1220)
:star:[code](https://github.com/liangchen527/SLADD8)
* [Improving the Transferability of Targeted Adversarial Examples Through Object-Based Diverse Input](https://arxiv.org/abs/2203.09123)
:star:[code](https://github.com/dreamflake/ODI)
* [Leveraging Adversarial Examples To Quantify Membership Information Leakage](https://arxiv.org/abs/2203.09566)
:star:[code](https://github.com/fra31/auto-atta)
* 对抗攻击
* [Shadows can be Dangerous: Stealthy and Effective Physical-world Adversarial Attack by Natural Phenomenon](https://arxiv.org/abs/2203.03818)
:star:[code](https://github.com/hncszyq/ShadowAttack)
* [Transferable Sparse Adversarial Attack](https://arxiv.org/abs/2105.14727)
:star:[code](https://github.com/shaguopohuaizhe/TSAA)
* [Towards Efficient Data Free Black-Box Adversarial Attack](https://openaccess.thecvf.com/content/CVPR2022/papers/Zhang_Towards_Efficient_Data_Free_Black-Box_Adversarial_Attack_CVPR_2022_paper.pdf)
* [Frequency-Driven Imperceptible Adversarial Attack on Semantic Similarity](https://arxiv.org/abs/2203.05151)
:star:[code](https://github.com/LinQinLiang/SSAH-adversarial-attack)
* [Stochastic Variance Reduced Ensemble Adversarial Attack for Boosting the Adversarial Transferability](https://arxiv.org/abs/2111.10752)
:star:[code](https://github.com/JHL-HUST/SVRE)
* 黑盒
* [Investigating Top-k White-Box and Transferable Black-box Attack](https://arxiv.org/abs/2204.00089)
:star:[code](https://github.com/ChaoningZhang/Top-k-Transferable-Attack)
* [DST: Dynamic Substitute Training for Data-free Black-box Attack](https://arxiv.org/abs/2204.00972)
:house:[project](https://wxwangiris.github.io/DST)
* [Bandits for Structure Perturbation-based Black-box Attacks to Graph Neural Networks with Theoretical Guarantees](https://arxiv.org/abs/2205.03546)
:open_mouth:oral:star:[code](https://github.com/Metaoblivion/Bandit_GNN_Attack)
* [Adversarial Eigen Attack on Black-Box Models](https://openaccess.thecvf.com/content/CVPR2022/papers/Zhou_Adversarial_Eigen_Attack_on_Black-Box_Models_CVPR_2022_paper.pdf)
* [Exploring Effective Data for Surrogate Training Towards Black-Box Attack](https://openaccess.thecvf.com/content/CVPR2022/papers/Sun_Exploring_Effective_Data_for_Surrogate_Training_Towards_Black-Box_Attack_CVPR_2022_paper.pdf)
:star:[code](https://github.com/xuxiangsun/ST-Data)
* [Boosting Black-Box Attack With Partially Transferred Conditional Adversarial Distribution](https://arxiv.org/abs/2006.08538)
:star:[code](https://github.com/Kira0096/CGATTACK)
* 对抗训练
* [LAS-AT: Adversarial Training with Learnable Attack Strategy](https://arxiv.org/pdf/2203.06616.pdf)
:open_mouth:oral:star:[code](https://github.com/jiaxiaojunQAQ/LAS-AT)
:newspaper:[CVPR 2022 中科院、腾讯提出LAS-AT，利用“可学习攻击策略”进行“对抗训练”](https://mp.weixin.qq.com/s/Aj9x61LY8tJICf8hUlz8ug)
## 39.Continual Learning(持续学习)
* [On Generalizing Beyond Domains in Cross-Domain Continual Learning](https://arxiv.org/abs/2203.03970)
* [Probing Representation Forgetting in Supervised and Unsupervised Continual Learning](https://arxiv.org/abs/2203.13381)
:star:[code](https://github.com/rezazzr/Probing-Representation-Forgetting)
* [Online Continual Learning on a Contaminated Data Stream with Blurry Task Boundaries](https://arxiv.org/abs/2203.15355)
:star:[code](https://github.com/clovaai/puridiver)
* [Learning To Prompt for Continual Learning](https://arxiv.org/abs/2112.08654)
:star:[code](https://github.com/google-research/l2p)
* [Learning Bayesian Sparse Networks With Full Experience Replay for Continual Learning](https://arxiv.org/abs/2202.10203)
* [Not Just Selection, but Exploration: Online Class-Incremental Continual Learning via Dual View Consistency](https://openaccess.thecvf.com/content/CVPR2022/papers/Gu_Not_Just_Selection_but_Exploration_Online_Class-Incremental_Continual_Learning_via_CVPR_2022_paper.pdf)
:star:[code](https://github.com/YananGu/DVC)
* [Continual Learning for Visual Search With Backward Consistent Feature Embedding](https://arxiv.org/abs/2205.13384)
:star:[code](https://github.com/ivclab/CVS)
* [Meta-Attention for ViT-Backed Continual Learning](https://arxiv.org/abs/2203.11684)
:star:[code](https://github.com/zju-vipa/MEAT-TIL)
* [Continual Learning with Lifelong Vision Transformer](https://openaccess.thecvf.com/content/CVPR2022/papers/Wang_Continual_Learning_With_Lifelong_Vision_Transformer_CVPR_2022_paper.pdf)
:newspaper:[解读](https://mp.weixin.qq.com/s/lSY1is6Fmm6A0Db0Jxo4qg)
* [DyTox: Transformers for Continual Learning With DYnamic TOken eXpansion](https://openaccess.thecvf.com/content/CVPR2022/papers/Douillard_DyTox_Transformers_for_Continual_Learning_With_DYnamic_TOken_eXpansion_CVPR_2022_paper.pdf)
:star:[code](https://github.com/arthurdouillard/dytox)
* [GCR: Gradient Coreset Based Replay Buffer Selection for Continual Learning](https://arxiv.org/abs/2111.11210)
:house:[project](https://gradientcoreset.github.io/)
## 38.Meta-Learning(元学习)
* [What Matters For Meta-Learning Vision Regression Tasks?](https://arxiv.org/abs/2203.04905)
:star:[code](https://github.com/boschresearch/what-matters-for-meta-learning)
* [Multidimensional Belief Quantification for Label-Efficient Meta-Learning](https://arxiv.org/abs/2203.12768)
* [Dynamic Kernel Selection for Improved Generalization and Memory Efficiency in Meta-learning](https://arxiv.org/abs/2206.01690)
:star:[code](https://github.com/transmuteAI/MetaDOCK)
* [Learning to Learn and Remember Super Long Multi-Domain Task Sequence](https://openaccess.thecvf.com/content/CVPR2022/papers/Wang_Learning_To_Learn_and_Remember_Super_Long_Multi-Domain_Task_Sequence_CVPR_2022_paper.pdf)
:open_mouth:oral:star:[code](https://github.com/joey-wang123/SDML)
:newspaper:[解读](https://mp.weixin.qq.com/s/lSY1is6Fmm6A0Db0Jxo4qg)
## 37.Contrastive Learning(对比学习)
* [Selective-Supervised Contrastive Learning with Noisy Labels](https://arxiv.org/abs/2203.04181)
:star:[code](https://github.com/ShikunLi/Sel-CL):newspaper:[粗解](https://zhuanlan.zhihu.com/p/478070143)
* [Frame-wise Action Representations for Long Videos via Sequence Contrastive Learning](https://arxiv.org/abs/2203.14957)
:star:[code](https://github.com/minghchen/CARL_code)
* [Cam-Ready: UNICON: Combating Label Noise Through Uniform Selection and Contrastive Learning](https://arxiv.org/abs/2203.14542)
:star:[code](https://github.com/nazmul-karim170/UNICON-Noisy-Labe)
* [Use All The Labels: A Hierarchical Multi-Label Contrastive Learning Framework](https://arxiv.org/abs/2204.13207)
:star:[code](https://github.com/salesforce/hierarchicalContrastiveLearning)
* [Crafting Better Contrastive Views for Siamese Representation Learning](https://arxiv.org/abs/2202.03278)
:open_mouth:oral:star:[code](https://github.com/xyupeng/ContrastiveCrop)
* [Dual Temperature Helps Contrastive Learning Without Many Negative Samples: Towards Understanding and Simplifying MoCo](https://arxiv.org/abs/2203.17248)
:star:[code](https://github.com/ChaoningZhang/Dual-temperature)
* [Estimating Fine-Grained Noise Model via Contrastive Learning](https://arxiv.org/abs/2204.01716)
* [Contextual Outpainting With Object-Level Contrastive Learning](https://openaccess.thecvf.com/content/CVPR2022/papers/Li_Contextual_Outpainting_With_Object-Level_Contrastive_Learning_CVPR_2022_paper.pdf)
:house:[project](https://ddlee-cn.github.io/cto-gan/)
* [Rethinking the Augmentation Module in Contrastive Learning: Learning Hierarchical Augmentation Invariance With Expanded Views](https://arxiv.org/abs/2206.00227)
* [Contrastive Dual Gating: Learning Sparse Features With Contrastive Learning](https://openaccess.thecvf.com/content/CVPR2022/papers/Meng_Contrastive_Dual_Gating_Learning_Sparse_Features_With_Contrastive_Learning_CVPR_2022_paper.pdf)
* [Noise Is Also Useful: Negative Correlation-Steered Latent Contrastive Learning](https://openaccess.thecvf.com/content/CVPR2022/papers/Yan_Noise_Is_Also_Useful_Negative_Correlation-Steered_Latent_Contrastive_Learning_CVPR_2022_paper.pdf)
* [On Learning Contrastive Representations for Learning With Noisy Labels](https://arxiv.org/abs/2203.01785)
* [Unsupervised Deraining: Where Contrastive Learning Meets Self-Similarity](https://arxiv.org/abs/2203.11509)
* [Robust Contrastive Learning Against Noisy Views](https://arxiv.org/abs/2201.04309)
:star:[code](https://github.com/chingyaoc/RINCE)
* [Unified Contrastive Learning in Image-Text-Label Space](https://arxiv.org/abs/2204.03610)
:star:[code](https://github.com/microsoft/UniCL)
* [Consistent Explanations by Contrastive Learning](https://arxiv.org/abs/2110.00527)
:star:[code](https://github.com/UCDvision/CGC)
* [Rethinking Minimal Sufficient Representation in Contrastive Learning](https://arxiv.org/abs/2203.07004)
:star:[code](https://github.com/Haoqing-Wang/InfoCL)
* [Contrastive Learning for Space-Time Correspondence via Self-Cycle Consistency](https://openaccess.thecvf.com/content/CVPR2022/papers/Son_Contrastive_Learning_for_Space-Time_Correspondence_via_Self-Cycle_Consistency_CVPR_2022_paper.pdf)
* [M5Product: Self-harmonized Contrastive Learning for E-commercial Multi-modal Pretraining](https://openaccess.thecvf.com/content/CVPR2022/papers/Dong_M5Product_Self-Harmonized_Contrastive_Learning_for_E-Commercial_Multi-Modal_Pretraining_CVPR_2022_paper.pdf)
:sunflower:[dataset](https://xiaodongsuper.github.io/M5Product_dataset/)
* [Multi-Marginal Contrastive Learning for Multi-Label Subcellular Protein Localization](https://openaccess.thecvf.com/content/CVPR2022/papers/Liu_Multi-Marginal_Contrastive_Learning_for_Multi-Label_Subcellular_Protein_Localization_CVPR_2022_paper.pdf)
:star:[code](https://github.com/ziniBRC/DeePSLoc)
* [Unpaired Deep Image Deraining Using Dual Contrastive Learning](https://arxiv.org/abs/2109.02973)
:star:[code](https://drive.google.com/file/d/1ZIEGR75jbBSKiCagkMBAijkUB65vsTfB/view):house:[project](https://cxtalk.github.io/projects/DCD-GAN.html)
## 36.Optical Flow(光流估计)
* [CRAFT: Cross-Attentional Flow Transformer for Robust Optical Flow](https://arxiv.org/abs/2203.16896)
:star:[code](https://github.com/askerlee/craft)
* [DIP: Deep Inverse Patchmatch for High-Resolution Optical Flow](https://arxiv.org/abs/2204.00330)
:star:[code](https://github.com/zihuazheng/DIP)
* [Imposing Consistency for Optical Flow Estimation](https://arxiv.org/abs/2204.07262)
* [Deep Equilibrium Optical Flow Estimation](https://arxiv.org/abs/2204.08442)
:star:[code](https://github.com/locuslab/deq-flow):newspaper:[解读](https://zhuanlan.zhihu.com/p/501027339)
* [GMFlow: Learning Optical Flow via Global Matching](https://arxiv.org/abs/2111.13680)
:open_mouth:oral:star:[code](https://github.com/haofeixu/gmflow):newspaper:[解读](https://mp.weixin.qq.com/s/lSY1is6Fmm6A0Db0Jxo4qg)
* [Optical Flow Estimation for Spiking Camera](https://arxiv.org/abs/2110.03916)
:star:[code](https://github.com/Acnext/Optical-Flow-For-Spiking-Camera)
* [Learning Optical Flow with Kernel Patch Attention](https://openaccess.thecvf.com/content/CVPR2022/papers/Luo_Learning_Optical_Flow_With_Kernel_Patch_Attention_CVPR_2022_paper.pdf)
:star:[code](https://github.com/megvii-research/KPAFlow):newspaper:[解读](https://mp.weixin.qq.com/s/kZue3ds348UXQI86xrwudQ)
* [CamLiFlow: Bidirectional Camera-LiDAR Fusion for Joint Optical Flow and Scene Flow Estimation](https://arxiv.org/abs/2111.10502)
:star:[code](https://github.com/MCG-NJU/CamLiFlow)
* [Global Matching With Overlapping Attention for Optical Flow Estimation](https://arxiv.org/abs/2203.11335)
:star:[code](https://github.com/xiaofeng94/GMFlowNet)
* [Towards Understanding Adversarial Robustness of Optical Flow Networks](https://arxiv.org/abs/2103.16255)
:star:[code](https://github.com/lmb-freiburg/understanding_flow_robustness)
## 35.OCR
* [XYLayoutLM: Towards Layout-Aware Multimodal Networks for Visually-Rich Document Understanding](https://arxiv.org/abs/2203.06947)
* [SwinTextSpotter: Scene Text Spotting via Better Synergy Between Text Detection and Text Recognition](https://arxiv.org/abs/2203.10209)
:star:[code](https://github.com/mxin262/SwinTextSpotter)
* 场景文本检测
* [Towards End-to-End Unified Scene Text Detection and Layout Analysis](https://arxiv.org/abs/2203.15143)
:star:[code](https://github.com/google-research-datasets/hiertext)
* [Pushing the Performance Limit of Scene Text Recognizer without Human Annotation](https://arxiv.org/abs/2204.07714)
* [Vision-Language Pre-Training for Boosting Scene Text Detectors](https://arxiv.org/abs/2204.13867)
:star:[code](https://github.com/CVI-SZU/STKM)
视觉语言预训练，场景文本检测,代码将开源，地址尚未公布。
* [Few Could Be Better Than All: Feature Sampling and Grouping for Scene Text Detection](https://arxiv.org/abs/2203.15221)
* 场景文本识别
* [SimAN: Exploring Self-Supervised Representation Learning of Scene Text via Similarity-Aware Normalization](https://arxiv.org/abs/2203.10492)
:star:[code](https://github.com/Canjie-Luo/Real-300K)
* Text Spotting
* [Text Spotting Transformers](https://arxiv.org/abs/2204.01918)
:star:[code](https://github.com/mlpc-ucsd/TESTR):newspaper:[粗解](https://zhuanlan.zhihu.com/p/493615566)
* [Towards Weakly-Supervised Text Spotting Using a Multi-Task Transformer](https://arxiv.org/abs/2202.05508)
* LOGO设计
* [Aesthetic Text Logo Synthesis via Content-aware Layout Inferring](https://arxiv.org/abs/2204.02701)
:star:[code](https://github.com/yizhiwang96/TextLogoLayout)
:newspaper:[CVPR 2022 | 北大、腾讯提出文字logo生成模型，脑洞大开堪比设计师](https://mp.weixin.qq.com/s/gjrdktxwbTDPWWeIK5wVNQ)
* 字体生成
* [XMP-Font: Self-Supervised Cross-Modality Pre-training for Few-Shot Font Generation](https://arxiv.org/abs/2204.05084)
* [(Oral)Look Closer to Supervise Better: One-Shot Font Generation via Component-Based Discriminator](https://arxiv.org/abs/2205.00146)
字体生成（很有商业价值的方向）
* [Few-Shot Font Generation by Learning Fine-Grained Local Styles](https://arxiv.org/abs/2205.09965)
* 文本识别
* [Open-set Text Recognition via Character-Context Decoupling](https://arxiv.org/abs/2204.05535)
* 表格结构识别
* [Neural Collaborative Graph Machines for Table Structure Recognition](https://arxiv.org/abs/2111.13359)
:newspaper:[解读](https://mp.weixin.qq.com/s/FYn0S46OA6xraH9ofNIctw)
* 文本美观预测评估
* [Does Text Attract Attention on E-Commerce Images: A Novel Saliency Prediction Dataset and Method](https://openaccess.thecvf.com/content/CVPR2022/papers/Jiang_Does_Text_Attract_Attention_on_E-Commerce_Images_A_Novel_Saliency_CVPR_2022_paper.pdf)
:star:[code](https://github.com/leafy-lee/E-commercial-dataset)
* 表结构理解
* [TableFormer: Table Structure Understanding with Transformers](https://arxiv.org/abs/2203.01017)
* 文本分割
* [BTS: A Bi-Lingual Benchmark for Text Segmentation in the Wild](https://openaccess.thecvf.com/content/CVPR2022/papers/Xu_BTS_A_Bi-Lingual_Benchmark_for_Text_Segmentation_in_the_Wild_CVPR_2022_paper.pdf)
* 表格检测
* [PubTables-1M: Towards Comprehensive Table Extraction From Unstructured Documents](https://openaccess.thecvf.com/content/CVPR2022/papers/Smock_PubTables-1M_Towards_Comprehensive_Table_Extraction_From_Unstructured_Documents_CVPR_2022_paper.pdf)
:star:[code](https://github.com/microsoft/table-transformer)
* 文本修复
* [Fourier Document Restoration for Robust Document Dewarping and Recognition](https://arxiv.org/abs/2203.09910)
:house:[project](https://sg-vilab.github.io/event/warpdoc/)
* 手写数学表达式识别
* [Syntax-Aware Network for Handwritten Mathematical Expression Recognition](https://arxiv.org/abs/2203.01601)
## 34.Model Compression/Knowledge Distillation/Pruning(模型压缩/知识蒸馏/剪枝)
* 知识蒸馏
* [Knowledge Distillation with the Reused Teacher Classifier](https://arxiv.org/abs/2203.14001)
* [DearKD: Data-Efficient Early Knowledge Distillation for Vision Transformers](https://arxiv.org/abs/2204.12997)
:newspaper:[解读](https://mp.weixin.qq.com/s/lSY1is6Fmm6A0Db0Jxo4qg)
* [Decoupled Knowledge Distillation](https://arxiv.org/abs/2203.08679)
:star:[code](https://github.com/megvii-research/mdistiller)
:newspaper:[解耦知识蒸馏，让Hinton在7年前提出的方法重回SOTA行列](https://mp.weixin.qq.com/s/ozLLnUf8KggVzbPxeegQ3g)
* [Knowledge Distillation via the Target-aware Transformer](https://arxiv.org/abs/2205.10793)
:open_mouth:oral:star:[code](https://github.com/sihaoevery/TaT)
:newspaper:[RMIT&阿里&UTS&中山提出Target-aware Transformer，进行one-to-all知识蒸馏！性能SOTA](https://mp.weixin.qq.com/s/hz8julfb0ahYeT8kxGvS9w)
* [Evaluation-oriented Knowledge Distillation for Deep Face Recognition](https://arxiv.org/abs/2206.02325)
:open_mouth:oral:star:[code](https://github.com/Tencent/TFace/tree/master/recognition/tasks/ekd)
:newspaper:[解读1](https://zhuanlan.zhihu.com/p/525331776)
:newspaper:[解读2](https://mp.weixin.qq.com/s/FYn0S46OA6xraH9ofNIctw)
* [Open-Vocabulary One-Stage Detection With Hierarchical Visual-Language Knowledge Distillation](https://arxiv.org/abs/2203.10593)
:star:[code](https://github.com/mengqiDyangge/HierKD)
* [Self-Distillation From the Last Mini-Batch for Consistency Regularization](https://arxiv.org/abs/2203.16172)
:star:[code](https://github.com/Meta-knowledge-Lab/DLB)
* [Knowledge Distillation As Efficient Pre-Training: Faster Convergence, Higher Data-Efficiency, and Better Transferability](https://openaccess.thecvf.com/content/CVPR2022/papers/He_Knowledge_Distillation_As_Efficient_Pre-Training_Faster_Convergence_Higher_Data-Efficiency_and_CVPR_2022_paper.pdf)
:star:[code](https://github.com/CVMI-Lab/KDEP)
* [Knowledge Distillation: A Good Teacher Is Patient and Consistent](https://arxiv.org/abs/2106.05237)
* [PCA-Based Knowledge Distillation Towards Lightweight and Content-Style Balanced Photorealistic Style Transfer Models](https://arxiv.org/abs/2203.13452)
:star:[code](https://github.com/chiutaiyin/PCA-Knowledge-Distillation)
* [Structural and Statistical Texture Knowledge Distillation for Semantic Segmentation](https://openaccess.thecvf.com/content/CVPR2022/papers/Ji_Structural_and_Statistical_Texture_Knowledge_Distillation_for_Semantic_Segmentation_CVPR_2022_paper.pdf)
* 模型压缩
* [CHEX: CHannel EXploration for CNN Model Compression](https://arxiv.org/abs/2203.15794)
* [DiSparse: Disentangled Sparsification for Multitask Model Compression](https://openaccess.thecvf.com/content/CVPR2022/papers/Sun_DiSparse_Disentangled_Sparsification_for_Multitask_Model_Compression_CVPR_2022_paper.pdf)
:star:[code](https://github.com/SHI-Labs/DiSparse-Multitask-Model-Compression)
* 剪枝
* [Revisiting Random Channel Pruning for Neural Network Compression](https://arxiv.org/abs/2205.05676)
:star:[code](https://github.com/ofsoundof/random_channel_pruning)
:newspaper:[解读](https://zhuanlan.zhihu.com/p/513130382)
* [Fire Together Wire Together: A Dynamic Pruning Approach With Self-Supervised Mask Prediction](https://arxiv.org/abs/2110.08232)
* [When To Prune? A Policy Towards Early Structural Pruning](https://arxiv.org/abs/2110.12007)
* [Interspace Pruning: Using Adaptive Filter Representations To Improve Training of Sparse CNNs](https://arxiv.org/abs/2203.07808)
* 量化
* [A Deeper Dive Into What Deep Spatiotemporal Networks Encode: Quantifying Static vs. Dynamic Information](https://arxiv.org/abs/2206.02846)
:star:[code](https://github.com/YorkUCVIL/Static-Dynamic-Interpretability/):house:[project](https://yorkucvil.github.io/Static-Dynamic-Interpretability/)
* [Mr.BiQ: Post-Training Non-Uniform Quantization Based on Minimizing the Reconstruction Error](https://openaccess.thecvf.com/content/CVPR2022/papers/Jeon_Mr.BiQ_Post-Training_Non-Uniform_Quantization_Based_on_Minimizing_the_Reconstruction_Error_CVPR_2022_paper.pdf)
* [Nonuniform-to-Uniform Quantization: Towards Accurate Quantization via Generalized Straight-Through Estimation](https://arxiv.org/abs/2111.14826)
:star:[code](https://github.com/liuzechun/Nonuniform-to-Uniform-Quantization)
* [AlignQ: Alignment Quantization With ADMM-Based Correlation Preservation](https://openaccess.thecvf.com/content/CVPR2022/papers/Chen_AlignQ_Alignment_Quantization_With_ADMM-Based_Correlation_Preservation_CVPR_2022_paper.pdf)
:star:[code](https://github.com/tinganchen/AlignQ)
* [Data-Free Network Compression via Parametric Non-Uniform Mixed Precision Quantization](https://openaccess.thecvf.com/content/CVPR2022/papers/Chikin_Data-Free_Network_Compression_via_Parametric_Non-Uniform_Mixed_Precision_Quantization_CVPR_2022_paper.pdf)
* [Mutual Quantization for Cross-Modal Search With Noisy Labels](https://openaccess.thecvf.com/content/CVPR2022/papers/Yang_Mutual_Quantization_for_Cross-Modal_Search_With_Noisy_Labels_CVPR_2022_paper.pdf)
* [Instance-Aware Dynamic Neural Network Quantization](https://openaccess.thecvf.com/content/CVPR2022/papers/Liu_Instance-Aware_Dynamic_Neural_Network_Quantization_CVPR_2022_paper.pdf)
:star:[code](https://github.com/huaweinoah/Efficient-Computing)
* [IntraQ: Learning Synthetic Images With Intra-Class Heterogeneity for Zero-Shot Network Quantization](https://arxiv.org/abs/2111.09136)
:star:[code](https://github.com/zysxmu/IntraQ)
* [Learnable Lookup Table for Neural Network Quantization](https://openaccess.thecvf.com/content/CVPR2022/papers/Wang_Learnable_Lookup_Table_for_Neural_Network_Quantization_CVPR_2022_paper.pdf)
* [Channel Balancing for Accurate Quantization of Winograd Convolutions](https://openaccess.thecvf.com/content/CVPR2022/papers/Chikin_Channel_Balancing_for_Accurate_Quantization_of_Winograd_Convolutions_CVPR_2022_paper.pdf)
* 超参数优化
* [AME: Attention and Memory Enhancement in Hyper-Parameter Optimization](https://openaccess.thecvf.com/content/CVPR2022/papers/Xu_AME_Attention_and_Memory_Enhancement_in_Hyper-Parameter_Optimization_CVPR_2022_paper.pdf)
## 33.Human-Object Interaction(人物交互)
* [HOI4D: A 4D Egocentric Dataset for Category-Level Human-Object Interaction](https://arxiv.org/abs/2203.01577)
:star:[code](https://hoi4d.github.io/)
* [MSTR: Multi-Scale Transformer for End-to-End Human-Object Interaction Detection](https://arxiv.org/abs/2203.14709)
* [GEN-VLKT: Simplify Association and Enhance Interaction Understanding for HOI Detection](https://arxiv.org/abs/2203.13954)
:star:[code](https://github.com/YueLiao/gen-vlkt)
* [Distillation Using Oracle Queries for Transformer-Based Human-Object Interaction Detection](https://openaccess.thecvf.com/content/CVPR2022/papers/Qu_Distillation_Using_Oracle_Queries_for_Transformer-Based_Human-Object_Interaction_Detection_CVPR_2022_paper.pdf)
:star:[code](https://github.com/SherlockHolmes221/DOQ)
* [OakInk: A Large-scale Knowledge Repository for Understanding Hand-Object Interaction](https://arxiv.org/abs/2203.15709)
:star:[code](https://github.com/lixiny/OakInk)
:newspaper:[粗解](https://news.sjt