Overview

  • Number of checkpoints: 229

  • Number of configs: 230

  • Number of papers: 26

    • ALGORITHM: 16

    • BACKBONE: 10

For supported datasets, see datasets overview.

Animal Models

  • Number of checkpoints: 39

  • Number of configs: 39

  • Number of papers: 2

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

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

Bottom Up Models

  • Number of checkpoints: 18

  • Number of configs: 18

  • Number of papers: 6

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

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

    • [ALGORITHM] Higherhrnet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimation ()

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

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

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

Face Models

  • Number of checkpoints: 6

  • Number of configs: 6

  • Number of papers: 5

    • [ALGORITHM] Deep High-Resolution Representation Learning for Visual Recognition ()

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

    • [ALGORITHM] Distribution-Aware Coordinate Representation for Human Pose Estimation ()

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

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

Fashion Models

  • Number of checkpoints: 6

  • Number of configs: 6

  • Number of papers: 2

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

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

Hand Models

  • Number of checkpoints: 23

  • Number of configs: 23

  • Number of papers: 6

    • [ALGORITHM] Deep High-Resolution Representation Learning for Visual Recognition ()

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

    • [ALGORITHM] Distribution-Aware Coordinate Representation for Human Pose Estimation ()

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

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

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

Mesh Models

  • Number of checkpoints: 1

  • Number of configs: 1

  • Number of papers: 1

    • [ALGORITHM] End-to-End Recovery of Human Shape and Pose ()

Top Down Models

  • Number of checkpoints: 124

  • Number of configs: 125

  • Number of papers: 20

    • [ALGORITHM] Albumentations: Fast and Flexible Image Augmentations ()

    • [ALGORITHM] Convolutional Pose Machines ()

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

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

    • [ALGORITHM] Distribution-Aware Coordinate Representation for Human Pose Estimation ()

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

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

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

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

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

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

    • [BACKBONE] Aggregated Residual Transformations for Deep Neural Networks ()

    • [BACKBONE] Bag of Tricks for Image Classification With Convolutional Neural Networks ()

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

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

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

    • [BACKBONE] Shufflenet V2: Practical Guidelines for Efficient CNN Architecture Design ()

    • [BACKBONE] Shufflenet: An Extremely Efficient Convolutional Neural Network for Mobile Devices ()

    • [BACKBONE] Squeeze-and-Excitation Networks ()

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

Whole-Body Models

  • Number of checkpoints: 12

  • Number of configs: 12

  • Number of papers: 3

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

    • [ALGORITHM] Distribution-Aware Coordinate Representation for Human Pose Estimation ()

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