官网链接:https://eccv2022.ecva.net/
截稿日期:2022年3月7日(9:59PM CET, 11:59AM PST)
会议日期:2022年10月24日-2022年10月28日
历年综述论文分类汇总戳这里↘️ CV-Surveys施工中~~~~~~~~~~
- 3D
- 去雪
- 医学图像分割
- HOS
- depth restoration
- 类增量学习
- GAN
- 跟踪
- 目标检测
- 光流
- 图像修复
- VL
- 无监督
- 异常检测
- OCR
- 检索
- human relighting
- 奇异值检测(Novelty Detection)
- Multi-attribute Learning
- 偏见识别
- 新类别发现(Novel Class Discovery)
- 密集预测
- 变分自动编码器(VAEs)
- 开集识别
- 草图
- Visual Grounding
- 互动结构理解
- HDR全景图生成
- 手语识别
- Panoptic Scene Graph Generation
⭐code🏠project - Meta Spatio-Temporal Debiasing for Video Scene Graph Generation
- 视听分割
- 语音合成
- 声音分离
- CCPL: Contrastive Coherence Preserving Loss for Versatile Style Transfer
😮oral⭐code - Learning Graph Neural Networks for Image Style Transfer
- 图像风格化
- InfiniteNature-Zero: Learning Perpetual View Generation of Natural Scenes from Single Images
😮oral - CompNVS: Novel View Synthesis with Scene Completion
- COO: Comic Onomatopoeia Dataset for Recognizing Arbitrary or Truncated Texts
⭐code
用于识别任意或截断文本的漫画拟声词数据集 - BRACE: The Breakdancing Competition Dataset for Dance Motion Synthesis
🌻dataset
用于舞蹈动作合成的霹雳舞比赛数据集 - CelebV-HQ: A Large-Scale Video Facial Attributes Dataset
🌻dataset🏠project
一个大规模的视频人脸属性数据集 - 数据集
- UnrealEgo: A New Dataset for Robust Egocentric 3D Human Motion Capture
⭐code🏠project
用于鲁棒性以自我为中心的三维人类运动捕捉的新数据集
- UnrealEgo: A New Dataset for Robust Egocentric 3D Human Motion Capture
- Bi-PointFlowNet: Bidirectional Learning for Point Cloud Based Scene Flow Estimation
⭐code - What Matters for 3D Scene Flow Network
⭐code
- Registration based Few-Shot Anomaly Detection
😮oral⭐code - Dynamic Local Aggregation Network with Adaptive Clusterer for Anomaly Detection
⭐code - HaloAE: An HaloNet based Local Transformer Auto-Encoder for Anomaly Detection and Localization
- 表面异常检测
- Relighting4D: Neural Relightable Human from Videos
⭐code🏠project📺video - MPIB: An MPI-Based Bokeh Rendering Framework for Realistic Partial Occlusion Effects
⭐code🏠project - Approximate Differentiable Rendering with Algebraic Surfaces
⭐code🏠project - AdaNeRF: Adaptive Sampling for Real-time Rendering of Neural Radiance Fields
⭐code🏠project - Generalizable Patch-Based Neural Rendering
😮oral⭐code🏠project - Deforming Radiance Fields with Cages
⭐code🏠project - NeuMesh: Learning Disentangled Neural Mesh-based Implicit Field for Geometry and Texture Editing
😮oral⭐code🏠project
- 小样本
- 零样本
- 域适应
- Prior Knowledge Guided Unsupervised Domain Adaptation
⭐code - CoSMix: Compositional Semantic Mix for Domain Adaptation in 3D LiDAR Segmentation
⭐code - GIPSO: Geometrically Informed Propagation for Online Adaptation in 3D LiDAR Segmentation
⭐code - Prototype-Guided Continual Adaptation for Class-Incremental Unsupervised Domain Adaptation
⭐code - MemSAC: Memory Augmented Sample Consistency for Large Scale Domain Adaptation
⭐code🏠project - Concurrent Subsidiary Supervision for Unsupervised Source-Free Domain Adaptation
⭐code🏠project
- Prior Knowledge Guided Unsupervised Domain Adaptation
- 域泛化
- Balancing Stability and Plasticity through Advanced Null Space in Continual Learning
😮oral - Online Continual Learning with Contrastive Vision Transformer
- Learning with Recoverable Forgetting
- Incremental Task Learning with Incremental Rank Updates
⭐code - 类增量
- Prior-Guided Adversarial Initialization for Fast Adversarial Training
⭐code - Decoupled Adversarial Contrastive Learning for Self-supervised Adversarial Robustness
😮oral⭐code - 对抗攻击
- Network Binarization via Contrastive Learning
- Adversarial Contrastive Learning via Asymmetric InfoNCE
⭐code - Fast-MoCo: Boost Momentum-based Contrastive Learning with Combinatorial Patches
⭐code
- DenseHybrid: Hybrid Anomaly Detection for Dense Open-set Recognition
- Difficulty-Aware Simulator for Open Set Recognition
⭐code
- 知识蒸馏
- 量化
- 剪枝
- Few 'Zero Level Set'-Shot Learning of Shape Signed Distance Functions in Feature Space
- Dynamic 3D Scene Analysis by Point Cloud Accumulation
⭐code🏠project - 点云定位
- 点云分割
- Dual Adaptive Transformations for Weakly Supervised Point Cloud Segmentation
- 点云补全
- 点云配准
- VR
- human volumetric capture(容积捕获)
- 虚拟试穿
- 视觉定位(相机姿势估计)
- Secrets of Event-Based Optical Flow
⭐code - Deep 360∘ Optical Flow Estimation Based on Multi-Projection Fusion
- Learning Omnidirectional Flow in 360-degree Video via Siamese Representation
🏠project
- 重识别
- 行人搜索
- ERA: Expert Retrieval and Assembly for Early Action Prediction
- Overlooked Poses Actually Make Sense: Distilling Privileged Knowledge for Human Motion Prediction
- SuperTickets: Drawing Task-Agnostic Lottery Tickets from Supernets via Jointly Architecture Searching and Parameter Pruning
⭐code - UniNet: Unified Architecture Search with Convolution, Transformer, and MLP
⭐code - ScaleNet: Searching for the Model to Scale
⭐code - CLOSE: Curriculum Learning On the Sharing Extent Towards Better One-shot NAS
⭐code
- Exploring Fine-Grained Audiovisual Categorization with the SSW60 Dataset
⭐code - Centrality and Consistency: Two-Stage Clean Samples Identification for Learning with Instance-Dependent Noisy Labels
⭐code - Constructing Balance from Imbalance for Long-tailed Image Recognition
⭐code - 小样本图像分类
- 长尾分类
- 视觉分类
- 细粒度识别
- 跨模态超分辨率
- 图像超分辨率
- 视频超分辨率
- 车辆轨迹预测
- 自动驾驶
- 轨迹预测
- 车道线检测
- 行人轨迹预测
- 医学图像分割
- 放射科报告生成
- 密集预测
- retinal image matching(视网膜图像匹配)
- 支架追踪
- 文本识别
- 手写数学表达式识别
- 场景文本检测
- Scene Text Recognition with Permuted Autoregressive Sequence Models
⭐code - Dynamic Low-Resolution Distillation for Cost-Efficient End-to-End Text Spotting
⭐code - SGBANet: Semantic GAN and Balanced Attention Network for Arbitrarily Oriented Scene Text Recognition
- Optimal Boxes: Boosting End-to-End Scene Text Recognition by Adjusting Annotated Bounding Boxes via Reinforcement Learning
- Contextual Text Block Detection towards Scene Text Understanding
🏠project - Toward Understanding WordArt: Corner-Guided Transformer for Scene Text Recognition
😮oral⭐code - GLASS: Global to Local Attention for Scene-Text Spotting
⭐code
- Scene Text Recognition with Permuted Autoregressive Sequence Models
- 视频文本检测
- 无监督
- 自监督
- 半监督
- 监督学习
- deepfake检测
- 三维人脸
- 活体检测
- 人脸识别
- 人脸聚类
- 谈话头像合成
- Injecting 3D Perception of Controllable NeRF-GAN into StyleGAN for Editable Portrait Image Synthesis
⭐code🏠project - Auto-regressive Image Synthesis with Integrated Quantization
😮oral - 图像生成
- 样本引导下的图像生成
- VecGAN: Image-to-Image Translation with Interpretable Latent Directions
- Vector Quantized Image-to-Image Translation
⭐code🏠project - 图像翻译
- RepMix: Representation Mixing for Robust Attribution of Synthesized Images
⭐code - FakeCLR: Exploring Contrastive Learning for Solving Latent Discontinuity in Data-Efficient GANs
⭐code - Generative Multiplane Images: Making a 2D GAN 3D-Aware
⭐code🏠project - Generator Knows What Discriminator Should Learn in Unconditional GANs
⭐code - Hierarchical Semantic Regularization of Latent Spaces in StyleGANs
⭐code🏠project - 线稿上色
- 图像生成
- k-means Mask Transformer
⭐code - Outpainting by Queries
⭐code - Locality Guidance for Improving Vision Transformers on Tiny Datasets
⭐code - TinyViT: Fast Pretraining Distillation for Small Vision Transformers
⭐code - MeshMAE: Masked Autoencoders for 3D Mesh Data Analysis
- An Impartial Take to the CNN vs Transformer Robustness Contest
- FashionViL: Fashion-Focused Vision-and-Language Representation Learning
⭐code - NewsStories: Illustrating articles with visual summaries
⭐code - Can Shuffling Video Benefit Temporal Bias Problem: A Novel Training Framework for Temporal Grounding
⭐code - Frozen CLIP Models are Efficient Video Learners
⭐code - 视觉表征学习
- Weakly Supervised Grounding for VQA in Vision-Language Transformers
⭐code - Rethinking Data Augmentation for Robust Visual Question Answering
⭐code - Video Question Answering with Iterative Video-Text Co-Tokenization
⭐code - Video-QA
- Towards Hard-Positive Query Mining for DETR-based Human-Object Interaction Detection
⭐code - Geometric Features Informed Multi-person Human-object Interaction Recognition in Videos
- IGFormer: Interaction Graph Transformer for Skeleton-based Human Interaction Recognition
- Mining Cross-Person Cues for Body-Part Interactiveness Learning in HOI Detection
⭐code - 交互式物体分割
- HOS
- 动作识别
- Collaborating Domain-shared and Target-specific Feature Clustering for Cross-domain 3D Action Recognition
- Hierarchically Self-Supervised Transformer for Human Skeleton Representation Learning
⭐code - An Efficient Spatio-Temporal Pyramid Transformer for Action Detection
- Spatiotemporal Self-attention Modeling with Temporal Patch Shift for Action Recognition
⭐code
- Privacy-Preserving Action Recognition via Motion Difference Quantization
⭐code - 基于骨架动作识别
- 小样本动作识别
- 社会群体活动识别
- 时序动作检测
- Semi-Supervised Temporal Action Detection with Proposal-Free Masking
⭐code - Temporal Action Detection with Global Segmentation Mask Learning
⭐code - ReAct: Temporal Action Detection with Relational Queries
⭐code - Zero-Shot Temporal Action Detection via Vision-Language Prompting
⭐code - Weakly-Supervised Temporal Action Detection for Fine-Grained Videos with Hierarchical Atomic Actions
⭐code
- Semi-Supervised Temporal Action Detection with Proposal-Free Masking
- Action Quality Assessment(行动质量评估)
- 视频-视频合成
- 视频生成
- 视频质量评估
- 视频修复
- 视频去模糊
- 视频对话
- 有源扬声器检测(视频会议)
- VOS
- XMem: Long-Term Video Object Segmentation with an Atkinson-Shiffrin Memory Model
⭐code🏠project📺video - Tackling Background Distraction in Video Object Segmentation
⭐code - Hierarchical Feature Alignment Network for Unsupervised Video Object Segmentation
⭐code - Learning Quality-aware Dynamic Memory for Video Object Segmentation
⭐code
- XMem: Long-Term Video Object Segmentation with an Atkinson-Shiffrin Memory Model
- VIS
- VSS
- 视频抠图
- 视频表征
- 视频传输
- 运动分割
- 视频异常检测
- 视频识别
- 视频理解
- 视频分类
- 视频卷帘快门(Rolling shutter)
- Video Transition Effects(视频转场特效)
- 视频编解码
- 物体姿势
- 抓取物体姿势估计
- 6D
- 9D
- Self-Constrained Inference Optimization on Structural Groups for Human Pose Estimation
- Pose for Everything: Towards Category-Agnostic Pose Estimation
😮oral⭐code - 运动捕捉
- 基于点的衣着人体建模
- 动态人体数字化
- 人体姿势与形状估计
- 三维人体姿势估计
- 三维人体重建
- 三维交互式手部姿势估计
- 姿势合成
- 手物重建
- 人体与场景的交互
- 人体姿势建模
- 姿势跟踪
- 三维人体网格恢复
- 三维人体运动预测
- 姿势迁移
- DeepPS2: Revisiting Photometric Stereo Using Two Differently Illuminated Images
- Towards High-Fidelity Single-view Holistic Reconstruction of Indoor Scenes
⭐code - Self-calibrating Photometric Stereo by Neural Inverse Rendering
⭐code - MVS
- 3D场景合成
- 场景重建
- 深度估计
- Physical Attack on Monocular Depth Estimation with Optimal Adversarial Patches
- Towards Scale-Aware, Robust, and Generalizable Unsupervised Monocular Depth Estimation by Integrating IMU Motion Dynamics
⭐code - JPerceiver: Joint Perception Network for Depth, Pose and Layout Estimation in Driving Scenes
⭐code - RA-Depth: Resolution Adaptive Self-Supervised Monocular Depth Estimation
⭐code - PointFix: Learning to Fix Domain Bias for Robust Online Stereo Adaptation
- Depth Field Networks for Generalizable Multi-view Scene Representation
🏠project - Gradient-based Uncertainty for Monocular Depth Estimation
⭐code
- 三维视觉
- 三维房间布局
- 三维重建
- Object-Compositional Neural Implicit Surfaces
⭐code🏠project📺video - Perspective Phase Angle Model for Polarimetric 3D Reconstruction
⭐code - Monocular 3D Object Reconstruction with GAN Inversion
⭐code🏠project - Structural Causal 3D Reconstruction
- 2D GANs Meet Unsupervised Single-view 3D Reconstruction
⭐code🏠project - Few-shot Single-view 3D Reconstruction with Memory Prior Contrastive Network
- PlaneFormers: From Sparse View Planes to 3D Reconstruction
⭐code🏠project📺video
- Object-Compositional Neural Implicit Surfaces
- 三维形状
- depth restoration
- Towards Grand Unification of Object Tracking
😮oral⭐code
📰ECCV 2022 Oral《Unicorn》首次统一了四项目标跟踪任务的网络结构与学习范式,在8个富有挑战性的数据集上SOTA - 3D跟踪
- 多目标跟踪
- Tracking Objects as Pixel-wise Distributions
😮oral - The Caltech Fish Counting Dataset: A Benchmark for Multiple-Object Tracking and Counting
- MOTCOM: The Multi-Object Tracking Dataset Complexity Metric
⭐code🏠project - Tracking Every Thing in the Wild
- PolarMOT: How Far Can Geometric Relations Take Us in 3D Multi-Object Tracking?
- SOMPT22: A Surveillance Oriented Multi-Pedestrian Tracking Dataset
- Robust Multi-Object Tracking by Marginal Inference
- Tracking Objects as Pixel-wise Distributions
- 视觉跟踪
- Should All Proposals be Treated Equally in Object Detection?
⭐code - HEAD: HEtero-Assists Distillation for Heterogeneous Object Detectors
⭐code - Adversarially-Aware Robust Object Detector
😮oral⭐code - ObjectBox: From Centers to Boxes for Anchor-Free Object Detection
😮oral⭐code - Point-to-Box Network for Accurate Object Detection via Single Point Supervision
⭐code - You Should Look at All Objects
⭐code - Class-agnostic Object Detection with Multi-modal Transformer
⭐code
使用多模态 ViTs 和人类可理解的文本查询来生成高质量的OP - Exploiting Unlabeled Data with Vision and Language Models for Object Detection
⭐code - PoserNet: Refining Relative Camera Poses Exploiting Object Detections
⭐code - Robust Object Detection With Inaccurate Bounding Boxes
⭐code - UC-OWOD: Unknown-Classified Open World Object Detection
⭐code - Exploring Resolution and Degradation Clues as Self-supervised Signal for Low Quality Object
⭐code - 3D目标检测
- DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection
⭐code - Rethinking IoU-based Optimization for Single-stage 3D Object Detection
⭐code - Densely Constrained Depth Estimator for Monocular 3D Object Detection
⭐code - AutoAlignV2: Deformable Feature Aggregation for Dynamic Multi-Modal 3D Object Detection
⭐code - DEVIANT: Depth EquiVarIAnt NeTwork for Monocular 3D Object Detection
⭐code - Label-Guided Auxiliary Training Improves 3D Object Detector
⭐code - Monocular 3D Object Detection with Depth from Motion
😮oral⭐code - MV-FCOS3D++: Multi-View Camera-Only 4D Object Detection with Pretrained Monocular Backbones
😮oral⭐code - Graph R-CNN: Towards Accurate 3D Object Detection with Semantic-Decorated Local Graph
😮oral⭐code
- DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection
- 半监督目标检测
- 小样本目标检测
- 显著目标检测
- 弱监督目标检测
- 目标定位
- 单阶目标检测
- 目标计数
- OOD
- 跨域检索
- 图像检索
- 视频检索
- LocVTP: Video-Text Pre-training for Temporal Localization
⭐code - Video Geo-localization(检索)
- LocVTP: Video-Text Pre-training for Temporal Localization
- 文本-视频检索
- 图像字幕
- 图像质量评估
- 图像修补(image retouching)
- 图像变形(Image Warping)
- 图像恢复
- 图像修复
- 图像增强
- 图像和谐化
- 去噪
- 去雪
- 去模糊
- 去摩尔纹
- 语义图像编辑
- PseudoClick: Interactive Image Segmentation with Click Imitation
- 语义分割
- 2DPASS: 2D Priors Assisted Semantic Segmentation on LiDAR Point Clouds
⭐code - Open-world Semantic Segmentation via Contrasting and Clustering Vision-Language Embedding
- ML-BPM: Multi-teacher Learning with Bidirectional Photometric Mixing for Open Compound Domain Adaptation in Semantic Segmentation
- 域适应语义分割
- 小样本语义分割
- 2DPASS: 2D Priors Assisted Semantic Segmentation on LiDAR Point Clouds
- 实例分割
- 小样本分割
- 抠图
- Differentiable Rendering for Synthetic Aperture Radar Imagery
- Batch-efficient EigenDecomposition for Small and Medium Matrices
⭐code - Accelerating Score-based Generative Models with Preconditioned Diffusion Sampling
- Improving Covariance Conditioning of the SVD Meta-layer by Orthogonality
⭐code - Contrastive Deep Supervision
⭐code - Organic Priors in Non-Rigid Structure from Motion
😮oral - Bootstrapped Masked Autoencoders for Vision BERT Pretraining
⭐code - Lipschitz Continuity Retained Binary Neural Network
⭐code - NeFSAC: Neurally Filtered Minimal Samples
⭐code - Towards Understanding The Semidefinite Relaxations of Truncated Least-Squares in Robust Rotation Search
- Latency-Aware Collaborative Perception
⭐code - MHR-Net: Multiple-Hypothesis Reconstruction of Non-Rigid Shapes from 2D Views
- SelectionConv: Convolutional Neural Networks for Non-rectilinear Image Data
- Overcoming Shortcut Learning in a Target Domain by Generalizing Basic Visual Factors from a Source Domain
⭐code - Discrete-Constrained Regression for Local Counting Models
- On the Versatile Uses of Partial Distance Correlation in Deep Learning
⭐code - Streamable Neural Fields
⭐code - Contributions of Shape, Texture, and Color in Visual Recognition
⭐code - Single Frame Atmospheric Turbulence Mitigation: A Benchmark Study and A New Physics-Inspired Transformer Model
- Latent Discriminant deterministic Uncertainty
⭐code - SPIN: An Empirical Evaluation on Sharing Parameters of Isotropic Networks
⭐code - UFO: Unified Feature Optimization
⭐code - POP: Mining POtential Performance of new fashion products via webly cross-modal query expansion
⭐code - My View is the Best View: Procedure Learning from Egocentric Videos
⭐code🏠project - Equivariance and Invariance Inductive Bias for Learning from Insufficient Data
⭐code - Contrastive Monotonic Pixel-Level Modulation
😮oral⭐code - Neural-Sim: Learning to Generate Training Data with NeRF
⭐code - Learning Hierarchy Aware Features for Reducing Mistake Severity
⭐code - Translating a Visual LEGO Manual to a Machine-Executable Plan
⭐code🏠project - Hardly Perceptible Trojan Attack against Neural Networks with Bit Flips
⭐code - LGV: Boosting Adversarial Example Transferability from Large Geometric Vicinity
- MonteBoxFinder: Detecting and Filtering Primitives to Fit a Noisy Point Cloud
⭐code🏠project - Neural Strands: Learning Hair Geometry and Appearance from Multi-View Images
🏠project - A Repulsive Force Unit for Garment Collision Handling in Neural Networks
🏠project - Minimal Neural Atlas: Parameterizing Complex Surfaces with Minimal Charts and Distortion
⭐code - Revisiting the Critical Factors of Augmentation-Invariant Representation Learning
⭐code - Fast Two-step Blind Optical Aberration Correction
⭐code - Transformers as Meta-Learners for Implicit Neural Representations
⭐code🏠project - Neighborhood Collective Estimation for Noisy Label Identification and Correction
⭐code