Deep Learning Applications

Papers

DeepPose: Human Pose Estimation via Deep Neural Networks

• intro: CVPR 2014
• arxiv: http://arxiv.org/abs/1312.4659
• slides: http://140.122.184.143/paperlinks/Slides/DeepPose_HumanPose_Estimation_via_Deep_Neural_Networks.pptx
• github: https://github.com/asanakoy/deeppose_tf

Heterogeneous multi-task learning for human pose estimation with deep convolutional neural network

• paper: www.cv-foundation.org/openaccess/content_cvpr_workshops_2014/W15/papers/LI_Heterogeneous_Multi-task_Learning_2014_CVPR_paper.pdf

Flowing ConvNets for Human Pose Estimation in Videos

• arxiv: http://arxiv.org/abs/1506.02897
• homepage: http://www.robots.ox.ac.uk/~vgg/software/cnn_heatmap/
• github: https://github.com/tpfister/caffe-heatmap

Sparseness Meets Deepness: 3D Human Pose Estimation from Monocular Video

• arxiv: http://arxiv.org/abs/1511.09439
• project page: https://fling.seas.upenn.edu/~xiaowz/dynamic/wordpress/monocular-human-pose/
• video: http://weibo.com/p/230444264a8772b7fff71cd23e40b8a88dcaad

Structured Feature Learning for Pose Estimation

• arxiv: http://arxiv.org/abs/1603.09065
• homepage: http://www.ee.cuhk.edu.hk/~xgwang/projectpage_structured_feature_pose.html

CPM

Convolutional Pose Machines

• intro: Convolutional Pose Machines(CPMs)
• arxiv: http://arxiv.org/abs/1602.00134
• github: https://github.com/shihenw/convolutional-pose-machines-release
• github(PyTorch): https://github.com/tensorboy/pytorch_Realtime_Multi-Person_Pose_Estimation
• github: https://github.com/timctho/convolutional-pose-machines-tensorflow

Stacked Hourglass Networks for Human Pose Estimation

• homepage: http://www-personal.umich.edu/~alnewell/pose/
• arxiv: http://arxiv.org/abs/1603.06937
• github: https://github.com/anewell/pose-hg-train
• demo: https://github.com/anewell/pose-hg-demo

Chained Predictions Using Convolutional Neural Networks

• intro: EECV 2016
• keywords: CNN, structured prediction, RNN, human pose estimation
• arxiv: http://arxiv.org/abs/1605.02346

DeeperCut: A Deeper, Stronger, and Faster Multi-Person Pose Estimation Model

• arxiv: http://arxiv.org/abs/1605.03170
• github: https://github.com/eldar/deepcut-cnn

Real-time Human Pose Estimation from Video with Convolutional Neural Networks

• arxiv: http://arxiv.org/abs/1609.07420

Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields

• intro: CVPR 2017 Oral
• keywords: Part Confidence Maps, Part Affinity Fields & Bipartite Matching & Part Association
• arxiv: https://arxiv.org/abs/1611.08050
• video: https://www.youtube.com/watch?v=pW6nZXeWlGM&feature=youtu.be
• slides: http://image-net.org/challenges/talks/2016/Multi-person%20pose%20estimation-CMU.pdf
• github: https://github.com/ZheC/Realtime_Multi-Person_Pose_Estimation

OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields

• intro: Journal version
• arxiv: https://arxiv.org/abs/1812.08008

Towards Accurate Multi-person Pose Estimation in the Wild

• intro: Google
• intro: CVPR 2017
• arxiv: https://arxiv.org/abs/1701.01779

Binarized Convolutional Landmark Localizers for Human Pose Estimation and Face Alignment with Limited Resources

• intro: ICCV 2017 Oral
• project page: https://www.adrianbulat.com/binary-cnn-landmarks
• arxiv: https://www.arxiv.org/abs/1703.00862

Adversarial PoseNet: A Structure-aware Convolutional Network for Human Pose Estimation

• arxiv: https://arxiv.org/abs/1705.00389
• video: http://v.qq.com/x/page/c039862eira.html
• video: http://v.qq.com/x/page/f0398zcvkl5.html
• video: http://v.qq.com/x/page/w0398ei9m1r.html

A simple yet effective baseline for 3d human pose estimation

• intro: ICCV 2017
• arxiv: https://arxiv.org/abs/1705.03098
• github: https://github.com/una-dinosauria/3d-pose-baseline

Human Pose Detection Mining Body Language from Videos

• blog: https://medium.com/@samim/human-pose-detection-51268e95ddc2

OpenPose: A Real-Time Multi-Person Keypoint Detection And Multi-Threading C++ Library

• intro: OpenPose is a library for real-time multi-person keypoint detection and multi-threading written in C++ using OpenCV and Caffe
• github: https://github.com/CMU-Perceptual-Computing-Lab/openpose

Learning Feature Pyramids for Human Pose Estimation

• arxiv: https://arxiv.org/abs/1708.01101
• github: https://github.com/bearpaw/PyraNet

Multi-Context Attention for Human Pose Estimation

• intro: CVPR 2017
• arxiv: https://arxiv.org/abs/1702.07432
• github(Torch): https://github.com/bearpaw/pose-attention

Human Pose Estimation with TensorFlow

https://github.com/eldar/pose-tensorflow

Cascaded Pyramid Network for Multi-Person Pose Estimation

• intro: CVPR 2018. Tsinghua University & HuaZhong Univerisity of Science and Technology & Megvii Inc
• arxiv: https://arxiv.org/abs/1711.07319
• github(official): https://github.com/chenyilun95/tf-cpn
• github: https://github.com/GengDavid/pytorch-cpn

LSTM Pose Machines

• intro: CVPR 2018. SenseTime Research & Sun Yat-sen University
• arxiv: https://arxiv.org/abs/1712.06316
• github(Caffe, officical): https://github.com/lawy623/LSTM_Pose_Machines

DenseReg: Fully Convolutional Dense Shape Regression In-the-Wild

• intro: CVPR 2017
• project page: http://alpguler.com/DenseReg.html
• arxiv: https://arxiv.org/abs/1612.01202
• github: https://github.com/ralpguler/DenseReg

DenseReg: Fully Convolutional Dense Shape Regression In-the-Wild

https://arxiv.org/abs/1803.02188

DensePose: Dense Human Pose Estimation In The Wild

• intro: CVPR 2018. INRIA & Facebook AI Research
• project page: http://densepose.org/
• arxiv: https://arxiv.org/abs/1802.00434
• github(CaffeO2): https://github.com/facebookresearch/DensePose

LCR-Net++: Multi-person 2D and 3D Pose Detection in Natural Images

• intro: journal version of the CVPR 2017 paper
• arxiv: https://arxiv.org/abs/1803.00455

Deep Pose Consensus Networks

https://arxiv.org/abs/1803.08190

3D Human Pose Estimation in the Wild by Adversarial Learning

• intro: CVPR 2018
• arxiv: https://arxiv.org/abs/1803.09722

Multi-Scale Structure-Aware Network for Human Pose Estimation

https://arxiv.org/abs/1803.09894

Co-occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation

• intro: IJCAI 2018 oral. Hikvision Research Institute
• arxiv: https://arxiv.org/abs/1804.06055

Learning to Refine Human Pose Estimation

• intro: CVPRW (2018). Workshop: Visual Understanding of Humans in Crowd Scene and the 2nd Look Into Person Challenge (VUHCS-LIP)
• arxiv: https://arxiv.org/abs/1804.07909

3D Human Pose Estimation with Relational Networks

https://arxiv.org/abs/1805.08961

Jointly Optimize Data Augmentation and Network Training: Adversarial Data Augmentation in Human Pose Estimation

• intro: CVPR 2018
• arxiv: https://arxiv.org/abs/1805.09707

AlphaPose

RMPE: Regional Multi-person Pose Estimation

• intro: ICCV 2017
• project page: https://fang-haoshu.github.io/publications/rmpe/
• arxiv: https://arxiv.org/abs/1612.00137
• paper: http://openaccess.thecvf.com/content_ICCV_2017/papers/Fang_RMPE_Regional_Multi-Person_ICCV_2017_paper.pdf
• github(Caffe, official): https://github.com/MVIG-SJTU/RMPE
• github: https://github.com/Fang-Haoshu/RMPE

Pose Flow: Efficient Online Pose Tracking

https://arxiv.org/abs/1802.00977

AlphaPose: Multi-Person Pose Estimation System

• intro: an accurate multi-person pose estimation system
• project page: http://www.mvig.org/research/alphapose.html

---

Computing CNN Loss and Gradients for Pose Estimation with Riemannian Geometry

https://arxiv.org/abs/1805.01026

Bi-directional Graph Structure Information Model for Multi-Person Pose Estimation

https://arxiv.org/abs/1805.00603

MultiPoseNet: Fast Multi-Person Pose Estimation using Pose Residual Network

• intro: ECCV 2018. Middle East Technical University
• keywords: Pose Residual Network (PRN), person detection, keypoint detection, person segmentation and pose estimation
• arxiv: https://arxiv.org/abs/1807.04067
• github: https://github.com/mkocabas/pose-residual-network

Deep Autoencoder for Combined Human Pose Estimation and body Model Upscaling

https://arxiv.org/abs/1807.01511

Learning Human Poses from Actions

• intro: BMVC 2018
• arxiv: https://arxiv.org/abs/1807.09075

Multi-Scale Supervised Network for Human Pose Estimation

• intro: ICIP 2018
• arxiv: https://arxiv.org/abs/1808.01623

CU-Net: Coupled U-Nets

• intro: BMVC 2018 (Oral)
• arxiv: https://arxiv.org/abs/1808.06521

Multi-Domain Pose Network for Multi-Person Pose Estimation and Tracking

https://arxiv.org/abs/1810.08338

Benchmarking and Error Diagnosis in Multi-Instance Pose Estimation

• intro: ICCV 2017
• project page: http://www.vision.caltech.edu/~mronchi/projects/PoseErrorDiagnosis/
• arxiv: https://arxiv.org/abs/1707.05388
• github: https://github.com/matteorr/coco-analyze

Improving Multi-Person Pose Estimation using Label Correction

https://arxiv.org/abs/1811.03331

Fast Human Pose Estimation

• intro: Fast Pose Distillation (FPD)
• arxiv: https://arxiv.org/abs/1811.05419

PoseFix: Model-agnostic General Human Pose Refinement Network

• intro: CVPR 2019
• arxiv: https://arxiv.org/abs/1812.03595
• github(TensorFlow): https://github.com/mks0601/PoseFix_RELEASE

Rethinking on Multi-Stage Networks for Human Pose Estimation

• intro: Megvii Inc. (Face++) & Shanghai Jiao Tong University & Beihang University & Beijing University of Posts and Telecommunications
• arxiv: https://arxiv.org/abs/1901.00148
• github: https://github.com/fenglinglwb/MSPN

Deep High-Resolution Representation Learning for Human Pose Estimation

• intro: CVPR 2019
• intro: University of Science and Technology of China & Microsoft Research Asia
• keywords: HRNet
• arxiv: https://arxiv.org/abs/1902.09212
• project page: https://jingdongwang2017.github.io/Projects/HRNet/PoseEstimation.html
• github(official): https://github.com/leoxiaobin/deep-high-resolution-net.pytorch

A Context-and-Spatial Aware Network for Multi-Person Pose Estimation

https://arxiv.org/abs/1905.05355

FastPose: Towards Real-time Pose Estimation and Tracking via Scale-normalized Multi-task Networks

• intro: Chinese Academy of Sciences & BUPT & Horizon Robotics
• arxiv: https://arxiv.org/abs/1908.06290

Single-Stage Multi-Person Pose Machines

• intro: ICCV 2019
• intro: Yitu Technology
• arxiv: https://arxiv.org/abs/1908.09220

Single-Network Whole-Body Pose Estimation

• intro: ICCV 2019
• project page: https://github.com/CMU-Perceptual-Computing-Lab/openpose_train
• arxiv: https://arxiv.org/abs/1909.13423

ZoomNAS: Searching for Whole-body Human Pose Estimation in the Wild

• intro: TPAMI
• arxiv: https://arxiv.org/abs/2208.11547

NADS-Net: A Nimble Architecture for Driver and Seat Belt Detection via Convolutional Neural Networks

https://arxiv.org/abs/1910.03695

Distribution-Aware Coordinate Representation for Human Pose Estimation

• intro: CVPR 2020
• keywords: Distribution-Aware coordinate Representation of Keypoint (DARK)
• intro: Results on the COCO keypoint detection challenge: 78.9% AP on the test-dev set (Top-1 in the leaderbord by 12 Oct 2019) and 76.4% AP on the test-challenge set.
• project page: https://ilovepose.github.io/coco/
• arxiv: https://arxiv.org/abs/1910.06278
• github: https://github.com/ilovepose/DarkPose

TRB: A Novel Triplet Representation for Understanding 2D Human Body

• intro: ICCV 2019
• arxiv: https://arxiv.org/abs/1910.11535

Chirality Nets for Human Pose Regression

• intro: NeurIPS 2019
• arxiv: https://arxiv.org/abs/1911.00029

Conservative Wasserstein Training for Pose Estimation

• intro: ICCV 2019
• arxiv: https://arxiv.org/abs/1911.00962

DirectPose: Direct End-to-End Multi-Person Pose Estimation

• intro: The University of Adelaide
• keywords: Keypoint Alignment (KPAlign)
• arxiv: https://arxiv.org/abs/1911.07451

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

• intro: CVPR 2020
• intro: XForwardAI Technology Co.,Ltd & Tsinghua University
• arxiv: https://arxiv.org/abs/1911.07524
• github: https://github.com/HuangJunJie2017/UDP-Pose

Simple Pose: Rethinking and Improving a Bottom-up Approach for Multi-Person Pose Estimation

• intro: AAAI 2020
• arxiv: https://arxiv.org/abs/1911.10529
• github: https://github.com/hellojialee/Improved-Body-Parts

HintPose

• intro: Joint COCO and Mapillary Workshop at ICCV 2019: Keypoint Detection Challenge Track
• arxiv: https://arxiv.org/abs/2003.02170

How to Train Your Robust Human Pose Estimator: Pay Attention to the Constraint Cue

• intro: XForwardAI Technology Co.,Ltd & Tsinghua University
• arxiv: https://arxiv.org/abs/2008.07139

CoKe: Localized Contrastive Learning for Robust Keypoint Detection

• intro: Johns Hopkins University
• arxiv: https://arxiv.org/abs/2009.14115

View-Invariant, Occlusion-Robust Probabilistic Embedding for Human Pose

• intro: Google Research & California Institute of Technology & Rutgers University
• arxiv: https://arxiv.org/abs/2010.13321
• gtihub: https://github.com/google-research/google-research/tree/master/poem

An Empirical Study of the Collapsing Problem in Semi-Supervised 2D Human Pose Estimation

• intro: Peking University & Microsoft Research Asia
• arxiv: https://arxiv.org/abs/2011.12498

EfficientPose: Efficient Human Pose Estimation with Neural Architecture Search

https://arxiv.org/abs/2012.07086

TransPose: Towards Explainable Human Pose Estimation by Transformer

• intro: Southeast University
• arxiv: https://arxiv.org/abs/2012.14214
• github: https://github.com/yangsenius/TransPose

Rethinking the Heatmap Regression for Bottom-up Human Pose Estimation

https://arxiv.org/abs/2012.15175

Multi-Hypothesis Pose Networks: Rethinking Top-Down Pose Estimation

https://arxiv.org/abs/2101.11223

OmniPose: A Multi-Scale Framework for Multi-Person Pose Estimation

https://arxiv.org/abs/2103.10180

End-to-End Trainable Multi-Instance Pose Estimation with Transformers

• intro: Swiss Federal Institute of Technology (EPFL)
• arxiv: https://arxiv.org/abs/2103.12115

TFPose: Direct Human Pose Estimation with Transformers

• intro: The University of Adelaide & Alibaba Group
• arxiv: https://arxiv.org/abs/2103.15320

TokenPose: Learning Keypoint Tokens for Human Pose Estimation

• intro: MEGVII Technology & Tsinghua University & Southeast University & Peng Cheng Laboratory
• arxiv: https://arxiv.org/abs/2104.03516

Pose Recognition with Cascade Transformers

• intro: CVPR 2021
• arxiv: https://arxiv.org/abs/2104.06976
• github: https://github.com/mlpc-ucsd/PRTR

Is 2D Heatmap Representation Even Necessary for Human Pose Estimation?

• intro: Tsinghua University & MEGVII Technology & Southeast University & Peng Cheng Laboratory
• arxiv: https://arxiv.org/abs/2107.03332
• github: https://github.com/leeyegy/SimDR

InsPose: Instance-Aware Networks for Single-Stage Multi-Person Pose Estimation

• intro: ACM MM 2021
• arxiv: https://arxiv.org/abs/2107.08982

Adaptive Dilated Convolution For Human Pose Estimation

• intro: Megvii & UCAS & CRIPAC & NLPR & CASIA
• arxiv: https://arxiv.org/abs/2107.10477

PoseDet: Fast Multi-Person Pose Estimation Using Pose Embedding

• intro: Tsinghua University & Northwestern University
• arxiv: https://arxiv.org/abs/2107.10466

Online Knowledge Distillation for Efficient Pose Estimation

• intro: ICCV 2021
• arxiv: https://arxiv.org/abs/2108.02092

SmoothNet: A Plug-and-Play Network for Refining Human Poses in Videos

• intro: The Chinese University of Hong Kong & Sensetime Group Ltd. &Shanghai Jiao Tong University & Nanyang Technological University
• project page: https://ailingzeng.site/smoothnet
• arxiv: https://arxiv.org/abs/2112.13715

AdaptivePose: Human Parts as Adaptive Points

• intro: AAAI 2022
• intro: Beijing University of Posts and Telecommunications & ByteDance Inc. & Tsinghua University & Horizon Robotics
• arxiv: https://arxiv.org/abs/2112.13635

Learning Quality-aware Representation for Multi-person Pose Regression

• intro: AAAI 2022
• intro: Beijing University of Posts and Telecommunications & ByteDance Inc. & Tsinghua University & Horizon Robotics
• arxiv: https://arxiv.org/abs/2201.01087

Recognition of Freely Selected Keypoints on Human Limbs

• intro: CVPR 2022 Workshops
• arxiv: https://arxiv.org/abs/2204.06326

YOLO-Pose: Enhancing YOLO for Multi Person Pose Estimation Using Object Keypoint Similarity Loss

• intro: Texas Instruments Inc
• arxiv: https://arxiv.org/abs/2204.06806
• github: https://github.com/TexasInstruments/edgeai-yolov5
• github: https://github.com/TexasInstruments/edgeai-yolox

Lite Pose: Efficient Architecture Design for 2D Human Pose Estimation

• intro: Tsinghua University & CMU & MIT
• arxiv: https://arxiv.org/abs/2205.01271
• github: https://github.com/mit-han-lab/litepose

PPT: token-Pruned Pose Transformer for monocular and multi-view human pose estimation

• intro: University of California & Tencent Inc & Meta AI
• arxiv: https://arxiv.org/abs/2209.08194
• github: https://github.com/HowieMa/PPT

Regression-based Method

Integral Human Pose Regression

• intro: ECCV 2018
• arxiv: https://arxiv.org/abs/1711.08229
• slides: https://jimmysuen.github.io/slides/xiaosun_integral_human_pose_regression.pptx
• github: https://github.com/JimmySuen/integral-human-pose

Human Pose Regression with Residual Log-likelihood Estimation

• intro: ICCV 2021 Oral
• intro: Shanghai Jiao Tong University & The Chinese University of Hong Kong & SenseTime Research
• arxiv: https://arxiv.org/abs/2107.11291
• github: https://github.com/Jeff-sjtu/res-loglikelihood-regression

Poseur: Direct Human Pose Regression with Transformers

• intro: The University of Adelaide & Alibaba Damo Academy & Zhejiang University
• arxiv: https://arxiv.org/abs/2201.07412

Location-free Human Pose Estimation

• intro: Beijing Jiaotong University & Tencent Youtu Lab
• arxiv: https://arxiv.org/abs/2205.12619

Top-Down

Point-Set Anchors for Object Detection, Instance Segmentation and Pose Estimation

• intro: ECCV 2020
• intro: MSRA & Peking University
• arxiv: https://arxiv.org/abs/2007.02846
• github: https://github.com/FangyunWei/PointSetAnchor

Bottom-Up

PifPaf: Composite Fields for Human Pose Estimation

• intro: CVPR 2019
• intro: EPFL VITA lab
• keywords: Part Intensity Field (PIF), Part Association Field (PAF)
• arxiv: https://arxiv.org/abs/1903.06593

OpenPifPaf: Composite Fields for Semantic Keypoint Detection and Spatio-Temporal Association

• project: https://openpifpaf.github.io/intro.html
• intro: https://arxiv.org/abs/2103.02440
• github: https://github.com/openpifpaf/openpifpaf

Bottom-Up Human Pose Estimation Via Disentangled Keypoint Regression

• intro: CVPR 2021
• keywords: DEKR
• intro: University of Science and Technology of China & University of Chinese Academy of Sciences & Microsof
• arxiv: https://arxiv.org/abs/2104.02300
• github: https://github.com/HRNet/DEKR

DeepSportLab: a Unified Framework for Ball Detection, Player Instance Segmentation and Pose Estimation in Team Sports Scenes

• intro: BMVC 2021
• arxiv: https://arxiv.org/abs/2112.00627
• github: https://github.com/ispgroupucl/DeepSportLab

Learning Local-Global Contextual Adaptation for Fully End-to-End Bottom-Up Human Pose Estimation

• intro: Wuhan University & North Carolina State University
• arxiv: https://arxiv.org/abs/2109.03622

Keypoint Communities

• intro: ICCV 2021
• arxiv: https://arxiv.org/abs/2110.00988

The Center of Attention: Center-Keypoint Grouping via Attention for Multi-Person Pose Estimation

• intro: ICCV 2021
• intro: Technical University of Munich
• arxiv: https://arxiv.org/abs/2110.05132

Self-Supervision and Spatial-Sequential Attention Based Loss for Multi-Person Pose Estimation

https://arxiv.org/abs/2110.10734

Learning Local-Global Contextual Adaptation for Multi-Person Pose Estimation

• intro: CVPR 2022
• intro: Wuhan University, NC State University
• arxiv: https://arxiv.org/abs/2109.03622

I^2R-Net: Intra- and Inter-Human Relation Network for Multi-Person Pose Estimation

• intro: Xiamen University & Microsoft Research Asia
• arxiv: https://arxiv.org/abs/2206.10892

End-to-End Multi-Person Pose Estimation with Transformers

• intro: CVPR 2022 Oral
• paper: https://openaccess.thecvf.com/content/CVPR2022/papers/Shi_End-to-End_Multi-Person_Pose_Estimation_With_Transformers_CVPR_2022_paper.pdf
• github: https://github.com/hikvision-research/opera/tree/main/configs/petr

Hand Pose

Model-based Deep Hand Pose Estimation

• paper: http://xingyizhou.xyz/zhou2016model.pdf
• github: https://github.com/tenstep/DeepModel

Region Ensemble Network: Improving Convolutional Network for Hand Pose Estimation

• arxiv: https://arxiv.org/abs/1702.02447

Crossing Nets: Combining GANs and VAEs with a Shared Latent Space for Hand Pose Estimation

• intro: CVPR 2017
• arxiv: https://arxiv.org/abs/1702.03431

Learning to Disambiguate Strongly Interacting Hands via Probabilistic Per-pixel Part Segmentation

• intro: 3DV 2021 Oral
• arxiv: https://arxiv.org/abs/2107.00434
• github: https://github.com/zc-alexfan/digit-interacting

3D Pose

Can 3D Pose be Learned from 2D Projections Alone?

• intro: ECCV 2018 workshop
• arxiv: https://arxiv.org/abs/1808.07182

Fast and Robust Multi-Person 3D Pose Estimation from Multiple Views

• project page: https://zju-3dv.github.io/mvpose/
• arxiv: https://arxiv.org/abs/1901.04111
• github: https://github.com/zju-3dv/mvpose

3D Human Pose Machines with Self-supervised Learning

• intro: T-PAMI 2019
• project page: http://www.sysu-hcp.net/3d_pose_ssl/
• arxiv: https://arxiv.org/abs/1901.03798
• github: https://github.com/chanyn/3Dpose_ssl

Feature Boosting Network For 3D Pose Estimation

• intro: Nanyang Technological University & Chalmers University of Technology & Peking University & Alibaba Group
• arxiv: https://arxiv.org/abs/1901.04877

View Invariant 3D Human Pose Estimation

• intro: MSRA & USTC
• arxiv: https://arxiv.org/abs/1901.10841

3D Human Pose Estimation from Deep Multi-View 2D Pose

https://arxiv.org/abs/1902.02841

RepNet: Weakly Supervised Training of an Adversarial Reprojection Network for 3D Human Pose Estimation

• intro: CVPR 2019
• arxiv: https://arxiv.org/abs/1902.09868

DenseBody: Directly Regressing Dense 3D Human Pose and Shape From a Single Color Image

• intro: Cloudwalk & Shanghai Jiao Tong University
• arxiv: https://arxiv.org/abs/1903.10153

Camera Distance-aware Top-down Approach for 3D Multi-person Pose Estimation from a Single RGB Image

• intro: ICCV 2019
• arxiv: https://arxiv.org/abs/1907.11346
• github: https://github.com/mks0601/3DMPPE_ROOTNET_RELEASE

Lightweight 3D Human Pose Estimation Network Training Using Teacher-Student Learning

https://arxiv.org/abs/2001.05097

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

• intro: CVPR 2020
• intro: Tsinghua University & AiFi Inc.
• arxiv: https://arxiv.org/abs/2003.03972

Skeletor: Skeletal Transformers for Robust Body-Pose Estimation

• intro: University of Surrey
• arxiv: https://arxiv.org/abs/2104.11712

UltraPose: Synthesizing Dense Pose with 1 Billion Points by Human-body Decoupling 3D Model

• intro: ICCV 2021
• intro: Beijing Momo Technology Co., Ltd. & Sun Yat-sen University
• arxiv: https://arxiv.org/abs/2110.15267
• github: https://github.com/MomoAILab/ultrapose

Distribution-Aware Single-Stage Models for Multi-Person 3D Pose Estimation

• intro: CVPR 2022
• intro: Beihang University & Meitu Inc.
• arxiv: https://arxiv.org/abs/2203.07697

DeciWatch: A Simple Baseline for 10x Efficient 2D and 3D Pose Estimation

• intro: The Chinese University of Hong Kong & Sensetime Group Ltd. & Shanghai Artificial Intelligence Laboratory
• project page: https://ailingzeng.site/deciwatch
• arxiv: https://arxiv.org/abs/2203.08713

3D Car keypoints Detection

Occlusion-Net: 2D/3D Occluded Keypoint Localization Using Graph Networks

• intro: CVPR 2019
• paper: http://openaccess.thecvf.com/content_CVPR_2019/papers/Reddy_Occlusion-Net_2D3D_Occluded_Keypoint_Localization_Using_Graph_Networks_CVPR_2019_paper.pdf
• github: https://github.com/dineshreddy91/Occlusion_Net

RTM3D: Real-time Monocular 3D Detection from Object Keypoints for Autonomous Driving

• intro: Chinese Academy of Sciences & University of Chinese Academy of Sciences
• arxiv: https://arxiv.org/abs/2001.03343
• github: https://github.com/Banconxuan/RTM3D

SMOKE: Single-Stage Monocular 3D Object Detection via Keypoint Estimation

• intro: 1ZongMu Tech & TU/e
• arxiv: https://arxiv.org/abs/2002.10111

Pose Estimation and Action Recognition

2D/3D Pose Estimation and Action Recognition using Multitask Deep Learning

• intro: CVPR 2018
• arxiv: https://arxiv.org/abs/1802.09232

Multi-task Deep Learning for Real-Time 3D Human Pose Estimation and Action Recognition

• arxiv: https://arxiv.org/abs/1912.08077
• github: https://github.com/dluvizon/deephar

Video Pose Distillation for Few-Shot, Fine-Grained Sports Action Recognition

• intro: ICCV 2021 poster
• intro: Stanford University & Adobe Research
• arxiv: https://arxiv.org/abs/2109.01305

Skeleton Sequence and RGB Frame Based Multi-Modality Feature Fusion Network for Action Recognition

• intro: ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM)
• arxiv: https://arxiv.org/abs/2202.11374

Revisiting Skeleton-based Action Recognition

• intro: CVPR 2022 Oral
• keywords: PoseConv3D
• arxiv: https://arxiv.org/abs/2104.13586
• github: https://github.com/kennymckormick/pyskl
• github: https://github.com/open-mmlab/mmaction2/blob/master/configs/skeleton/posec3d/README.md

Pose Tracking

Detect-and-Track: Efficient Pose Estimation in Videos

• intro: CVPR 2018. CMU & Facebook & Dartmouth
• intro: Ranked first in ICCV 2017 PoseTrack challenge (keypoint tracking in videos)
• project page: https://rohitgirdhar.github.io/DetectAndTrack/
• arxiv: https://arxiv.org/abs/1712.09184
• paper: http://openaccess.thecvf.com/content_cvpr_2018/papers/Girdhar_Detect-and-Track_Efficient_Pose_CVPR_2018_paper.pdf
• github: https://github.com/facebookresearch/DetectAndTrack/

Simple Baselines for Human Pose Estimation and Tracking

• intro: ECCV 2018
• intro: MSRA
• keywords: optical flow based pose propagation and similarity measurement
• arxiv: https://arxiv.org/abs/1804.06208
• github(official): https://github.com/Microsoft/human-pose-estimation.pytorch

A Top-down Approach to Articulated Human Pose Estimation and Tracking

• intro: JD.com Silicon Valley Research Center
• intro: ECCVW 2018. Workshop: 2nd PoseTrack Challenge
• arxiv: https://arxiv.org/abs/1901.07680

15 Keypoints Is All You Need

• intro: Brown University & NEC Labs America
• arxiv: https://arxiv.org/abs/1912.02323

Learning Dynamics via Graph Neural Networks for Human Pose Estimation and Tracking

• intro: CVPR 2021
• intro: Stevens Institute of Technology & Wormpex AI Research & National University of Singapore
• arxiv: https://arxiv.org/abs/2106.03772

Object Pose Estimation

Real-Time Object Pose Estimation with Pose Interpreter Networks

• intro: 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2018)
• arxiv: https://arxiv.org/abs/1808.01099
• github: https://github.com/jimmyyhwu/pose-interpreter-networks

Projects

MobilePose: Single Person Pose Estimation for Mobile Device

• intro: a Tiny PyTorch implementation of single person 2D pose estimation framework
• github: https://github.com/YuliangXiu/MobilePose-pytorch github: https://github.com/MVIG-SJTU/AlphaPose

PyTorch-Pose: A PyTorch toolkit for 2D Human Pose Estimation

• intro: a PyTorch implementation of the general pipeline for 2D single human pose estimation.
• github: https://github.com/bearpaw/pytorch-pose

Hourglass, DHN and CPN model in TensorFlow for 2018-FashionAI Key Points Detection of Apparel at TianChi

• intro: Full pipeline for TianChi FashionAI clothes keypoints detection compitetion in TensorFlow
• github: https://github.com/HiKapok/tf.fashionAI

FashionAI: KeyPoint Detection Challenge in Keras

• intro: Code for TianChi 2018 FashionAI Cloth KeyPoint Detection Challenge
• github: https://github.com/yuanyuanli85/FashionAI_KeyPoint_Detection_Challenge_Keras

Challenge

POSETRACK CHALLENGE: ARTICULATED PEOPLE TRACKING IN THE WILD

https://posetrack.net/workshops/eccv2018/#challenges