MasterBin-IIAU / ICCV2021-Papers-with-Code

ICCV 2021 论文和开源项目合集

ICCV2021-Papers-with-Code

ICCV 2021 论文和开源项目合集(papers with code)！

1617 papers accepted - 25.9% acceptance rate

ICCV 2021 收录论文IDs：https://docs.google.com/spreadsheets/u/1/d/e/2PACX-1vRfaTmsNweuaA0Gjyu58H_Cx56pGwFhcTYII0u1pg0U7MbhlgY0R6Y-BbK3xFhAiwGZ26u3TAtN5MnS/pubhtml

注1：欢迎各位大佬提交issue，分享ICCV 2021论文和开源项目！

注2：关于往年CV顶会论文以及其他优质CV论文和大盘点，详见： https://github.com/amusi/daily-paper-computer-vision

【ICCV 2021 论文和开源目录】

• Backbone
• Transformer
• GAN
• NAS
• NeRF
• Loss
• Zero-Shot Learning
• Few-Shot Learning
• 长尾(Long-tailed)
• 无监督/自监督(Self-Supervised)
• 2D目标检测(Object Detection)
• 语义分割(Semantic Segmentation)
• 实例分割(Instance Segmentation)
• 医学图像分割(Medical Image Segmentation)
• Few-shot Segmentation
• 人体运动分割(Human Motion Segmentation)
• 目标跟踪(Object Tracking)
• 3D Point Cloud
• Point Cloud Object Detection(点云目标检测)
• Point Cloud Semantic Segmenation(点云语义分割)
• Point Cloud Denoising(点云去噪)
• Point Cloud Registration(点云配准)
• 超分辨率(Super-Resolution)
• 行人重识别(Person Re-identification)
• 2D/3D人体姿态估计(2D/3D Human Pose Estimation)
• 3D人头重建(3D Head Reconstruction)
• 行为识别(Action Recognition)
• 时序动作定位(Temporal Action Localization)
• 文本检测(Text Detection)
• 文本识别(Text Recognition)
• 视觉问答(Visual Question Answering, VQA)
• 对抗攻击(Adversarial Attack)
• 深度估计(Depth Estimation)
• 视线估计(Gaze Estimation)
• 人群计数(Crowd Counting)
• 轨迹预测(Trajectory Prediction)
• 异常检测(Anomaly Detection)
• 场景图生成(Scene Graph Generation)
• Unsupervised Domain Adaptation
• Video Rescaling
• Hand-Object Interaction
• 数据集(Datasets)
• 其他(Others)

Backbone

Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions

• Paper(Oral): https://arxiv.org/abs/2102.12122
• Code: https://github.com/whai362/PVT

AutoFormer: Searching Transformers for Visual Recognition

• Paper: https://arxiv.org/abs/2107.00651
• Code: https://github.com/microsoft/AutoML

Bias Loss for Mobile Neural Networks

• Paper: https://arxiv.org/abs/2107.11170
• Code: None

Visual Transformer

Swin Transformer: Hierarchical Vision Transformer using Shifted Windows

• Paper: https://arxiv.org/abs/2103.14030
• Code: https://github.com/microsoft/Swin-Transformer

An Empirical Study of Training Self-Supervised Vision Transformers

• Paper(Oral): https://arxiv.org/abs/2104.02057
• MoCo v3 Code: None

Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions

• Paper(Oral): https://arxiv.org/abs/2102.12122
• Code: https://github.com/whai362/PVT

Group-Free 3D Object Detection via Transformers

• Paper: https://arxiv.org/abs/2104.00678
• Code: None

Spatial-Temporal Transformer for Dynamic Scene Graph Generation

• Paper: https://arxiv.org/abs/2107.12309
• Code: None

Rethinking and Improving Relative Position Encoding for Vision Transformer

• Paper: https://arxiv.org/abs/2107.14222
• Code: None

Emerging Properties in Self-Supervised Vision Transformers

• Paper: https://arxiv.org/abs/2104.14294
• Code: https://github.com/facebookresearch/dino

GAN

Labels4Free: Unsupervised Segmentation using StyleGAN

• Homepage: https://rameenabdal.github.io/Labels4Free/
• Paper: https://arxiv.org/abs/2103.14968

GNeRF: GAN-based Neural Radiance Field without Posed Camera

• Paper(Oral): https://arxiv.org/abs/2103.15606

• Code: https://github.com/MQ66/gnerf

EigenGAN: Layer-Wise Eigen-Learning for GANs

• Paper: https://arxiv.org/abs/2104.12476
• Code: https://github.com/LynnHo/EigenGAN-Tensorflow

From Continuity to Editability: Inverting GANs with Consecutive Images

• Paper: https://arxiv.org/abs/2107.13812
• Code: https://github.com/Qingyang-Xu/InvertingGANs_with_ConsecutiveImgs

NAS

AutoFormer: Searching Transformers for Visual Recognition

• Paper: https://arxiv.org/abs/2107.00651

• Code: https://github.com/microsoft/AutoML

NeRF

GNeRF: GAN-based Neural Radiance Field without Posed Camera

• Paper(Oral): https://arxiv.org/abs/2103.15606

• Code: https://github.com/MQ66/gnerf

KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPs

• Paper: https://arxiv.org/abs/2103.13744

• Code: https://github.com/creiser/kilonerf

In-Place Scene Labelling and Understanding with Implicit Scene Representation

• Homepage: https://shuaifengzhi.com/Semantic-NeRF/
• Paper(Oral): https://arxiv.org/abs/2103.15875

Putting NeRF on a Diet: Semantically Consistent Few-Shot View Synthesis

• Homepage: https://ajayj.com/dietnerf
• Paper(DietNeRF): https://arxiv.org/abs/2104.00677

Loss

Rank & Sort Loss for Object Detection and Instance Segmentation

• Paper(Oral): https://arxiv.org/abs/2107.11669
• Code: https://github.com/kemaloksuz/RankSortLoss

Bias Loss for Mobile Neural Networks

• Paper: https://arxiv.org/abs/2107.11170
• Code: None

Zero-Shot Learning

FREE: Feature Refinement for Generalized Zero-Shot Learning

• Paper: https://arxiv.org/abs/2107.13807
• Code: https://github.com/shiming-chen/FREE

Few-Shot Learning

Few-Shot and Continual Learning with Attentive Independent Mechanisms

• Paper: https://arxiv.org/abs/2107.14053
• Code: https://github.com/huang50213/AIM-Fewshot-Continual

长尾(Long-tailed)

Parametric Contrastive Learning

• Paper: https://arxiv.org/abs/2107.12028
• Code: https://github.com/jiequancui/Parametric-Contrastive-Learning

无监督/自监督(Un/Self-Supervised)

An Empirical Study of Training Self-Supervised Vision Transformers

• Paper(Oral): https://arxiv.org/abs/2104.02057
• MoCo v3 Code: None

DetCo: Unsupervised Contrastive Learning for Object Detection

• Paper: https://arxiv.org/abs/2102.04803

• Code: https://github.com/xieenze/DetCo

2D目标检测(Object Detection)

DetCo: Unsupervised Contrastive Learning for Object Detection

• Paper: https://arxiv.org/abs/2102.04803
• Code: https://github.com/xieenze/DetCo

Detecting Invisible People

• Homepage: http://www.cs.cmu.edu/~tkhurana/invisible.htm
• Code: https://arxiv.org/abs/2012.08419

Active Learning for Deep Object Detection via Probabilistic Modeling

• Paper: https://arxiv.org/abs/2103.16130
• Code: None

Conditional Variational Capsule Network for Open Set Recognition

• Paper: https://arxiv.org/abs/2104.09159
• Code: https://github.com/guglielmocamporese/cvaecaposr

MDETR : Modulated Detection for End-to-End Multi-Modal Understanding

• Homepage: https://ashkamath.github.io/mdetr_page/
• Paper(Oral): https://arxiv.org/abs/2104.12763
• Code: https://github.com/ashkamath/mdetr

Rank & Sort Loss for Object Detection and Instance Segmentation

• Paper(Oral): https://arxiv.org/abs/2107.11669
• Code: https://github.com/kemaloksuz/RankSortLoss

SimROD: A Simple Adaptation Method for Robust Object Detection

• Paper(Oral): https://arxiv.org/abs/2107.13389
• Code: None

GraphFPN: Graph Feature Pyramid Network for Object Detection

• Paper: https://arxiv.org/abs/2108.00580
• Code: None

半监督目标检测

End-to-End Semi-Supervised Object Detection with Soft Teacher

• Paper: https://arxiv.org/abs/2106.09018
• Code: None

语义分割(Semantic Segmentation)

Personalized Image Semantic Segmentation

• Paper: https://arxiv.org/abs/2107.13978
• Code: https://github.com/zhangyuygss/PIS
• Dataset: https://github.com/zhangyuygss/PIS

Standardized Max Logits: A Simple yet Effective Approach for Identifying Unexpected Road Obstacles in Urban-Scene Segmentation

• Paper(Oral): https://arxiv.org/abs/2107.11264
• Code: None

半监督语义分割(Semi-supervised Semantic Segmentation)

Leveraging Auxiliary Tasks with Affinity Learning for Weakly Supervised Semantic Segmentation

• Paper: https://arxiv.org/abs/2107.11787
• Code: None

Re-distributing Biased Pseudo Labels for Semi-supervised Semantic Segmentation: A Baseline Investigation

• Paper(Oral): https://arxiv.org/abs/2107.11279
• Code: https://github.com/CVMI-Lab/DARS

无监督分割(Unsupervised Segmentation)

Labels4Free: Unsupervised Segmentation using StyleGAN

• Homepage: https://rameenabdal.github.io/Labels4Free/
• Paper: https://arxiv.org/abs/2103.14968

实例分割(Instance Segmentation)

Instances as Queries

• Paper: https://arxiv.org/abs/2105.01928
• Code: https://github.com/hustvl/QueryInst

Crossover Learning for Fast Online Video Instance Segmentation

• Paper: https://arxiv.org/abs/2104.05970
• Code: https://github.com/hustvl/CrossVIS

Rank & Sort Loss for Object Detection and Instance Segmentation

• Paper(Oral): https://arxiv.org/abs/2107.11669
• Code: https://github.com/kemaloksuz/RankSortLoss

医学图像分割(Medical Image Segmentation)

Recurrent Mask Refinement for Few-Shot Medical Image Segmentation

• Paper: https://arxiv.org/abs/2108.00622
• Code: https://github.com/uci-cbcl/RP-Net

Few-shot Segmentation

Mining Latent Classes for Few-shot Segmentation

• Paper(Oral): https://arxiv.org/abs/2103.15402
• Code: https://github.com/LiheYoung/MiningFSS

人体运动分割(Human Motion Segmentation)

Graph Constrained Data Representation Learning for Human Motion Segmentation

• Paper: https://arxiv.org/abs/2107.13362
• Code: None

目标跟踪(Object Tracking)

Learning Spatio-Temporal Transformer for Visual Tracking

• Paper: https://arxiv.org/abs/2103.17154
• Code: https://github.com/researchmm/Stark

Learning to Adversarially Blur Visual Object Tracking

• Paper: https://arxiv.org/abs/2107.12085
• Code: https://github.com/tsingqguo/ABA

HiFT: Hierarchical Feature Transformer for Aerial Tracking

• Paper: https://arxiv.org/abs/2108.00202
• Code: https://github.com/vision4robotics/HiFT

Learn to Match: Automatic Matching Network Design for Visual Tracking

• Paper: https://arxiv.org/abs/2108.00803
• Code: https://github.com/JudasDie/SOTS

3D Point Cloud

Unsupervised Point Cloud Pre-Training via View-Point Occlusion, Completion

• Homepage: https://hansen7.github.io/OcCo/
• Paper: https://arxiv.org/abs/2010.01089
• Code: https://github.com/hansen7/OcCo

Point Cloud Object Detection(点云目标检测)

Group-Free 3D Object Detection via Transformers

• Paper: https://arxiv.org/abs/2104.00678
• Code: None

Point Cloud Semantic Segmentation(点云语义分割)

ReDAL: Region-based and Diversity-aware Active Learning for Point Cloud Semantic Segmentation

• Paper: https://arxiv.org/abs/2107.11769
• Code: None

Learning with Noisy Labels for Robust Point Cloud Segmentation

• Homepage: https://shuquanye.com/PNAL_website/
• Paper(Oral): https://arxiv.org/abs/2107.14230

VMNet: Voxel-Mesh Network for Geodesic-Aware 3D Semantic Segmentation

• Paper(Oral): https://arxiv.org/abs/2107.13824
• Code: https://github.com/hzykent/VMNet

Point Cloud Denoising(点云去噪)

Score-Based Point Cloud Denoising

• Paper: https://arxiv.org/abs/2107.10981
• Code: None

Point Cloud Registration(点云配准)

HRegNet: A Hierarchical Network for Large-scale Outdoor LiDAR Point Cloud Registration

• Homepage: https://ispc-group.github.io/hregnet
• Paper: https://arxiv.org/abs/2107.11992
• Code: https://github.com/ispc-lab/HRegNet

超分辨率(Super-Resolution)

Learning for Scale-Arbitrary Super-Resolution from Scale-Specific Networks

• Paper: https://arxiv.org/abs/2004.03791
• Code: https://github.com/LongguangWang/ArbSR

行人重识别(Person Re-identification)

TransReID: Transformer-based Object Re-Identification

• Paper: https://arxiv.org/abs/2102.04378
• Code: https://github.com/heshuting555/TransReID

2D/3D人体姿态估计(2D/3D Human Pose Estimation)

2D 人体姿态估计

Human Pose Regression with Residual Log-likelihood Estimation

• Paper(Oral): https://arxiv.org/abs/2107.11291
• Code(RLE): https://github.com/Jeff-sjtu/res-loglikelihood-regression

3D 人体姿态估计

Probabilistic Monocular 3D Human Pose Estimation with Normalizing Flows

• Paper: https://arxiv.org/abs/2107.13788
• Code: https://github.com/twehrbein/Probabilistic-Monocular-3D-Human-Pose-Estimation-with-Normalizing-Flows

3D人头重建(3D Head Reconstruction)

H3D-Net: Few-Shot High-Fidelity 3D Head Reconstruction

• Homepage: https://crisalixsa.github.io/h3d-net/

• Paper: https://arxiv.org/abs/2107.12512

行为识别(Action Recognition)

MGSampler: An Explainable Sampling Strategy for Video Action Recognition

• Paper: https://arxiv.org/abs/2104.09952
• Code: None

时序动作定位(Temporal Action Localization)

Enriching Local and Global Contexts for Temporal Action Localization

• Paper: https://arxiv.org/abs/2107.12960
• Code: None

文本检测(Text Detection)

Adaptive Boundary Proposal Network for Arbitrary Shape Text Detection

• Paper: https://arxiv.org/abs/2107.12664
• Code: https://github.com/GXYM/TextBPN

文本识别(Text Recognition)

Joint Visual Semantic Reasoning: Multi-Stage Decoder for Text Recognition

• Paper: https://arxiv.org/abs/2107.12090
• Code: None

视觉问答(Visual Question Answering, VQA)

Greedy Gradient Ensemble for Robust Visual Question Answering

• Paper: https://arxiv.org/abs/2107.12651

• Code: https://github.com/GeraldHan/GGE

对抗攻击(Adversarial Attack)

Feature Importance-aware Transferable Adversarial Attacks

• Paper: https://arxiv.org/abs/2107.14185
• Code: https://github.com/hcguoO0/FIA

深度估计(Depth Estimation)

单目深度估计

MonoIndoor: Towards Good Practice of Self-Supervised Monocular Depth Estimation for Indoor Environments

• Paper: https://arxiv.org/abs/2107.12429
• Code: None

视线估计(Gaze Estimation)

Generalizing Gaze Estimation with Outlier-guided Collaborative Adaptation

• Paper: https://arxiv.org/abs/2107.13780
• Code: https://github.com/DreamtaleCore/PnP-GA

人群计数(Crowd Counting)

Rethinking Counting and Localization in Crowds:A Purely Point-Based Framework

• Paper(Oral): https://arxiv.org/abs/2107.12746
• Code(P2PNet): https://github.com/TencentYoutuResearch/CrowdCounting-P2PNet

Uniformity in Heterogeneity:Diving Deep into Count Interval Partition for Crowd Counting

• Paper: https://arxiv.org/abs/2107.12619
• Code: https://github.com/TencentYoutuResearch/CrowdCounting-UEPNet

轨迹预测(Trajectory Prediction)

Human Trajectory Prediction via Counterfactual Analysis

• Paper: https://arxiv.org/abs/2107.14202
• Code: https://github.com/CHENGY12/CausalHTP

Personalized Trajectory Prediction via Distribution Discrimination

• Paper: https://arxiv.org/abs/2107.14204
• Code: https://github.com/CHENGY12/DisDis

异常检测(Anomaly Detection)

Weakly-supervised Video Anomaly Detection with Robust Temporal Feature Magnitude Learning

• Paper: https://arxiv.org/abs/2101.10030
• Code: https://github.com/tianyu0207/RTFM

场景图生成(Scene Graph Generation)

Spatial-Temporal Transformer for Dynamic Scene Graph Generation

• Paper: https://arxiv.org/abs/2107.12309
• Code: None

Unsupervised Domain Adaptation

Recursively Conditional Gaussian for Ordinal Unsupervised Domain Adaptation

• Paper(Oral): https://arxiv.org/abs/2107.13467
• Code: None

Video Rescaling

Self-Conditioned Probabilistic Learning of Video Rescaling

• Paper: https://arxiv.org/abs/2107.11639

• Code: None

Hand-Object Interaction

Learning a Contact Potential Field to Model the Hand-Object Interaction

• Paper: https://arxiv.org/abs/2012.00924
• Code: https://lixiny.github.io/CPF

数据集(Datasets)

Personalized Image Semantic Segmentation

• Paper: https://arxiv.org/abs/2107.13978
• Code: https://github.com/zhangyuygss/PIS
• Dataset: https://github.com/zhangyuygss/PIS

H3D-Net: Few-Shot High-Fidelity 3D Head Reconstruction

• Homepage: https://crisalixsa.github.io/h3d-net/

• Paper: https://arxiv.org/abs/2107.12512

其他(Others)

Progressive Correspondence Pruning by Consensus Learning

• Homepage: https://sailor-z.github.io/projects/CLNet.html
• Paper: https://arxiv.org/abs/2101.00591
• Code: https://github.com/sailor-z/CLNet

项目主页：

Energy-Based Open-World Uncertainty Modeling for Confidence Calibration

• Paper: https://arxiv.org/abs/2107.12628
• Code: None

Generalized Shuffled Linear Regression

• Paper: https://drive.google.com/file/d/1Qu21VK5qhCW8WVjiRnnBjehrYVmQrDNh/view?usp=sharing
• Code: https://github.com/SILI1994/Generalized-Shuffled-Linear-Regression

Discovering 3D Parts from Image Collections

• Homepage: https://chhankyao.github.io/lpd/

• Paper: https://arxiv.org/abs/2107.13629

Semi-Supervised Active Learning with Temporal Output Discrepancy

• Paper: https://arxiv.org/abs/2107.14153
• Code: https://github.com/siyuhuang/TOD

Why Approximate Matrix Square Root Outperforms Accurate SVD in Global Covariance Pooling?

Paper: https://arxiv.org/abs/2105.02498

Code: https://github.com/KingJamesSong/DifferentiableSVD

Hand-Object Contact Consistency Reasoning for Human Grasps Generation

• Homepage: https://hwjiang1510.github.io/GraspTTA/
• Paper(Oral): https://arxiv.org/abs/2104.03304
• Code: None

Equivariant Imaging: Learning Beyond the Range Space

• Paper(Oral): https://arxiv.org/abs/2103.14756
• Code: https://github.com/edongdongchen/EI

Just Ask: Learning to Answer Questions from Millions of Narrated Videos

• Paper(Oral): https://arxiv.org/abs/2012.00451
• Code: https://github.com/antoyang/just-ask