Table of Contents
Shortcuts

Overview

  • Number of papers: 83

    • ALGORITHM: 32

    • BACKBONE: 20

    • DATASET: 28

    • OTHERS: 3

For supported pose algorithms, see modelzoo overview.

Animal 2D Keypoint

  • Number of papers: 9

    • [ALGORITHM] Ap-10k: A Benchmark for Animal Pose Estimation in the Wild (Ap10k Dataset ⇨)

    • [ALGORITHM] Deep High-Resolution Representation Learning for Human Pose Estimation (Ak Dataset ⇨, Animalpose Dataset ⇨, Ap10k Dataset ⇨)

    • [ALGORITHM] Rtmdet: An Empirical Study of Designing Real-Time Object Detectors (Ap10k Dataset ⇨)

    • [ALGORITHM] Simple Baselines for Human Pose Estimation and Tracking (Locust Dataset ⇨, Animalpose Dataset ⇨, Ap10k Dataset ⇨, Zebra Dataset ⇨)

    • [ALGORITHM] The Devil Is in the Details: Delving Into Unbiased Data Processing for Human Pose Estimation (Ap10k Dataset ⇨)

    • [DATASET] Animal Kingdom: A Large and Diverse Dataset for Animal Behavior Understanding (Ak Dataset ⇨)

    • [DATASET] Ap-10k: A Benchmark for Animal Pose Estimation in the Wild (Ap10k Dataset ⇨)

    • [DATASET] Cross-Domain Adaptation for Animal Pose Estimation (Animalpose Dataset ⇨)

    • [DATASET] Deepposekit, a Software Toolkit for Fast and Robust Animal Pose Estimation Using Deep Learning (Locust Dataset ⇨, Zebra Dataset ⇨)

Body 2D Keypoint

  • Number of papers: 54

    • [ALGORITHM] Associative Embedding: End-to-End Learning for Joint Detection and Grouping (Coco Dataset ⇨)

    • [ALGORITHM] Bottom-Up Human Pose Estimation via Disentangled Keypoint Regression (Coco Dataset ⇨, Crowdpose Dataset ⇨)

    • [ALGORITHM] Contextual Instance Decoupling for Robust Multi-Person Pose Estimation (Coco Dataset ⇨)

    • [ALGORITHM] Convolutional Pose Machines (JHMDB Dataset ⇨, Mpii Dataset ⇨, Coco Dataset ⇨)

    • [ALGORITHM] Deep High-Resolution Representation Learning for Human Pose Estimation (Posetrack18 Dataset ⇨, Humanart Dataset ⇨, Aic Dataset ⇨, Exlpose Dataset ⇨, Mpii Dataset ⇨, Coco Dataset ⇨, Crowdpose Dataset ⇨)

    • [ALGORITHM] Deeppose: Human Pose Estimation via Deep Neural Networks (Mpii Dataset ⇨, Coco Dataset ⇨)

    • [ALGORITHM] Distribution-Aware Coordinate Representation for Human Pose Estimation (Mpii Dataset ⇨, Coco Dataset ⇨)

    • [ALGORITHM] Explicit Box Detection Unifies End-to-End Multi-Person Pose Estimation (Coco Dataset ⇨)

    • [ALGORITHM] Hrformer: High-Resolution Vision Transformer for Dense Predict (Coco Dataset ⇨)

    • [ALGORITHM] Human Pose Regression With Residual Log-Likelihood Estimation (Mpii Dataset ⇨, Coco Dataset ⇨)

    • [ALGORITHM] Improving Convolutional Networks With Self-Calibrated Convolutions (Mpii Dataset ⇨, Coco Dataset ⇨)

    • [ALGORITHM] Integral Human Pose Regression (Coco Dataset ⇨)

    • [ALGORITHM] Learning Delicate Local Representations for Multi-Person Pose Estimation (Coco Dataset ⇨)

    • [ALGORITHM] Lite-Hrnet: A Lightweight High-Resolution Network (Mpii Dataset ⇨, Coco Dataset ⇨)

    • [ALGORITHM] Numerical Coordinate Regression With Convolutional Neural Networks (Coco Dataset ⇨)

    • [ALGORITHM] Removing the Bias of Integral Pose Regression (Coco Dataset ⇨)

    • [ALGORITHM] Rethinking on Multi-Stage Networks for Human Pose Estimation (Coco Dataset ⇨)

    • [ALGORITHM] Rtmdet: An Empirical Study of Designing Real-Time Object Detectors (Body8 Dataset ⇨, Humanart Dataset ⇨, Mpii Dataset ⇨, Coco Dataset ⇨, Crowdpose Dataset ⇨)

    • [ALGORITHM] Rtmpose: Real-Time Multi-Person Pose Estimation Based on Mmpose (Body8 Dataset ⇨, Humanart Dataset ⇨, Mpii Dataset ⇨, Coco Dataset ⇨, Crowdpose Dataset ⇨)

    • [ALGORITHM] Simcc: A Simple Coordinate Classification Perspective for Human Pose Estimation (Coco Dataset ⇨)

    • [ALGORITHM] Simple Baselines for Human Pose Estimation and Tracking (Posetrack18 Dataset ⇨, JHMDB Dataset ⇨, Aic Dataset ⇨, Mpii Dataset ⇨, Coco Dataset ⇨, Crowdpose Dataset ⇨)

    • [ALGORITHM] Stacked Hourglass Networks for Human Pose Estimation (Mpii Dataset ⇨, Coco Dataset ⇨)

    • [ALGORITHM] The Devil Is in the Details: Delving Into Unbiased Data Processing for Human Pose Estimation (Mpii Dataset ⇨, Coco Dataset ⇨, Crowdpose Dataset ⇨)

    • [ALGORITHM] Vipnas: Efficient Video Pose Estimation via Neural Architecture Search (Coco Dataset ⇨)

    • [ALGORITHM] Yolo-Pose: Enhancing Yolo for Multi Person Pose Estimation Using Object Keypoint Similarity Loss (Coco Dataset ⇨)

    • [ALGORITHM] {Rtmo (Coco Dataset ⇨, Crowdpose Dataset ⇨, Body7 Dataset ⇨)

    • [BACKBONE] Aggregated Residual Transformations for Deep Neural Networks (Mpii Dataset ⇨, Coco Dataset ⇨)

    • [BACKBONE] Bag of Tricks for Image Classification With Convolutional Neural Networks (Mpii Dataset ⇨, Coco Dataset ⇨)

    • [BACKBONE] Deep Residual Learning for Image Recognition (Posetrack18 Dataset ⇨, JHMDB Dataset ⇨, Aic Dataset ⇨, Mpii Dataset ⇨, Coco Dataset ⇨, Crowdpose Dataset ⇨)

    • [BACKBONE] Imagenet Classification With Deep Convolutional Neural Networks (Coco Dataset ⇨)

    • [BACKBONE] Mobilenetv2: Inverted Residuals and Linear Bottlenecks (Mpii Dataset ⇨, Coco Dataset ⇨)

    • [BACKBONE] Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction Without Convolutions (Coco Dataset ⇨)

    • [BACKBONE] Resnest: Split-Attention Networks (Coco Dataset ⇨)

    • [BACKBONE] Rtmdet: An Empirical Study of Designing Real-Time Object Detectors (Body8 Dataset ⇨, Humanart Dataset ⇨, Mpii Dataset ⇨, Coco Dataset ⇨, Crowdpose Dataset ⇨)

    • [BACKBONE] Shufflenet V2: Practical Guidelines for Efficient CNN Architecture Design (Mpii Dataset ⇨, Coco Dataset ⇨)

    • [BACKBONE] Shufflenet: An Extremely Efficient Convolutional Neural Network for Mobile Devices (Mpii Dataset ⇨, Coco Dataset ⇨)

    • [BACKBONE] Squeeze-and-Excitation Networks (Mpii Dataset ⇨, Coco Dataset ⇨)

    • [BACKBONE] Swin Transformer: Hierarchical Vision Transformer Using Shifted Windows (Coco Dataset ⇨)

    • [BACKBONE] Very Deep Convolutional Networks for Large-Scale Image Recognition (Coco Dataset ⇨)

    • [BACKBONE] Vipnas: Efficient Video Pose Estimation via Neural Architecture Search (Coco Dataset ⇨)

    • [BACKBONE] Vi{tp (Humanart Dataset ⇨, Coco Dataset ⇨)

    • [BACKBONE] Yolox: Exceeding Yolo Series in 2021 (Coco Dataset ⇨)

    • [DATASET] 2d Human Pose Estimation: New Benchmark and State of the Art Analysis (Mpii Dataset ⇨, Body7 Dataset ⇨)

    • [DATASET] Ai Challenger: A Large-Scale Dataset for Going Deeper in Image Understanding (Aic Dataset ⇨, Coco Dataset ⇨, Body7 Dataset ⇨)

    • [DATASET] Alphapose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Time (Body8 Dataset ⇨)

    • [DATASET] Crowdpose: Efficient Crowded Scenes Pose Estimation and a New Benchmark (Crowdpose Dataset ⇨, Body7 Dataset ⇨)

    • [DATASET] Human Pose Estimation in Extremely Low-Light Conditions (Exlpose Dataset ⇨)

    • [DATASET] Microsoft Coco: Common Objects in Context (Body8 Dataset ⇨, Humanart Dataset ⇨, Coco Dataset ⇨, Body7 Dataset ⇨)

    • [DATASET] Posetrack: A Benchmark for Human Pose Estimation and Tracking (Posetrack18 Dataset ⇨, Body7 Dataset ⇨)

    • [DATASET] Towards Understanding Action Recognition (JHMDB Dataset ⇨, Body7 Dataset ⇨)

    • [DATASET] Whole-Body Human Pose Estimation in the Wild (Humanart Dataset ⇨, Coco Dataset ⇨)

    • [OTHERS] Albumentations: Fast and Flexible Image Augmentations (Coco Dataset ⇨)

    • [OTHERS] Feature Pyramid Networks for Object Detection (Coco Dataset ⇨)

    • [OTHERS] Mixed Precision Training (Coco Dataset ⇨)

Body 3D Keypoint

  • Number of papers: 4

    • [BACKBONE] 3d Human Pose Estimation in Video With Temporal Convolutions and Semi-Supervised Training (H36m Dataset ⇨)

    • [BACKBONE] A Simple Yet Effective Baseline for 3d Human Pose Estimation (H36m Dataset ⇨)

    • [BACKBONE] Learning Human Motion Representations: A Unified Perspective (H36m Dataset ⇨)

    • [DATASET] Human3.6m: Large Scale Datasets and Predictive Methods for 3d Human Sensing in Natural Environments (H36m Dataset ⇨)

Face 2D Keypoint

  • Number of papers: 22

    • [ALGORITHM] Adaptive Wing Loss for Robust Face Alignment via Heatmap Regression (WFLW Dataset ⇨)

    • [ALGORITHM] Deep High-Resolution Representation Learning for Visual Recognition (Aflw Dataset ⇨, 300wlp Dataset ⇨, 300w Dataset ⇨, Cofw Dataset ⇨, Coco_wholebody_face Dataset ⇨, WFLW Dataset ⇨)

    • [ALGORITHM] Deeppose: Human Pose Estimation via Deep Neural Networks (WFLW Dataset ⇨)

    • [ALGORITHM] Distribution-Aware Coordinate Representation for Human Pose Estimation (Aflw Dataset ⇨, Coco_wholebody_face Dataset ⇨, WFLW Dataset ⇨)

    • [ALGORITHM] Improving Convolutional Networks With Self-Calibrated Convolutions (Coco_wholebody_face Dataset ⇨)

    • [ALGORITHM] Rtmdet: An Empirical Study of Designing Real-Time Object Detectors (Lapa Dataset ⇨, Face6 Dataset ⇨, Coco_wholebody_face Dataset ⇨, WFLW Dataset ⇨)

    • [ALGORITHM] Rtmpose: Real-Time Multi-Person Pose Estimation Based on Mmpose (Face6 Dataset ⇨)

    • [ALGORITHM] Simple Baselines for Human Pose Estimation and Tracking (Coco_wholebody_face Dataset ⇨)

    • [ALGORITHM] Stacked Hourglass Networks for Human Pose Estimation (Coco_wholebody_face Dataset ⇨)

    • [ALGORITHM] Structure-Coherent Deep Feature Learning for Robust Face Alignment (WFLW Dataset ⇨)

    • [ALGORITHM] Wing Loss for Robust Facial Landmark Localisation With Convolutional Neural Networks (WFLW Dataset ⇨)

    • [BACKBONE] Deep Residual Learning for Image Recognition (Coco_wholebody_face Dataset ⇨, WFLW Dataset ⇨)

    • [BACKBONE] Mobilenetv2: Inverted Residuals and Linear Bottlenecks (Coco_wholebody_face Dataset ⇨)

    • [BACKBONE] Rtmdet: An Empirical Study of Designing Real-Time Object Detectors (Lapa Dataset ⇨, Face6 Dataset ⇨, Coco_wholebody_face Dataset ⇨, WFLW Dataset ⇨)

    • [DATASET] 300 Faces in-the-Wild Challenge: Database and Results (300w Dataset ⇨)

    • [DATASET] A New Dataset and Boundary-Attention Semantic Segmentation for Face Parsing. (Lapa Dataset ⇨)

    • [DATASET] Annotated Facial Landmarks in the Wild: A Large-Scale, Real-World Database for Facial Landmark Localization (Aflw Dataset ⇨)

    • [DATASET] Face Alignment in Full Pose Range: A 3d Total Solution (300wlp Dataset ⇨)

    • [DATASET] Look at Boundary: A Boundary-Aware Face Alignment Algorithm (WFLW Dataset ⇨)

    • [DATASET] Microsoft Coco: Common Objects in Context (Face6 Dataset ⇨)

    • [DATASET] Robust Face Landmark Estimation Under Occlusion (Cofw Dataset ⇨)

    • [DATASET] Whole-Body Human Pose Estimation in the Wild (Coco_wholebody_face Dataset ⇨)

Fashion 2D Keypoint

  • Number of papers: 7

    • [ALGORITHM] Deep High-Resolution Representation Learning for Human Pose Estimation (Deepfashion Dataset ⇨)

    • [ALGORITHM] Simple Baselines for Human Pose Estimation and Tracking (Deepfashion2 Dataset ⇨, Deepfashion Dataset ⇨)

    • [BACKBONE] Deep Residual Learning for Image Recognition (Deepfashion2 Dataset ⇨, Deepfashion Dataset ⇨)

    • [BACKBONE] The Devil Is in the Details: Delving Into Unbiased Data Processing for Human Pose Estimation (Deepfashion Dataset ⇨)

    • [DATASET] A Versatile Benchmark for Detection, Pose Estimation, Segmentation and Re-Identification of Clothing Images (Deepfashion2 Dataset ⇨)

    • [DATASET] Deepfashion: Powering Robust Clothes Recognition and Retrieval With Rich Annotations (Deepfashion Dataset ⇨)

    • [DATASET] Fashion Landmark Detection in the Wild (Deepfashion Dataset ⇨)

Hand 2D Keypoint

  • Number of papers: 18

    • [ALGORITHM] Deep High-Resolution Representation Learning for Visual Recognition (Coco_wholebody_hand Dataset ⇨, Rhd2d Dataset ⇨, Onehand10k Dataset ⇨)

    • [ALGORITHM] Deeppose: Human Pose Estimation via Deep Neural Networks (Rhd2d Dataset ⇨, Onehand10k Dataset ⇨)

    • [ALGORITHM] Distribution-Aware Coordinate Representation for Human Pose Estimation (Coco_wholebody_hand Dataset ⇨, Rhd2d Dataset ⇨, Onehand10k Dataset ⇨)

    • [ALGORITHM] Improving Convolutional Networks With Self-Calibrated Convolutions (Coco_wholebody_hand Dataset ⇨)

    • [ALGORITHM] Lite-Hrnet: A Lightweight High-Resolution Network (Coco_wholebody_hand Dataset ⇨)

    • [ALGORITHM] Rtmdet: An Empirical Study of Designing Real-Time Object Detectors (Hand5 Dataset ⇨, Coco_wholebody_hand Dataset ⇨)

    • [ALGORITHM] Rtmpose: Real-Time Multi-Person Pose Estimation Based on Mmpose (Hand5 Dataset ⇨)

    • [ALGORITHM] Simple Baselines for Human Pose Estimation and Tracking (Freihand2d Dataset ⇨, Coco_wholebody_hand Dataset ⇨, Rhd2d Dataset ⇨, Onehand10k Dataset ⇨)

    • [ALGORITHM] Stacked Hourglass Networks for Human Pose Estimation (Coco_wholebody_hand Dataset ⇨)

    • [ALGORITHM] The Devil Is in the Details: Delving Into Unbiased Data Processing for Human Pose Estimation (Rhd2d Dataset ⇨, Onehand10k Dataset ⇨)

    • [BACKBONE] Deep Residual Learning for Image Recognition (Freihand2d Dataset ⇨, Coco_wholebody_hand Dataset ⇨, Rhd2d Dataset ⇨, Onehand10k Dataset ⇨)

    • [BACKBONE] Mobilenetv2: Inverted Residuals and Linear Bottlenecks (Coco_wholebody_hand Dataset ⇨, Rhd2d Dataset ⇨, Onehand10k Dataset ⇨)

    • [BACKBONE] Rtmdet: An Empirical Study of Designing Real-Time Object Detectors (Hand5 Dataset ⇨, Coco_wholebody_hand Dataset ⇨)

    • [DATASET] Freihand: A Dataset for Markerless Capture of Hand Pose and Shape From Single RGB Images (Freihand2d Dataset ⇨)

    • [DATASET] Learning to Estimate 3d Hand Pose From Single RGB Images (Rhd2d Dataset ⇨)

    • [DATASET] Mask-Pose Cascaded CNN for 2d Hand Pose Estimation From Single Color Image (Onehand10k Dataset ⇨)

    • [DATASET] Microsoft Coco: Common Objects in Context (Hand5 Dataset ⇨)

    • [DATASET] Whole-Body Human Pose Estimation in the Wild (Coco_wholebody_hand Dataset ⇨)

Hand 3D Keypoint

  • Number of papers: 3

    • [ALGORITHM] Interhand2.6m: A Dataset and Baseline for 3d Interacting Hand Pose Estimation From a Single RGB Image (Interhand3d Dataset ⇨)

    • [BACKBONE] Deep Residual Learning for Image Recognition (Interhand3d Dataset ⇨)

    • [DATASET] Interhand2.6m: A Dataset and Baseline for 3d Interacting Hand Pose Estimation From a Single RGB Image (Interhand3d Dataset ⇨)

Wholebody 2D Keypoint

  • Number of papers: 10

    • [ALGORITHM] Deep High-Resolution Representation Learning for Human Pose Estimation (Coco-Wholebody Dataset ⇨, Ubody2d Dataset ⇨)

    • [ALGORITHM] Distribution-Aware Coordinate Representation for Human Pose Estimation (Coco-Wholebody Dataset ⇨)

    • [ALGORITHM] Rtmdet: An Empirical Study of Designing Real-Time Object Detectors (Cocktail14 Dataset ⇨, Coco-Wholebody Dataset ⇨)

    • [ALGORITHM] Rtmpose: Real-Time Multi-Person Pose Estimation Based on Mmpose (Cocktail14 Dataset ⇨, Coco-Wholebody Dataset ⇨)

    • [ALGORITHM] Simple Baselines for Human Pose Estimation and Tracking (Coco-Wholebody Dataset ⇨)

    • [ALGORITHM] The Devil Is in the Details: Delving Into Unbiased Data Processing for Human Pose Estimation (Coco-Wholebody Dataset ⇨)

    • [ALGORITHM] Vipnas: Efficient Video Pose Estimation via Neural Architecture Search (Coco-Wholebody Dataset ⇨)

    • [BACKBONE] Rtmdet: An Empirical Study of Designing Real-Time Object Detectors (Cocktail14 Dataset ⇨, Coco-Wholebody Dataset ⇨)

    • [DATASET] One-Stage 3d Whole-Body Mesh Recovery With Component Aware Transformer (Ubody2d Dataset ⇨)

    • [DATASET] Whole-Body Human Pose Estimation in the Wild (Cocktail14 Dataset ⇨, Coco-Wholebody Dataset ⇨)