CVPR 2020 papers

本文轉載自查看原文 2020-10-29 16:26 540

# CVPR2020-Code

[CVPR 2020](https://openaccess.thecvf.com/CVPR2020) 論文開源項目合集，同時歡迎各位大佬提交issue，分享CVPR 2020開源項目

**【推薦閱讀】**

- [CVPR 2020 virtual](http://cvpr20.com/)
- ECCV 2020 論文開源項目合集來了：https://github.com/amusi/ECCV2020-Code

- 關於往年CV頂會論文（如ECCV 2020、CVPR 2019、ICCV 2019）以及其他優質CV論文和大盤點，詳見： https://github.com/amusi/daily-paper-computer-vision

**【CVPR 2020 論文開源目錄】**

- [CNN](#CNN)
- [圖像分類](#Image-Classification)
- [視頻分類](#Video-Classification)
- [目標檢測](#Object-Detection)
- [3D目標檢測](#3D-Object-Detection)
- [視頻目標檢測](#Video-Object-Detection)
- [目標跟蹤](#Object-Tracking)
- [語義分割](#Semantic-Segmentation)
- [實例分割](#Instance-Segmentation)
- [全景分割](#Panoptic-Segmentation)
- [視頻目標分割](#VOS)
- [超像素分割](#Superpixel)
- [交互式圖像分割](#IIS)
- [NAS](#NAS)
- [GAN](#GAN)
- [Re-ID](#Re-ID)
- [3D點雲（分類/分割/配准/跟蹤等）](#3D-PointCloud)
- [人臉（識別/檢測/重建等）](#Face)
- [人體姿態估計(2D/3D)](#Human-Pose-Estimation)
- [人體解析](#Human-Parsing)
- [場景文本檢測](#Scene-Text-Detection)
- [場景文本識別](#Scene-Text-Recognition)
- [特征(點)檢測和描述](#Feature)
- [超分辨率](#Super-Resolution)
- [模型壓縮/剪枝](#Model-Compression)
- [視頻理解/行為識別](#Action-Recognition)
- [人群計數](#Crowd-Counting)
- [深度估計](#Depth-Estimation)
- [6D目標姿態估計](#6DOF)
- [手勢估計](#Hand-Pose)
- [顯著性檢測](#Saliency)
- [去噪](#Denoising)
- [去雨](#Deraining)
- [去模糊](#Deblurring)
- [去霧](#Dehazing)
- [特征點檢測與描述](#Feature)
- [視覺問答(VQA)](#VQA)
- [視頻問答(VideoQA)](#VideoQA)
- [視覺語言導航](#VLN)
- [視頻壓縮](#Video-Compression)
- [視頻插幀](#Video-Frame-Interpolation)
- [風格遷移](#Style-Transfer)
- [車道線檢測](#Lane-Detection)
- ["人-物"交互(HOI)檢測](#HOI)
- [軌跡預測](#TP)
- [運動預測](#Motion-Predication)
- [光流估計](#OF)
- [圖像檢索](#IR)
- [虛擬試衣](#Virtual-Try-On)
- [HDR](#HDR)
- [對抗樣本](#AE)
- [三維重建](#3D-Reconstructing)
- [深度補全](#DC)
- [語義場景補全](#SSC)
- [圖像/視頻描述](#Captioning)
- [線框解析](#WP)
- [數據集](#Datasets)
- [其他](#Others)
- [不確定中沒中](#Not-Sure)

# CNN

**Exploring Self-attention for Image Recognition**

- 論文：https://hszhao.github.io/papers/cvpr20_san.pdf

- 代碼：https://github.com/hszhao/SAN

**Improving Convolutional Networks with Self-Calibrated Convolutions**

- 主頁：https://mmcheng.net/scconv/

- 論文：http://mftp.mmcheng.net/Papers/20cvprSCNet.pdf
- 代碼：https://github.com/backseason/SCNet

**Rethinking Depthwise Separable Convolutions: How Intra-Kernel Correlations Lead to Improved MobileNets**

- 論文：https://arxiv.org/abs/2003.13549
- 代碼：https://github.com/zeiss-microscopy/BSConv

# 圖像分類

**Interpretable and Accurate Fine-grained Recognition via Region Grouping**

- 論文：https://arxiv.org/abs/2005.10411

- 代碼：https://github.com/zxhuang1698/interpretability-by-parts

**Compositional Convolutional Neural Networks: A Deep Architecture with Innate Robustness to Partial Occlusion**

- 論文：https://arxiv.org/abs/2003.04490

- 代碼：https://github.com/AdamKortylewski/CompositionalNets

**Spatially Attentive Output Layer for Image Classification**

- 論文：https://arxiv.org/abs/2004.07570
- 代碼（好像被原作者刪除了）：https://github.com/ildoonet/spatially-attentive-output-layer

# 視頻分類

**SmallBigNet: Integrating Core and Contextual Views for Video Classification**

- 論文：https://arxiv.org/abs/2006.14582
- 代碼：https://github.com/xhl-video/SmallBigNet

# 目標檢測

**Overcoming Classifier Imbalance for Long-tail Object Detection with Balanced Group Softmax**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/papers/Li_Overcoming_Classifier_Imbalance_for_Long-Tail_Object_Detection_With_Balanced_Group_CVPR_2020_paper.pdf
- 代碼：https://github.com/FishYuLi/BalancedGroupSoftmax

**AugFPN: Improving Multi-scale Feature Learning for Object Detection**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/papers/Guo_AugFPN_Improving_Multi-Scale_Feature_Learning_for_Object_Detection_CVPR_2020_paper.pdf
- 代碼：https://github.com/Gus-Guo/AugFPN

**Noise-Aware Fully Webly Supervised Object Detection**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/html/Shen_Noise-Aware_Fully_Webly_Supervised_Object_Detection_CVPR_2020_paper.html
- 代碼：https://github.com/shenyunhang/NA-fWebSOD/

**Learning a Unified Sample Weighting Network for Object Detection**

- 論文：https://arxiv.org/abs/2006.06568
- 代碼：https://github.com/caiqi/sample-weighting-network

**D2Det: Towards High Quality Object Detection and Instance Segmentation**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/papers/Cao_D2Det_Towards_High_Quality_Object_Detection_and_Instance_Segmentation_CVPR_2020_paper.pdf

- 代碼：https://github.com/JialeCao001/D2Det

**Dynamic Refinement Network for Oriented and Densely Packed Object Detection**

- 論文下載鏈接：https://arxiv.org/abs/2005.09973

- 代碼和數據集：https://github.com/Anymake/DRN_CVPR2020

**Scale-Equalizing Pyramid Convolution for Object Detection**

論文：https://arxiv.org/abs/2005.03101

代碼：https://github.com/jshilong/SEPC

**Revisiting the Sibling Head in Object Detector**

- 論文：https://arxiv.org/abs/2003.07540

- 代碼：https://github.com/Sense-X/TSD

**Scale-equalizing Pyramid Convolution for Object Detection**

- 論文：暫無
- 代碼：https://github.com/jshilong/SEPC

**Detection in Crowded Scenes: One Proposal, Multiple Predictions**

- 論文：https://arxiv.org/abs/2003.09163
- 代碼：https://github.com/megvii-model/CrowdDetection

**Instance-aware, Context-focused, and Memory-efficient Weakly Supervised Object Detection**

- 論文：https://arxiv.org/abs/2004.04725
- 代碼：https://github.com/NVlabs/wetectron

**Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selection**

- 論文：https://arxiv.org/abs/1912.02424
- 代碼：https://github.com/sfzhang15/ATSS

**BiDet: An Efficient Binarized Object Detector**

- 論文：https://arxiv.org/abs/2003.03961
- 代碼：https://github.com/ZiweiWangTHU/BiDet

**Harmonizing Transferability and Discriminability for Adapting Object Detectors**

- 論文：https://arxiv.org/abs/2003.06297
- 代碼：https://github.com/chaoqichen/HTCN

**CentripetalNet: Pursuing High-quality Keypoint Pairs for Object Detection**

- 論文：https://arxiv.org/abs/2003.09119
- 代碼：https://github.com/KiveeDong/CentripetalNet

**Hit-Detector: Hierarchical Trinity Architecture Search for Object Detection**

- 論文：https://arxiv.org/abs/2003.11818
- 代碼：https://github.com/ggjy/HitDet.pytorch

**EfficientDet: Scalable and Efficient Object Detection**

- 論文：https://arxiv.org/abs/1911.09070
- 代碼：https://github.com/google/automl/tree/master/efficientdet

# 3D目標檢測

**SESS: Self-Ensembling Semi-Supervised 3D Object Detection**

- 論文： https://arxiv.org/abs/1912.11803

- 代碼：https://github.com/Na-Z/sess

**Associate-3Ddet: Perceptual-to-Conceptual Association for 3D Point Cloud Object Detection**

- 論文： https://arxiv.org/abs/2006.04356

- 代碼：https://github.com/dleam/Associate-3Ddet

**What You See is What You Get: Exploiting Visibility for 3D Object Detection**

- 主頁：https://www.cs.cmu.edu/~peiyunh/wysiwyg/

- 論文：https://arxiv.org/abs/1912.04986
- 代碼：https://github.com/peiyunh/wysiwyg

**Learning Depth-Guided Convolutions for Monocular 3D Object Detection**

- 論文：https://arxiv.org/abs/1912.04799
- 代碼：https://github.com/dingmyu/D4LCN

**Structure Aware Single-stage 3D Object Detection from Point Cloud**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/html/He_Structure_Aware_Single-Stage_3D_Object_Detection_From_Point_Cloud_CVPR_2020_paper.html

- 代碼：https://github.com/skyhehe123/SA-SSD

**IDA-3D: Instance-Depth-Aware 3D Object Detection from Stereo Vision for Autonomous Driving**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/papers/Peng_IDA-3D_Instance-Depth-Aware_3D_Object_Detection_From_Stereo_Vision_for_Autonomous_CVPR_2020_paper.pdf

- 代碼：https://github.com/swords123/IDA-3D

**Train in Germany, Test in The USA: Making 3D Object Detectors Generalize**

- 論文：https://arxiv.org/abs/2005.08139

- 代碼：https://github.com/cxy1997/3D_adapt_auto_driving

**MLCVNet: Multi-Level Context VoteNet for 3D Object Detection**

- 論文：https://arxiv.org/abs/2004.05679
- 代碼：https://github.com/NUAAXQ/MLCVNet

**3DSSD: Point-based 3D Single Stage Object Detector**

- CVPR 2020 Oral

- 論文：https://arxiv.org/abs/2002.10187

- 代碼：https://github.com/tomztyang/3DSSD

**Disp R-CNN: Stereo 3D Object Detection via Shape Prior Guided Instance Disparity Estimation**

- 論文：https://arxiv.org/abs/2004.03572

- 代碼：https://github.com/zju3dv/disprcn

**End-to-End Pseudo-LiDAR for Image-Based 3D Object Detection**

- 論文：https://arxiv.org/abs/2004.03080

- 代碼：https://github.com/mileyan/pseudo-LiDAR_e2e

**DSGN: Deep Stereo Geometry Network for 3D Object Detection**

- 論文：https://arxiv.org/abs/2001.03398
- 代碼：https://github.com/chenyilun95/DSGN

**LiDAR-based Online 3D Video Object Detection with Graph-based Message Passing and Spatiotemporal Transformer Attention**

- 論文：https://arxiv.org/abs/2004.01389
- 代碼：https://github.com/yinjunbo/3DVID

**PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection**

- 論文：https://arxiv.org/abs/1912.13192

- 代碼：https://github.com/sshaoshuai/PV-RCNN

**Point-GNN: Graph Neural Network for 3D Object Detection in a Point Cloud**

- 論文：https://arxiv.org/abs/2003.01251
- 代碼：https://github.com/WeijingShi/Point-GNN

# 視頻目標檢測

**Memory Enhanced Global-Local Aggregation for Video Object Detection**

論文：https://arxiv.org/abs/2003.12063

代碼：https://github.com/Scalsol/mega.pytorch

# 目標跟蹤

**SiamCAR: Siamese Fully Convolutional Classification and Regression for Visual Tracking**

- 論文：https://arxiv.org/abs/1911.07241
- 代碼：https://github.com/ohhhyeahhh/SiamCAR

**D3S -- A Discriminative Single Shot Segmentation Tracker**

- 論文：https://arxiv.org/abs/1911.08862
- 代碼：https://github.com/alanlukezic/d3s

**ROAM: Recurrently Optimizing Tracking Model**

- 論文：https://arxiv.org/abs/1907.12006

- 代碼：https://github.com/skyoung/ROAM

**Siam R-CNN: Visual Tracking by Re-Detection**

- 主頁：https://www.vision.rwth-aachen.de/page/siamrcnn
- 論文：https://arxiv.org/abs/1911.12836
- 論文2：https://www.vision.rwth-aachen.de/media/papers/192/siamrcnn.pdf
- 代碼：https://github.com/VisualComputingInstitute/SiamR-CNN

**Cooling-Shrinking Attack: Blinding the Tracker with Imperceptible Noises**

- 論文：https://arxiv.org/abs/2003.09595
- 代碼：https://github.com/MasterBin-IIAU/CSA

**High-Performance Long-Term Tracking with Meta-Updater**

- 論文：https://arxiv.org/abs/2004.00305

- 代碼：https://github.com/Daikenan/LTMU

**AutoTrack: Towards High-Performance Visual Tracking for UAV with Automatic Spatio-Temporal Regularization**

- 論文：https://arxiv.org/abs/2003.12949

- 代碼：https://github.com/vision4robotics/AutoTrack

**Probabilistic Regression for Visual Tracking**

- 論文：https://arxiv.org/abs/2003.12565
- 代碼：https://github.com/visionml/pytracking

**MAST: A Memory-Augmented Self-supervised Tracker**

- 論文：https://arxiv.org/abs/2002.07793
- 代碼：https://github.com/zlai0/MAST

**Siamese Box Adaptive Network for Visual Tracking**

- 論文：https://arxiv.org/abs/2003.06761
- 代碼：https://github.com/hqucv/siamban

## 多目標跟蹤

**3D-ZeF: A 3D Zebrafish Tracking Benchmark Dataset**

- 主頁：https://vap.aau.dk/3d-zef/
- 論文：https://arxiv.org/abs/2006.08466
- 代碼：https://bitbucket.org/aauvap/3d-zef/src/master/
- 數據集：https://motchallenge.net/data/3D-ZeF20

# 語義分割

**FDA: Fourier Domain Adaptation for Semantic Segmentation**

- 論文：https://arxiv.org/abs/2004.05498

- 代碼：https://github.com/YanchaoYang/FDA

**Super-BPD: Super Boundary-to-Pixel Direction for Fast Image Segmentation**

- 論文：暫無

- 代碼：https://github.com/JianqiangWan/Super-BPD

**Single-Stage Semantic Segmentation from Image Labels**

- 論文：https://arxiv.org/abs/2005.08104

- 代碼：https://github.com/visinf/1-stage-wseg

**Learning Texture Invariant Representation for Domain Adaptation of Semantic Segmentation**

- 論文：https://arxiv.org/abs/2003.00867
- 代碼：https://github.com/MyeongJin-Kim/Learning-Texture-Invariant-Representation

**MSeg: A Composite Dataset for Multi-domain Semantic Segmentation**

- 論文：http://vladlen.info/papers/MSeg.pdf
- 代碼：https://github.com/mseg-dataset/mseg-api

**CascadePSP: Toward Class-Agnostic and Very High-Resolution Segmentation via Global and Local Refinement**

- 論文：https://arxiv.org/abs/2005.02551
- 代碼：https://github.com/hkchengrex/CascadePSP

**Unsupervised Intra-domain Adaptation for Semantic Segmentation through Self-Supervision**

- Oral
- 論文：https://arxiv.org/abs/2004.07703
- 代碼：https://github.com/feipan664/IntraDA

**Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentation**

- 論文：https://arxiv.org/abs/2004.04581
- 代碼：https://github.com/YudeWang/SEAM

**Temporally Distributed Networks for Fast Video Segmentation**

- 論文：https://arxiv.org/abs/2004.01800

- 代碼：https://github.com/feinanshan/TDNet

**Context Prior for Scene Segmentation**

- 論文：https://arxiv.org/abs/2004.01547

- 代碼：https://git.io/ContextPrior

**Strip Pooling: Rethinking Spatial Pooling for Scene Parsing**

- 論文：https://arxiv.org/abs/2003.13328

- 代碼：https://github.com/Andrew-Qibin/SPNet

**Cars Can't Fly up in the Sky: Improving Urban-Scene Segmentation via Height-driven Attention Networks**

- 論文：https://arxiv.org/abs/2003.05128
- 代碼：https://github.com/shachoi/HANet

**Learning Dynamic Routing for Semantic Segmentation**

- 論文：https://arxiv.org/abs/2003.10401

- 代碼：https://github.com/yanwei-li/DynamicRouting

# 實例分割

**D2Det: Towards High Quality Object Detection and Instance Segmentation**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/papers/Cao_D2Det_Towards_High_Quality_Object_Detection_and_Instance_Segmentation_CVPR_2020_paper.pdf

- 代碼：https://github.com/JialeCao001/D2Det

**PolarMask: Single Shot Instance Segmentation with Polar Representation**

- 論文：https://arxiv.org/abs/1909.13226
- 代碼：https://github.com/xieenze/PolarMask
- 解讀：https://zhuanlan.zhihu.com/p/84890413

**CenterMask : Real-Time Anchor-Free Instance Segmentation**

- 論文：https://arxiv.org/abs/1911.06667
- 代碼：https://github.com/youngwanLEE/CenterMask

**BlendMask: Top-Down Meets Bottom-Up for Instance Segmentation**

- 論文：https://arxiv.org/abs/2001.00309
- 代碼：https://github.com/aim-uofa/AdelaiDet

**Deep Snake for Real-Time Instance Segmentation**

- 論文：https://arxiv.org/abs/2001.01629
- 代碼：https://github.com/zju3dv/snake

**Mask Encoding for Single Shot Instance Segmentation**

- 論文：https://arxiv.org/abs/2003.11712

- 代碼：https://github.com/aim-uofa/AdelaiDet

# 全景分割

**Video Panoptic Segmentation**

- 論文：https://arxiv.org/abs/2006.11339
- 代碼：https://github.com/mcahny/vps
- 數據集：https://www.dropbox.com/s/ecem4kq0fdkver4/cityscapes-vps-dataset-1.0.zip?dl=0

**Pixel Consensus Voting for Panoptic Segmentation**

- 論文：https://arxiv.org/abs/2004.01849
- 代碼：還未公布

**BANet: Bidirectional Aggregation Network with Occlusion Handling for Panoptic Segmentation**

論文：https://arxiv.org/abs/2003.14031

代碼：https://github.com/Mooonside/BANet

# 視頻目標分割

**A Transductive Approach for Video Object Segmentation**

- 論文：https://arxiv.org/abs/2004.07193

- 代碼：https://github.com/microsoft/transductive-vos.pytorch

**State-Aware Tracker for Real-Time Video Object Segmentation**

- 論文：https://arxiv.org/abs/2003.00482

- 代碼：https://github.com/MegviiDetection/video_analyst

**Learning Fast and Robust Target Models for Video Object Segmentation**

- 論文：https://arxiv.org/abs/2003.00908
- 代碼：https://github.com/andr345/frtm-vos

**Learning Video Object Segmentation from Unlabeled Videos**

- 論文：https://arxiv.org/abs/2003.05020
- 代碼：https://github.com/carrierlxk/MuG

# 超像素分割

**Superpixel Segmentation with Fully Convolutional Networks**

- 論文：https://arxiv.org/abs/2003.12929
- 代碼：https://github.com/fuy34/superpixel_fcn

# 交互式圖像分割

**Interactive Object Segmentation with Inside-Outside Guidance**

- 論文下載鏈接：http://openaccess.thecvf.com/content_CVPR_2020/papers/Zhang_Interactive_Object_Segmentation_With_Inside-Outside_Guidance_CVPR_2020_paper.pdf
- 代碼：https://github.com/shiyinzhang/Inside-Outside-Guidance
- 數據集：https://github.com/shiyinzhang/Pixel-ImageNet

# NAS

**AOWS: Adaptive and optimal network width search with latency constraints**

- 論文：https://arxiv.org/abs/2005.10481
- 代碼：https://github.com/bermanmaxim/AOWS

**Densely Connected Search Space for More Flexible Neural Architecture Search**

- 論文：https://arxiv.org/abs/1906.09607

- 代碼：https://github.com/JaminFong/DenseNAS

**MTL-NAS: Task-Agnostic Neural Architecture Search towards General-Purpose Multi-Task Learning**

- 論文：https://arxiv.org/abs/2003.14058

- 代碼：https://github.com/bhpfelix/MTLNAS

**FBNetV2: Differentiable Neural Architecture Search for Spatial and Channel Dimensions**

- 論文下載鏈接：https://arxiv.org/abs/2004.05565

- 代碼：https://github.com/facebookresearch/mobile-vision

**Neural Architecture Search for Lightweight Non-Local Networks**

- 論文：https://arxiv.org/abs/2004.01961
- 代碼：https://github.com/LiYingwei/AutoNL

**Rethinking Performance Estimation in Neural Architecture Search**

- 論文：https://arxiv.org/abs/2005.09917
- 代碼：https://github.com/zhengxiawu/rethinking_performance_estimation_in_NAS
- 解讀1：https://www.zhihu.com/question/372070853/answer/1035234510
- 解讀2：https://zhuanlan.zhihu.com/p/111167409

**CARS: Continuous Evolution for Efficient Neural Architecture Search**

- 論文：https://arxiv.org/abs/1909.04977
- 代碼（即將開源）：https://github.com/huawei-noah/CARS

# GAN

**SEAN: Image Synthesis with Semantic Region-Adaptive Normalization**

- 論文：https://arxiv.org/abs/1911.12861
- 代碼：https://github.com/ZPdesu/SEAN

**Reusing Discriminators for Encoding: Towards Unsupervised Image-to-Image Translation**

- 論文地址：http://openaccess.thecvf.com/content_CVPR_2020/html/Chen_Reusing_Discriminators_for_Encoding_Towards_Unsupervised_Image-to-Image_Translation_CVPR_2020_paper.html
- 代碼地址：https://github.com/alpc91/NICE-GAN-pytorch

**Distribution-induced Bidirectional Generative Adversarial Network for Graph Representation Learning**

- 論文：https://arxiv.org/abs/1912.01899
- 代碼：https://github.com/SsGood/DBGAN

**PSGAN: Pose and Expression Robust Spatial-Aware GAN for Customizable Makeup Transfer**

- 論文：https://arxiv.org/abs/1909.06956
- 代碼：https://github.com/wtjiang98/PSGAN

**Semantically Mutil-modal Image Synthesis**

- 主頁：http://seanseattle.github.io/SMIS
- 論文：https://arxiv.org/abs/2003.12697
- 代碼：https://github.com/Seanseattle/SMIS

**Unpaired Portrait Drawing Generation via Asymmetric Cycle Mapping**

- 論文：https://yiranran.github.io/files/CVPR2020_Unpaired%20Portrait%20Drawing%20Generation%20via%20Asymmetric%20Cycle%20Mapping.pdf
- 代碼：https://github.com/yiranran/Unpaired-Portrait-Drawing

**Learning to Cartoonize Using White-box Cartoon Representations**

- 論文：https://github.com/SystemErrorWang/White-box-Cartoonization/blob/master/paper/06791.pdf

- 主頁：https://systemerrorwang.github.io/White-box-Cartoonization/
- 代碼：https://github.com/SystemErrorWang/White-box-Cartoonization
- 解讀：https://zhuanlan.zhihu.com/p/117422157
- Demo視頻：https://www.bilibili.com/video/av56708333

**GAN Compression: Efficient Architectures for Interactive Conditional GANs**

- 論文：https://arxiv.org/abs/2003.08936

- 代碼：https://github.com/mit-han-lab/gan-compression

**Watch your Up-Convolution: CNN Based Generative Deep Neural Networks are Failing to Reproduce Spectral Distributions**

- 論文：https://arxiv.org/abs/2003.01826
- 代碼：https://github.com/cc-hpc-itwm/UpConv

# Re-ID

**High-Order Information Matters: Learning Relation and Topology for Occluded Person Re-Identification**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/html/Wang_High-Order_Information_Matters_Learning_Relation_and_Topology_for_Occluded_Person_CVPR_2020_paper.html
- 代碼：https://github.com/wangguanan/HOReID

**COCAS: A Large-Scale Clothes Changing Person Dataset for Re-identification**

- 論文：https://arxiv.org/abs/2005.07862

- 數據集：暫無

**Transferable, Controllable, and Inconspicuous Adversarial Attacks on Person Re-identification With Deep Mis-Ranking**

- 論文：https://arxiv.org/abs/2004.04199

- 代碼：https://github.com/whj363636/Adversarial-attack-on-Person-ReID-With-Deep-Mis-Ranking

**Pose-guided Visible Part Matching for Occluded Person ReID**

- 論文：https://arxiv.org/abs/2004.00230
- 代碼：https://github.com/hh23333/PVPM

**Weakly supervised discriminative feature learning with state information for person identification**

- 論文：https://arxiv.org/abs/2002.11939
- 代碼：https://github.com/KovenYu/state-information

# 3D點雲（分類/分割/配准等）

## 3D點雲卷積

**PointASNL: Robust Point Clouds Processing using Nonlocal Neural Networks with Adaptive Sampling**

- 論文：https://arxiv.org/abs/2003.00492
- 代碼：https://github.com/yanx27/PointASNL

**Global-Local Bidirectional Reasoning for Unsupervised Representation Learning of 3D Point Clouds**

- 論文下載鏈接：https://arxiv.org/abs/2003.12971

- 代碼：https://github.com/raoyongming/PointGLR

**Grid-GCN for Fast and Scalable Point Cloud Learning**

- 論文：https://arxiv.org/abs/1912.02984

- 代碼：https://github.com/Xharlie/Grid-GCN

**FPConv: Learning Local Flattening for Point Convolution**

- 論文：https://arxiv.org/abs/2002.10701
- 代碼：https://github.com/lyqun/FPConv

## 3D點雲分類

**PointAugment: an Auto-Augmentation Framework for Point Cloud Classification**

- 論文：https://arxiv.org/abs/2002.10876
- 代碼（即將開源）： https://github.com/liruihui/PointAugment/

## 3D點雲語義分割

**RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds**

- 論文：https://arxiv.org/abs/1911.11236
- 代碼：https://github.com/QingyongHu/RandLA-Net

- 解讀：https://zhuanlan.zhihu.com/p/105433460

**Weakly Supervised Semantic Point Cloud Segmentation:Towards 10X Fewer Labels**

- 論文：https://arxiv.org/abs/2004.0409

- 代碼：https://github.com/alex-xun-xu/WeakSupPointCloudSeg

**PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation**

- 論文：https://arxiv.org/abs/2003.14032
- 代碼：https://github.com/edwardzhou130/PolarSeg

**Learning to Segment 3D Point Clouds in 2D Image Space**

- 論文：https://arxiv.org/abs/2003.05593

- 代碼：https://github.com/WPI-VISLab/Learning-to-Segment-3D-Point-Clouds-in-2D-Image-Space

## 3D點雲實例分割

PointGroup: Dual-Set Point Grouping for 3D Instance Segmentation

- 論文：https://arxiv.org/abs/2004.01658
- 代碼：https://github.com/Jia-Research-Lab/PointGroup

## 3D點雲配准

**Feature-metric Registration: A Fast Semi-supervised Approach for Robust Point Cloud Registration without Correspondences**

- 論文：https://arxiv.org/abs/2005.01014
- 代碼：https://github.com/XiaoshuiHuang/fmr

**D3Feat: Joint Learning of Dense Detection and Description of 3D Local Features**

- 論文：https://arxiv.org/abs/2003.03164
- 代碼：https://github.com/XuyangBai/D3Feat

**RPM-Net: Robust Point Matching using Learned Features**

- 論文：https://arxiv.org/abs/2003.13479
- 代碼：https://github.com/yewzijian/RPMNet

## 3D點雲補全

**Cascaded Refinement Network for Point Cloud Completion**

- 論文：https://arxiv.org/abs/2004.03327
- 代碼：https://github.com/xiaogangw/cascaded-point-completion

## 3D點雲目標跟蹤

**P2B: Point-to-Box Network for 3D Object Tracking in Point Clouds**

- 論文：https://arxiv.org/abs/2005.13888
- 代碼：https://github.com/HaozheQi/P2B

## 其他

**An Efficient PointLSTM for Point Clouds Based Gesture Recognition**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/html/Min_An_Efficient_PointLSTM_for_Point_Clouds_Based_Gesture_Recognition_CVPR_2020_paper.html
- 代碼：https://github.com/Blueprintf/pointlstm-gesture-recognition-pytorch

# 人臉

## 人臉識別

**CurricularFace: Adaptive Curriculum Learning Loss for Deep Face Recognition**

- 論文：https://arxiv.org/abs/2004.00288

- 代碼：https://github.com/HuangYG123/CurricularFace

**Learning Meta Face Recognition in Unseen Domains**

- 論文：https://arxiv.org/abs/2003.07733
- 代碼：https://github.com/cleardusk/MFR
- 解讀：https://mp.weixin.qq.com/s/YZoEnjpnlvb90qSI3xdJqQ

## 人臉檢測

## 人臉活體檢測

**Searching Central Difference Convolutional Networks for Face Anti-Spoofing**

- 論文：https://arxiv.org/abs/2003.04092

- 代碼：https://github.com/ZitongYu/CDCN

## 人臉表情識別

**Suppressing Uncertainties for Large-Scale Facial Expression Recognition**

- 論文：https://arxiv.org/abs/2002.10392

- 代碼（即將開源）：https://github.com/kaiwang960112/Self-Cure-Network

## 人臉轉正

**Rotate-and-Render: Unsupervised Photorealistic Face Rotation from Single-View Images**

- 論文：https://arxiv.org/abs/2003.08124
- 代碼：https://github.com/Hangz-nju-cuhk/Rotate-and-Render

## 人臉3D重建

**AvatarMe: Realistically Renderable 3D Facial Reconstruction "in-the-wild"**

- 論文：https://arxiv.org/abs/2003.13845
- 數據集：https://github.com/lattas/AvatarMe

**FaceScape: a Large-scale High Quality 3D Face Dataset and Detailed Riggable 3D Face Prediction**

- 論文：https://arxiv.org/abs/2003.13989
- 代碼：https://github.com/zhuhao-nju/facescape

# 人體姿態估計(2D/3D)

## 2D人體姿態估計

**TransMoMo: Invariance-Driven Unsupervised Video Motion Retargeting**

- 主頁：https://yzhq97.github.io/transmomo/

- 論文：https://arxiv.org/abs/2003.14401
- 代碼：https://github.com/yzhq97/transmomo.pytorch

**HigherHRNet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimation**

- 論文：https://arxiv.org/abs/1908.10357
- 代碼：https://github.com/HRNet/HigherHRNet-Human-Pose-Estimation

**The Devil is in the Details: Delving into Unbiased Data Processing for Human Pose Estimation**

- 論文：https://arxiv.org/abs/1911.07524
- 代碼：https://github.com/HuangJunJie2017/UDP-Pose
- 解讀：https://zhuanlan.zhihu.com/p/92525039

**Distribution-Aware Coordinate Representation for Human Pose Estimation**

- 主頁：https://ilovepose.github.io/coco/

- 論文：https://arxiv.org/abs/1910.06278

- 代碼：https://github.com/ilovepose/DarkPose

## 3D人體姿態估計

**Cascaded Deep Monocular 3D Human Pose Estimation With Evolutionary Training Data**

- 論文：https://arxiv.org/abs/2006.07778
- 代碼：https://github.com/Nicholasli1995/EvoSkeleton

**Fusing Wearable IMUs with Multi-View Images for Human Pose Estimation: A Geometric Approach**

- 主頁：https://www.zhe-zhang.com/cvpr2020
- 論文：https://arxiv.org/abs/2003.11163

- 代碼：https://github.com/CHUNYUWANG/imu-human-pose-pytorch

**Bodies at Rest: 3D Human Pose and Shape Estimation from a Pressure Image using Synthetic Data**

- 論文下載鏈接：https://arxiv.org/abs/2004.01166

- 代碼：https://github.com/Healthcare-Robotics/bodies-at-rest
- 數據集：https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/KOA4ML

**Self-Supervised 3D Human Pose Estimation via Part Guided Novel Image Synthesis**

- 主頁：http://val.cds.iisc.ac.in/pgp-human/
- 論文：https://arxiv.org/abs/2004.04400

**Compressed Volumetric Heatmaps for Multi-Person 3D Pose Estimation**

- 論文：https://arxiv.org/abs/2004.00329
- 代碼：https://github.com/fabbrimatteo/LoCO

**VIBE: Video Inference for Human Body Pose and Shape Estimation**

- 論文：https://arxiv.org/abs/1912.05656
- 代碼：https://github.com/mkocabas/VIBE

**Back to the Future: Joint Aware Temporal Deep Learning 3D Human Pose Estimation**

- 論文：https://arxiv.org/abs/2002.11251
- 代碼：https://github.com/vnmr/JointVideoPose3D

**Cross-View Tracking for Multi-Human 3D Pose Estimation at over 100 FPS**

- 論文：https://arxiv.org/abs/2003.03972
- 數據集：暫無

# 人體解析

**Correlating Edge, Pose with Parsing**

- 論文：https://arxiv.org/abs/2005.01431

- 代碼：https://github.com/ziwei-zh/CorrPM

# 場景文本檢測

**STEFANN: Scene Text Editor using Font Adaptive Neural Network**

- 主頁：https://prasunroy.github.io/stefann/

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/html/Roy_STEFANN_Scene_Text_Editor_Using_Font_Adaptive_Neural_Network_CVPR_2020_paper.html
- 代碼：https://github.com/prasunroy/stefann
- 數據集：https://drive.google.com/open?id=1sEDiX_jORh2X-HSzUnjIyZr-G9LJIw1k

**ContourNet: Taking a Further Step Toward Accurate Arbitrary-Shaped Scene Text Detection**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/papers/Wang_ContourNet_Taking_a_Further_Step_Toward_Accurate_Arbitrary-Shaped_Scene_Text_CVPR_2020_paper.pdf
- 代碼：https://github.com/wangyuxin87/ContourNet

**UnrealText: Synthesizing Realistic Scene Text Images from the Unreal World**

- 論文：https://arxiv.org/abs/2003.10608
- 代碼和數據集：https://github.com/Jyouhou/UnrealText/

**ABCNet: Real-time Scene Text Spotting with Adaptive Bezier-Curve Network**

- 論文：https://arxiv.org/abs/2002.10200
- 代碼（即將開源）：https://github.com/Yuliang-Liu/bezier_curve_text_spotting
- 代碼（即將開源）：https://github.com/aim-uofa/adet

**Deep Relational Reasoning Graph Network for Arbitrary Shape Text Detection**

- 論文：https://arxiv.org/abs/2003.07493

- 代碼：https://github.com/GXYM/DRRG

# 場景文本識別

**SEED: Semantics Enhanced Encoder-Decoder Framework for Scene Text Recognition**

- 論文：https://arxiv.org/abs/2005.10977
- 代碼：https://github.com/Pay20Y/SEED

**UnrealText: Synthesizing Realistic Scene Text Images from the Unreal World**

- 論文：https://arxiv.org/abs/2003.10608
- 代碼和數據集：https://github.com/Jyouhou/UnrealText/

**ABCNet: Real-time Scene Text Spotting with Adaptive Bezier-Curve Network**

- 論文：https://arxiv.org/abs/2002.10200
- 代碼（即將開源）：https://github.com/aim-uofa/adet

**Learn to Augment: Joint Data Augmentation and Network Optimization for Text Recognition**

- 論文：https://arxiv.org/abs/2003.06606

- 代碼：https://github.com/Canjie-Luo/Text-Image-Augmentation

# 特征(點)檢測和描述

**SuperGlue: Learning Feature Matching with Graph Neural Networks**

- 論文：https://arxiv.org/abs/1911.11763
- 代碼：https://github.com/magicleap/SuperGluePretrainedNetwork

# 超分辨率

## 圖像超分辨率

**Closed-Loop Matters: Dual Regression Networks for Single Image Super-Resolution**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/html/Guo_Closed-Loop_Matters_Dual_Regression_Networks_for_Single_Image_Super-Resolution_CVPR_2020_paper.html
- 代碼：https://github.com/guoyongcs/DRN

**Learning Texture Transformer Network for Image Super-Resolution**

- 論文：https://arxiv.org/abs/2006.04139

- 代碼：https://github.com/FuzhiYang/TTSR

**Image Super-Resolution with Cross-Scale Non-Local Attention and Exhaustive Self-Exemplars Mining**

- 論文：https://arxiv.org/abs/2006.01424
- 代碼：https://github.com/SHI-Labs/Cross-Scale-Non-Local-Attention

**Structure-Preserving Super Resolution with Gradient Guidance**

- 論文：https://arxiv.org/abs/2003.13081

- 代碼：https://github.com/Maclory/SPSR

**Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Analysis and a New Strategy**

論文：https://arxiv.org/abs/2004.00448

代碼：https://github.com/clovaai/cutblur

## 視頻超分辨率

**TDAN: Temporally-Deformable Alignment Network for Video Super-Resolution**

- 論文：https://arxiv.org/abs/1812.02898
- 代碼：https://github.com/YapengTian/TDAN-VSR-CVPR-2020

**Space-Time-Aware Multi-Resolution Video Enhancement**

- 主頁：https://alterzero.github.io/projects/STAR.html
- 論文：http://arxiv.org/abs/2003.13170
- 代碼：https://github.com/alterzero/STARnet

**Zooming Slow-Mo: Fast and Accurate One-Stage Space-Time Video Super-Resolution**

- 論文：https://arxiv.org/abs/2002.11616
- 代碼：https://github.com/Mukosame/Zooming-Slow-Mo-CVPR-2020

# 模型壓縮/剪枝

**DMCP: Differentiable Markov Channel Pruning for Neural Networks**

- 論文：https://arxiv.org/abs/2005.03354
- 代碼：https://github.com/zx55/dmcp

**Forward and Backward Information Retention for Accurate Binary Neural Networks**

- 論文：https://arxiv.org/abs/1909.10788

- 代碼：https://github.com/htqin/IR-Net

**Towards Efficient Model Compression via Learned Global Ranking**

- 論文：https://arxiv.org/abs/1904.12368
- 代碼：https://github.com/cmu-enyac/LeGR

**HRank: Filter Pruning using High-Rank Feature Map**

- 論文：http://arxiv.org/abs/2002.10179
- 代碼：https://github.com/lmbxmu/HRank

**GAN Compression: Efficient Architectures for Interactive Conditional GANs**

- 論文：https://arxiv.org/abs/2003.08936

- 代碼：https://github.com/mit-han-lab/gan-compression

**Group Sparsity: The Hinge Between Filter Pruning and Decomposition for Network Compression**

- 論文：https://arxiv.org/abs/2003.08935

- 代碼：https://github.com/ofsoundof/group_sparsity

# 視頻理解/行為識別

**Oops! Predicting Unintentional Action in Video**

- 主頁：https://oops.cs.columbia.edu/

- 論文：https://arxiv.org/abs/1911.11206
- 代碼：https://github.com/cvlab-columbia/oops
- 數據集：https://oops.cs.columbia.edu/data

**PREDICT & CLUSTER: Unsupervised Skeleton Based Action Recognition**

- 論文：https://arxiv.org/abs/1911.12409
- 代碼：https://github.com/shlizee/Predict-Cluster

**Intra- and Inter-Action Understanding via Temporal Action Parsing**

- 論文：https://arxiv.org/abs/2005.10229
- 主頁和數據集：https://sdolivia.github.io/TAPOS/

**3DV: 3D Dynamic Voxel for Action Recognition in Depth Video**

- 論文：https://arxiv.org/abs/2005.05501
- 代碼：https://github.com/3huo/3DV-Action

**FineGym: A Hierarchical Video Dataset for Fine-grained Action Understanding**

- 主頁：https://sdolivia.github.io/FineGym/
- 論文：https://arxiv.org/abs/2004.06704

**TEA: Temporal Excitation and Aggregation for Action Recognition**

- 論文：https://arxiv.org/abs/2004.01398

- 代碼：https://github.com/Phoenix1327/tea-action-recognition

**X3D: Expanding Architectures for Efficient Video Recognition**

- 論文：https://arxiv.org/abs/2004.04730

- 代碼：https://github.com/facebookresearch/SlowFast

**Temporal Pyramid Network for Action Recognition**

- 主頁：https://decisionforce.github.io/TPN

- 論文：https://arxiv.org/abs/2004.03548
- 代碼：https://github.com/decisionforce/TPN

## 基於骨架的動作識別

**Disentangling and Unifying Graph Convolutions for Skeleton-Based Action Recognition**

- 論文：https://arxiv.org/abs/2003.14111
- 代碼：https://github.com/kenziyuliu/ms-g3d

# 人群計數

# 深度估計

**BiFuse: Monocular 360◦ Depth Estimation via Bi-Projection Fusion**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/papers/Wang_BiFuse_Monocular_360_Depth_Estimation_via_Bi-Projection_Fusion_CVPR_2020_paper.pdf
- 代碼：https://github.com/Yeh-yu-hsuan/BiFuse

**Focus on defocus: bridging the synthetic to real domain gap for depth estimation**

- 論文：https://arxiv.org/abs/2005.09623
- 代碼：https://github.com/dvl-tum/defocus-net

**Bi3D: Stereo Depth Estimation via Binary Classifications**

- 論文：https://arxiv.org/abs/2005.07274

- 代碼：https://github.com/NVlabs/Bi3D

**AANet: Adaptive Aggregation Network for Efficient Stereo Matching**

- 論文：https://arxiv.org/abs/2004.09548
- 代碼：https://github.com/haofeixu/aanet

**Towards Better Generalization: Joint Depth-Pose Learning without PoseNet**

- 論文：https://github.com/B1ueber2y/TrianFlow

- 代碼：https://github.com/B1ueber2y/TrianFlow

## 單目深度估計

**On the uncertainty of self-supervised monocular depth estimation**

- 論文：https://arxiv.org/abs/2005.06209
- 代碼：https://github.com/mattpoggi/mono-uncertainty

**3D Packing for Self-Supervised Monocular Depth Estimation**

- 論文：https://arxiv.org/abs/1905.02693
- 代碼：https://github.com/TRI-ML/packnet-sfm
- Demo視頻：https://www.bilibili.com/video/av70562892/

**Domain Decluttering: Simplifying Images to Mitigate Synthetic-Real Domain Shift and Improve Depth Estimation**

- 論文：https://arxiv.org/abs/2002.12114
- 代碼：https://github.com/yzhao520/ARC

# 6D目標姿態估計

**PVN3D: A Deep Point-wise 3D Keypoints Voting Network for 6DoF Pose Estimation**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/papers/He_PVN3D_A_Deep_Point-Wise_3D_Keypoints_Voting_Network_for_6DoF_CVPR_2020_paper.pdf
- 代碼：https://github.com/ethnhe/PVN3D

**MoreFusion: Multi-object Reasoning for 6D Pose Estimation from Volumetric Fusion**

- 論文：https://arxiv.org/abs/2004.04336
- 代碼：https://github.com/wkentaro/morefusion

**EPOS: Estimating 6D Pose of Objects with Symmetries**

主頁：http://cmp.felk.cvut.cz/epos

論文：https://arxiv.org/abs/2004.00605

**G2L-Net: Global to Local Network for Real-time 6D Pose Estimation with Embedding Vector Features**

- 論文：https://arxiv.org/abs/2003.11089

- 代碼：https://github.com/DC1991/G2L_Net

# 手勢估計

**HOPE-Net: A Graph-based Model for Hand-Object Pose Estimation**

- 論文：https://arxiv.org/abs/2004.00060

- 主頁：http://vision.sice.indiana.edu/projects/hopenet

**Monocular Real-time Hand Shape and Motion Capture using Multi-modal Data**

- 論文：https://arxiv.org/abs/2003.09572

- 代碼：https://github.com/CalciferZh/minimal-hand

# 顯著性檢測

**JL-DCF: Joint Learning and Densely-Cooperative Fusion Framework for RGB-D Salient Object Detection**

- 論文：https://arxiv.org/abs/2004.08515

- 代碼：https://github.com/kerenfu/JLDCF/

**UC-Net: Uncertainty Inspired RGB-D Saliency Detection via Conditional Variational Autoencoders**

- 主頁：http://dpfan.net/d3netbenchmark/

- 論文：https://arxiv.org/abs/2004.05763
- 代碼：https://github.com/JingZhang617/UCNet

# 去噪

**A Physics-based Noise Formation Model for Extreme Low-light Raw Denoising**

- 論文：https://arxiv.org/abs/2003.12751

- 代碼：https://github.com/Vandermode/NoiseModel

**CycleISP: Real Image Restoration via Improved Data Synthesis**

- 論文：https://arxiv.org/abs/2003.07761

- 代碼：https://github.com/swz30/CycleISP

# 去雨

**Multi-Scale Progressive Fusion Network for Single Image Deraining**

- 論文：https://arxiv.org/abs/2003.10985
- 代碼：https://github.com/kuihua/MSPFN

**Detail-recovery Image Deraining via Context Aggregation Networks**

- 論文：https://openaccess.thecvf.com/content_CVPR_2020/html/Deng_Detail-recovery_Image_Deraining_via_Context_Aggregation_Networks_CVPR_2020_paper.html
- 代碼：https://github.com/Dengsgithub/DRD-Net

# 去模糊

## 視頻去模糊

**Cascaded Deep Video Deblurring Using Temporal Sharpness Prior**

- 主頁：https://csbhr.github.io/projects/cdvd-tsp/index.html
- 論文：https://arxiv.org/abs/2004.02501
- 代碼：https://github.com/csbhr/CDVD-TSP

# 去霧

**Domain Adaptation for Image Dehazing**

- 論文：https://arxiv.org/abs/2005.04668

- 代碼：https://github.com/HUSTSYJ/DA_dahazing

**Multi-Scale Boosted Dehazing Network with Dense Feature Fusion**

- 論文：https://arxiv.org/abs/2004.13388

- 代碼：https://github.com/BookerDeWitt/MSBDN-DFF

# 特征點檢測與描述

**ASLFeat: Learning Local Features of Accurate Shape and Localization**

- 論文：https://arxiv.org/abs/2003.10071

- 代碼：https://github.com/lzx551402/aslfeat

# 視覺問答(VQA)

**VC R-CNN：Visual Commonsense R-CNN**

- 論文：https://arxiv.org/abs/2002.12204
- 代碼：https://github.com/Wangt-CN/VC-R-CNN

# 視頻問答(VideoQA)

**Hierarchical Conditional Relation Networks for Video Question Answering**

- 論文：https://arxiv.org/abs/2002.10698
- 代碼：https://github.com/thaolmk54/hcrn-videoqa

# 視覺語言導航

**Towards Learning a Generic Agent for Vision-and-Language Navigation via Pre-training**

- 論文：https://arxiv.org/abs/2002.10638
- 代碼（即將開源）：https://github.com/weituo12321/PREVALENT

# 視頻壓縮

**Learning for Video Compression with Hierarchical Quality and Recurrent Enhancement**

- 論文：https://arxiv.org/abs/2003.01966
- 代碼：https://github.com/RenYang-home/HLVC

# 視頻插幀

**AdaCoF: Adaptive Collaboration of Flows for Video Frame Interpolation**

- 論文：https://arxiv.org/abs/1907.10244
- 代碼：https://github.com/HyeongminLEE/AdaCoF-pytorch

**FeatureFlow: Robust Video Interpolation via Structure-to-Texture Generation**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/html/Gui_FeatureFlow_Robust_Video_Interpolation_via_Structure-to-Texture_Generation_CVPR_2020_paper.html

- 代碼：https://github.com/CM-BF/FeatureFlow

**Zooming Slow-Mo: Fast and Accurate One-Stage Space-Time Video Super-Resolution**

- 論文：https://arxiv.org/abs/2002.11616
- 代碼：https://github.com/Mukosame/Zooming-Slow-Mo-CVPR-2020

**Space-Time-Aware Multi-Resolution Video Enhancement**

- 主頁：https://alterzero.github.io/projects/STAR.html
- 論文：http://arxiv.org/abs/2003.13170
- 代碼：https://github.com/alterzero/STARnet

**Scene-Adaptive Video Frame Interpolation via Meta-Learning**

- 論文：https://arxiv.org/abs/2004.00779
- 代碼：https://github.com/myungsub/meta-interpolation

**Softmax Splatting for Video Frame Interpolation**

- 主頁：http://sniklaus.com/papers/softsplat
- 論文：https://arxiv.org/abs/2003.05534
- 代碼：https://github.com/sniklaus/softmax-splatting

# 風格遷移

**Diversified Arbitrary Style Transfer via Deep Feature Perturbation**

- 論文：https://arxiv.org/abs/1909.08223
- 代碼：https://github.com/EndyWon/Deep-Feature-Perturbation

**Collaborative Distillation for Ultra-Resolution Universal Style Transfer**

- 論文：https://arxiv.org/abs/2003.08436

- 代碼：https://github.com/mingsun-tse/collaborative-distillation

# 車道線檢測

**Inter-Region Affinity Distillation for Road Marking Segmentation**

- 論文：https://arxiv.org/abs/2004.05304
- 代碼：https://github.com/cardwing/Codes-for-IntRA-KD

# "人-物"交互(HOT)檢測

**PPDM: Parallel Point Detection and Matching for Real-time Human-Object Interaction Detection**

- 論文：https://arxiv.org/abs/1912.12898
- 代碼：https://github.com/YueLiao/PPDM

**Detailed 2D-3D Joint Representation for Human-Object Interaction**

- 論文：https://arxiv.org/abs/2004.08154

- 代碼：https://github.com/DirtyHarryLYL/DJ-RN

**Cascaded Human-Object Interaction Recognition**

- 論文：https://arxiv.org/abs/2003.04262

- 代碼：https://github.com/tfzhou/C-HOI

**VSGNet: Spatial Attention Network for Detecting Human Object Interactions Using Graph Convolutions**

- 論文：https://arxiv.org/abs/2003.05541
- 代碼：https://github.com/ASMIftekhar/VSGNet

# 軌跡預測

**The Garden of Forking Paths: Towards Multi-Future Trajectory Prediction**

- 論文：https://arxiv.org/abs/1912.06445
- 代碼：https://github.com/JunweiLiang/Multiverse
- 數據集：https://next.cs.cmu.edu/multiverse/

**Social-STGCNN: A Social Spatio-Temporal Graph Convolutional Neural Network for Human Trajectory Prediction**

- 論文：https://arxiv.org/abs/2002.11927
- 代碼：https://github.com/abduallahmohamed/Social-STGCNN

# 運動預測

**Collaborative Motion Prediction via Neural Motion Message Passing**

- 論文：https://arxiv.org/abs/2003.06594
- 代碼：https://github.com/PhyllisH/NMMP

**MotionNet: Joint Perception and Motion Prediction for Autonomous Driving Based on Bird's Eye View Maps**

- 論文：https://arxiv.org/abs/2003.06754

- 代碼：https://github.com/pxiangwu/MotionNet

# 光流估計

**Learning by Analogy: Reliable Supervision from Transformations for Unsupervised Optical Flow Estimation**

- 論文：https://arxiv.org/abs/2003.13045
- 代碼：https://github.com/lliuz/ARFlow

# 圖像檢索

**Evade Deep Image Retrieval by Stashing Private Images in the Hash Space**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/html/Xiao_Evade_Deep_Image_Retrieval_by_Stashing_Private_Images_in_the_CVPR_2020_paper.html
- 代碼：https://github.com/sugarruy/hashstash

# 虛擬試衣

**Towards Photo-Realistic Virtual Try-On by Adaptively Generating↔Preserving Image Content**

- 論文：https://arxiv.org/abs/2003.05863
- 代碼：https://github.com/switchablenorms/DeepFashion_Try_On

# HDR

**Single-Image HDR Reconstruction by Learning to Reverse the Camera Pipeline**

- 主頁：https://www.cmlab.csie.ntu.edu.tw/~yulunliu/SingleHDR

- 論文下載鏈接：https://www.cmlab.csie.ntu.edu.tw/~yulunliu/SingleHDR_/00942.pdf

- 代碼：https://github.com/alex04072000/SingleHDR

# 對抗樣本

**Enhancing Cross-Task Black-Box Transferability of Adversarial Examples With Dispersion Reduction**

- 論文：https://openaccess.thecvf.com/content_CVPR_2020/papers/Lu_Enhancing_Cross-Task_Black-Box_Transferability_of_Adversarial_Examples_With_Dispersion_Reduction_CVPR_2020_paper.pdf
- 代碼：https://github.com/erbloo/dr_cvpr20

**Towards Large yet Imperceptible Adversarial Image Perturbations with Perceptual Color Distance**

- 論文：https://arxiv.org/abs/1911.02466
- 代碼：https://github.com/ZhengyuZhao/PerC-Adversarial

# 三維重建

**Unsupervised Learning of Probably Symmetric Deformable 3D Objects from Images in the Wild**

- **CVPR 2020 Best Paper**
- 主頁：https://elliottwu.com/projects/unsup3d/
- 論文：https://arxiv.org/abs/1911.11130
- 代碼：https://github.com/elliottwu/unsup3d

**Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization**

- 主頁：https://shunsukesaito.github.io/PIFuHD/
- 論文：https://arxiv.org/abs/2004.00452
- 代碼：https://github.com/facebookresearch/pifuhd

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/papers/Patel_TailorNet_Predicting_Clothing_in_3D_as_a_Function_of_Human_CVPR_2020_paper.pdf
- 代碼：https://github.com/chaitanya100100/TailorNet
- 數據集：https://github.com/zycliao/TailorNet_dataset

**Implicit Functions in Feature Space for 3D Shape Reconstruction and Completion**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/papers/Chibane_Implicit_Functions_in_Feature_Space_for_3D_Shape_Reconstruction_and_CVPR_2020_paper.pdf
- 代碼：https://github.com/jchibane/if-net

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/papers/Mir_Learning_to_Transfer_Texture_From_Clothing_Images_to_3D_Humans_CVPR_2020_paper.pdf
- 代碼：https://github.com/aymenmir1/pix2surf

# 深度補全

**Uncertainty-Aware CNNs for Depth Completion: Uncertainty from Beginning to End**

論文：https://arxiv.org/abs/2006.03349

代碼：https://github.com/abdo-eldesokey/pncnn

# 語義場景補全

**3D Sketch-aware Semantic Scene Completion via Semi-supervised Structure Prior**

- 論文：https://arxiv.org/abs/2003.14052
- 代碼：https://github.com/charlesCXK/3D-SketchAware-SSC

# 圖像/視頻描述

**Syntax-Aware Action Targeting for Video Captioning**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/papers/Zheng_Syntax-Aware_Action_Targeting_for_Video_Captioning_CVPR_2020_paper.pdf
- 代碼：https://github.com/SydCaption/SAAT

# 線框解析

**Holistically-Attracted Wireframe Parser**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/html/Xue_Holistically-Attracted_Wireframe_Parsing_CVPR_2020_paper.html

- 代碼：https://github.com/cherubicXN/hawp

# 數據集

**OASIS: A Large-Scale Dataset for Single Image 3D in the Wild**

- 論文：https://arxiv.org/abs/2007.13215
- 數據集：https://oasis.cs.princeton.edu/

**STEFANN: Scene Text Editor using Font Adaptive Neural Network**

- 主頁：https://prasunroy.github.io/stefann/

**Interactive Object Segmentation with Inside-Outside Guidance**

**Video Panoptic Segmentation**

- 論文：https://arxiv.org/abs/2006.11339
- 代碼：https://github.com/mcahny/vps
- 數據集：https://www.dropbox.com/s/ecem4kq0fdkver4/cityscapes-vps-dataset-1.0.zip?dl=0

**FSS-1000: A 1000-Class Dataset for Few-Shot Segmentation**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/html/Li_FSS-1000_A_1000-Class_Dataset_for_Few-Shot_Segmentation_CVPR_2020_paper.html

- 代碼：https://github.com/HKUSTCV/FSS-1000

- 數據集：https://github.com/HKUSTCV/FSS-1000

**3D-ZeF: A 3D Zebrafish Tracking Benchmark Dataset**

- 主頁：https://vap.aau.dk/3d-zef/
- 論文：https://arxiv.org/abs/2006.08466
- 代碼：https://bitbucket.org/aauvap/3d-zef/src/master/
- 數據集：https://motchallenge.net/data/3D-ZeF20

**TailorNet: Predicting Clothing in 3D as a Function of Human Pose, Shape and Garment Style**

**Oops! Predicting Unintentional Action in Video**

- 主頁：https://oops.cs.columbia.edu/

- 論文：https://arxiv.org/abs/1911.11206
- 代碼：https://github.com/cvlab-columbia/oops
- 數據集：https://oops.cs.columbia.edu/data

**The Garden of Forking Paths: Towards Multi-Future Trajectory Prediction**

- 論文：https://arxiv.org/abs/1912.06445
- 代碼：https://github.com/JunweiLiang/Multiverse
- 數據集：https://next.cs.cmu.edu/multiverse/

**Open Compound Domain Adaptation**

- 主頁：https://liuziwei7.github.io/projects/CompoundDomain.html
- 數據集：https://drive.google.com/drive/folders/1_uNTF8RdvhS_sqVTnYx17hEOQpefmE2r?usp=sharing
- 論文：https://arxiv.org/abs/1909.03403
- 代碼：https://github.com/zhmiao/OpenCompoundDomainAdaptation-OCDA

**Intra- and Inter-Action Understanding via Temporal Action Parsing**

- 論文：https://arxiv.org/abs/2005.10229
- 主頁和數據集：https://sdolivia.github.io/TAPOS/

**Dynamic Refinement Network for Oriented and Densely Packed Object Detection**

- 論文下載鏈接：https://arxiv.org/abs/2005.09973

- 代碼和數據集：https://github.com/Anymake/DRN_CVPR2020

**COCAS: A Large-Scale Clothes Changing Person Dataset for Re-identification**

- 論文：https://arxiv.org/abs/2005.07862

- 數據集：暫無

**KeypointNet: A Large-scale 3D Keypoint Dataset Aggregated from Numerous Human Annotations**

- 論文：https://arxiv.org/abs/2002.12687

- 數據集：https://github.com/qq456cvb/KeypointNet

**MSeg: A Composite Dataset for Multi-domain Semantic Segmentation**

- 論文：http://vladlen.info/papers/MSeg.pdf
- 代碼：https://github.com/mseg-dataset/mseg-api
- 數據集：https://github.com/mseg-dataset/mseg-semantic

**AvatarMe: Realistically Renderable 3D Facial Reconstruction "in-the-wild"**

- 論文：https://arxiv.org/abs/2003.13845
- 數據集：https://github.com/lattas/AvatarMe

**Learning to Autofocus**

- 論文：https://arxiv.org/abs/2004.12260
- 數據集：暫無

**FaceScape: a Large-scale High Quality 3D Face Dataset and Detailed Riggable 3D Face Prediction**

- 論文：https://arxiv.org/abs/2003.13989
- 代碼：https://github.com/zhuhao-nju/facescape

**Bodies at Rest: 3D Human Pose and Shape Estimation from a Pressure Image using Synthetic Data**

- 論文下載鏈接：https://arxiv.org/abs/2004.01166

- 代碼：https://github.com/Healthcare-Robotics/bodies-at-rest
- 數據集：https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/KOA4ML

**FineGym: A Hierarchical Video Dataset for Fine-grained Action Understanding**

- 主頁：https://sdolivia.github.io/FineGym/
- 論文：https://arxiv.org/abs/2004.06704

**A Local-to-Global Approach to Multi-modal Movie Scene Segmentation**

- 主頁：https://anyirao.com/projects/SceneSeg.html

- 論文下載鏈接：https://arxiv.org/abs/2004.02678

- 代碼：https://github.com/AnyiRao/SceneSeg

**Deep Homography Estimation for Dynamic Scenes**

- 論文：https://arxiv.org/abs/2004.02132

- 數據集：https://github.com/lcmhoang/hmg-dynamics

**Assessing Image Quality Issues for Real-World Problems**

- 主頁：https://vizwiz.org/tasks-and-datasets/image-quality-issues/
- 論文：https://arxiv.org/abs/2003.12511

**UnrealText: Synthesizing Realistic Scene Text Images from the Unreal World**

- 論文：https://arxiv.org/abs/2003.10608
- 代碼和數據集：https://github.com/Jyouhou/UnrealText/

**PANDA: A Gigapixel-level Human-centric Video Dataset**

- 論文：https://arxiv.org/abs/2003.04852

- 數據集：http://www.panda-dataset.com/

**IntrA: 3D Intracranial Aneurysm Dataset for Deep Learning**

- 論文：https://arxiv.org/abs/2003.02920
- 數據集：https://github.com/intra3d2019/IntrA

**Cross-View Tracking for Multi-Human 3D Pose Estimation at over 100 FPS**

- 論文：https://arxiv.org/abs/2003.03972
- 數據集：暫無

# 其他

**CONSAC: Robust Multi-Model Fitting by Conditional Sample Consensus**

- 論文：http://openaccess.thecvf.com/content_CVPR_2020/html/Kluger_CONSAC_Robust_Multi-Model_Fitting_by_Conditional_Sample_Consensus_CVPR_2020_paper.html
- 代碼：https://github.com/fkluger/consac

**Learning to Learn Single Domain Generalization**

- 論文：https://arxiv.org/abs/2003.13216
- 代碼：https://github.com/joffery/M-ADA

**Open Compound Domain Adaptation**

**Differentiable Volumetric Rendering: Learning Implicit 3D Representations without 3D Supervision**

- 論文：http://www.cvlibs.net/publications/Niemeyer2020CVPR.pdf

- 代碼：https://github.com/autonomousvision/differentiable_volumetric_rendering

**QEBA: Query-Efficient Boundary-Based Blackbox Attack**

- 論文：https://arxiv.org/abs/2005.14137
- 代碼：https://github.com/AI-secure/QEBA

**Equalization Loss for Long-Tailed Object Recognition**

- 論文：https://arxiv.org/abs/2003.05176
- 代碼：https://github.com/tztztztztz/eql.detectron2

**Instance-aware Image Colorization**

- 主頁：https://ericsujw.github.io/InstColorization/
- 論文：https://arxiv.org/abs/2005.10825
- 代碼：https://github.com/ericsujw/InstColorization

**Contextual Residual Aggregation for Ultra High-Resolution Image Inpainting**

- 論文：https://arxiv.org/abs/2005.09704

- 代碼：https://github.com/Atlas200dk/sample-imageinpainting-HiFill

**Where am I looking at? Joint Location and Orientation Estimation by Cross-View Matching**

- 論文：https://arxiv.org/abs/2005.03860
- 代碼：https://github.com/shiyujiao/cross_view_localization_DSM

**Epipolar Transformers**

- 論文：https://arxiv.org/abs/2005.04551

- 代碼：https://github.com/yihui-he/epipolar-transformers

**Bringing Old Photos Back to Life**

- 主頁：http://raywzy.com/Old_Photo/
- 論文：https://arxiv.org/abs/2004.09484

**MaskFlownet: Asymmetric Feature Matching with Learnable Occlusion Mask**

- 論文：https://arxiv.org/abs/2003.10955

- 代碼：https://github.com/microsoft/MaskFlownet

**Self-Supervised Viewpoint Learning from Image Collections**

- 論文：https://arxiv.org/abs/2004.01793
- 論文2：https://research.nvidia.com/sites/default/files/pubs/2020-03_Self-Supervised-Viewpoint-Learning/SSV-CVPR2020.pdf
- 代碼：https://github.com/NVlabs/SSV

**Towards Discriminability and Diversity: Batch Nuclear-norm Maximization under Label Insufficient Situations**

- Oral

- 論文：https://arxiv.org/abs/2003.12237
- 代碼：https://github.com/cuishuhao/BNM

**Towards Learning Structure via Consensus for Face Segmentation and Parsing**

- 論文：https://arxiv.org/abs/1911.00957
- 代碼：https://github.com/isi-vista/structure_via_consensus

**Plug-and-Play Algorithms for Large-scale Snapshot Compressive Imaging**

- Oral
- 論文：https://arxiv.org/abs/2003.13654

- 代碼：https://github.com/liuyang12/PnP-SCI

**Lightweight Photometric Stereo for Facial Details Recovery**

- 論文：https://arxiv.org/abs/2003.12307
- 代碼：https://github.com/Juyong/FacePSNet

**Footprints and Free Space from a Single Color Image**

- 論文：https://arxiv.org/abs/2004.06376

- 代碼：https://github.com/nianticlabs/footprints

**Self-Supervised Monocular Scene Flow Estimation**

- 論文：https://arxiv.org/abs/2004.04143
- 代碼：https://github.com/visinf/self-mono-sf

**Quasi-Newton Solver for Robust Non-Rigid Registration**

- 論文：https://arxiv.org/abs/2004.04322
- 代碼：https://github.com/Juyong/Fast_RNRR

**A Local-to-Global Approach to Multi-modal Movie Scene Segmentation**

- 主頁：https://anyirao.com/projects/SceneSeg.html

- 論文下載鏈接：https://arxiv.org/abs/2004.02678

- 代碼：https://github.com/AnyiRao/SceneSeg

**DeepFLASH: An Efficient Network for Learning-based Medical Image Registration**

- 論文：https://arxiv.org/abs/2004.02097

- 代碼：https://github.com/jw4hv/deepflash

**Self-Supervised Scene De-occlusion**

- 主頁：https://xiaohangzhan.github.io/projects/deocclusion/
- 論文：https://arxiv.org/abs/2004.02788
- 代碼：https://github.com/XiaohangZhan/deocclusion

**Polarized Reflection Removal with Perfect Alignment in the Wild**

- 主頁：https://leichenyang.weebly.com/project-polarized.html
- 代碼：https://github.com/ChenyangLEI/CVPR2020-Polarized-Reflection-Removal-with-Perfect-Alignment

**Background Matting: The World is Your Green Screen**

- 論文：https://arxiv.org/abs/2004.00626
- 代碼：http://github.com/senguptaumd/Background-Matting

**What Deep CNNs Benefit from Global Covariance Pooling: An Optimization Perspective**

- 論文：https://arxiv.org/abs/2003.11241

- 代碼：https://github.com/ZhangLi-CS/GCP_Optimization

**Look-into-Object: Self-supervised Structure Modeling for Object Recognition**

- 論文：暫無
- 代碼：https://github.com/JDAI-CV/LIO

**Video Object Grounding using Semantic Roles in Language Description**

- 論文：https://arxiv.org/abs/2003.10606
- 代碼：https://github.com/TheShadow29/vognet-pytorch

**Dynamic Hierarchical Mimicking Towards Consistent Optimization Objectives**

- 論文：https://arxiv.org/abs/2003.10739
- 代碼：https://github.com/d-li14/DHM

**SDFDiff: Differentiable Rendering of Signed Distance Fields for 3D Shape Optimization**

- 論文：http://www.cs.umd.edu/~yuejiang/papers/SDFDiff.pdf
- 代碼：https://github.com/YueJiang-nj/CVPR2020-SDFDiff

**On Translation Invariance in CNNs: Convolutional Layers can Exploit Absolute Spatial Location**

- 論文：https://arxiv.org/abs/2003.07064

- 代碼：https://github.com/oskyhn/CNNs-Without-Borders

**GhostNet: More Features from Cheap Operations**

- 論文：https://arxiv.org/abs/1911.11907

- 代碼：https://github.com/iamhankai/ghostnet

**AdderNet: Do We Really Need Multiplications in Deep Learning?**

- 論文：https://arxiv.org/abs/1912.13200
- 代碼：https://github.com/huawei-noah/AdderNet

**Deep Image Harmonization via Domain Verification**

- 論文：https://arxiv.org/abs/1911.13239
- 代碼：https://github.com/bcmi/Image_Harmonization_Datasets

**Blurry Video Frame Interpolation**

- 論文：https://arxiv.org/abs/2002.12259
- 代碼：https://github.com/laomao0/BIN

**Extremely Dense Point Correspondences using a Learned Feature Descriptor**

- 論文：https://arxiv.org/abs/2003.00619
- 代碼：https://github.com/lppllppl920/DenseDescriptorLearning-Pytorch

**Filter Grafting for Deep Neural Networks**

- 論文：https://arxiv.org/abs/2001.05868
- 代碼：https://github.com/fxmeng/filter-grafting
- 論文解讀：https://www.zhihu.com/question/372070853/answer/1041569335

**Action Segmentation with Joint Self-Supervised Temporal Domain Adaptation**

- 論文：https://arxiv.org/abs/2003.02824
- 代碼：https://github.com/cmhungsteve/SSTDA

**Detecting Attended Visual Targets in Video**

- 論文：https://arxiv.org/abs/2003.02501

- 代碼：https://github.com/ejcgt/attention-target-detection

**Deep Image Spatial Transformation for Person Image Generation**

- 論文：https://arxiv.org/abs/2003.00696
- 代碼：https://github.com/RenYurui/Global-Flow-Local-Attention

**Rethinking Zero-shot Video Classification: End-to-end Training for Realistic Applications**

- 論文：https://arxiv.org/abs/2003.01455
- 代碼：https://github.com/bbrattoli/ZeroShotVideoClassification

https://github.com/charlesCXK/3D-SketchAware-SSC

https://github.com/Anonymous20192020/Anonymous_CVPR5767

https://github.com/avirambh/ScopeFlow

https://github.com/csbhr/CDVD-TSP

https://github.com/ymcidence/TBH

https://github.com/yaoyao-liu/mnemonics

https://github.com/meder411/Tangent-Images

https://github.com/KaihuaTang/Scene-Graph-Benchmark.pytorch

https://github.com/sjmoran/deep_local_parametric_filters

https://github.com/charlesCXK/3D-SketchAware-SSC

https://github.com/bermanmaxim/AOWS

https://github.com/dc3ea9f/look-into-object

# 不確定中沒中

**FADNet: A Fast and Accurate Network for Disparity Estimation**

- 論文：還沒出來
- 代碼：https://github.com/HKBU-HPML/FADNet

https://github.com/rFID-submit/RandomFID：不確定中沒中

https://github.com/JackSyu/AE-MSR：不確定中沒中

https://github.com/fastconvnets/cvpr2020：不確定中沒中

https://github.com/aimagelab/meshed-memory-transformer：不確定中沒中

https://github.com/TWSFar/CRGNet：不確定中沒中

https://github.com/CVPR-2020/CDARTS：不確定中沒中

https://github.com/anucvml/ddn-cvprw2020：不確定中沒中

https://github.com/dl-model-recommend/model-trust：不確定中沒中

https://github.com/apratimbhattacharyya18/CVPR-2020-Corr-Prior：不確定中沒中

https://github.com/onetcvpr/O-Net：不確定中沒中

https://github.com/502463708/Microcalcification_Detection：不確定中沒中

https://github.com/anonymous-for-review/cvpr-2020-deep-smoke-machine：不確定中沒中

https://github.com/anonymous-for-review/cvpr-2020-smoke-recognition-dataset：不確定中沒中

https://github.com/cvpr-nonrigid/dataset：不確定中沒中

https://github.com/theFool32/PPBA：不確定中沒中

https://github.com/Realtime-Action-Recognition/Realtime-Action-Recognition

免責聲明！

本站轉載的文章為個人學習借鑒使用，本站對版權不負任何法律責任。如果侵犯了您的隱私權益，請聯系本站郵箱yoyou2525@163.com刪除。

猜您在找 CVPR-2021-Papers CVPR歷年Best Papers 多篇開源CVPR 2020 語義分割論文 CVPR 2020 三篇有趣的論文解讀 CVPR 2020目標跟蹤多篇開源論文（上）圖像分割 | Context Prior CPNet | CVPR2020 cvpr2020 | 圖像增強與恢復論文盤點模糊視頻幀插值：CVPR2020論文點評 CVPR 2020目標跟蹤多篇開源論文（下）【CVPR2020】MMTM: Multimodal Transfer Module for CNN Fusion