Human-Pose-DeepLearning-List

Some recent (2016-now) Human-Pose related deep learning studies. If you find any errors or problems, please feel free to comment and PR. Early version is shared on [Zhihu] and [Mind-map].

Content

Human-Pose Estimation
Human-Pose Recognition
Human-Pose Generation
Human-Pose Reconstruction and Rendering (Mesh or Appearance)
Open-source Toolbox

Human-Pose Estimation

1. monocular single image

2D human pose (heatmap-based)

[Top-down]

2D human pose (regression-based)

Integral Human Pose Regression
LCR-Net: Localization-Classification-Regression for Human Pose

2D human pose (vector-based)

3D human pose

[Non-rigid Structure from Motion]

c3dpo
Deep Interpretable Non-Rigid Structure from Motion

[2D lift to 3D]

A simple yet effective baseline for 3d human pose estimation

[Depth-aware]

HMOR

[Others]

Unsupervised Cross-Modal Alignment for Multi-Person 3D Pose Estimation
PI-Net: Pose Interacting Network for Multi-Person Monocular 3D Pose Estimation
Semantic Graph Convolutional Networks for 3D Human Pose Regression

2. heatmap-aware

UDP
DarkPose
Rethink

3. occlusion-aware

GAN

Adversarial PoseNet
Adversarial Semantic Data Augmentation for Human Pose Estimation

Post-processing

CrowdPose
Peeking into occluded joints

Data augmentation

Adversarial Semantic Data Augmentation for Human Pose Estimation
3D Human Pose Estimation using Spatio-Temporal Networks with Explicit Occlusion Training
Occlusion-Aware Networks for 3D Human Pose Estimation in Video
Occlusion-Aware Siamese Network for Human Pose Estimation

Others

A Semantic Occlusion Model for Human Pose Estimation from a Single Depth
Occluded Joints Recovery in 3D Human Pose Estimation based on Distance Matrix
Object-Occluded Human Shape and Pose Estimation from a Single Color Image
LCR-Net Localization-Classification-Regression for Human Pose

4. multi-view images

Cross-view

Adaptive Multiview Fusion for Accurate Human Pose Estimation in the wild
Cross View Fusion for 3D Human Pose Estimation
Cross-View Tracking for Multi-Human 3D Pose Estimation at over 100 FPS
Fast and Robust Multi-Person 3D Pose Estimation from Multiple Views
Fusing Wearable IMUs with Multi-View Images for Human Pose Estimation A Geometric Approach
View-Invariant Probabilistic Embedding for Human Pose

Mirror

5. video

2D

Combining detection and tracking for human pose estimation in videos
Key Frame Proposal Network for Efficient Pose Estimation in Videos
Learning Temporal Pose Estimation from Sparsely-Labeled Videos
Personalizing Human Video Pose Estimation
Self-supervised Keypoint Correspondences for Multi-Person Pose Estimation and Tracking in Videos

3D

VideoPose3D
PoseNet3D
Motion Guided 3D Pose Estimation from Videos
2D or 3D Pose Estimation and Action Recognition using Multitask Deep Learning
3D Human Pose Estimation from Monocular Video
Attention Mechanism Exploits Temporal Contexts Real-time 3D Human Pose Reconstruction
VNect

6. RGBD image

3D Human Pose Estimation in RGBD Images for Robotic Task Learning
Human Pose Estimation for RGBD Imagery with Multi-Channel Mixture of Parts and Kinematic Constraints

Human-Pose Recognition

1. action classification

2D or 3D Pose Estimation and Action Recognition using Multitask Deep Learning
SlowFast
Long-term Feature Bank
AlphAction
2D or 3D Pose Estimation and Action Recognition using Multitask Deep Learning

2. human-object interaction

HOI-Learning-List

Human-Pose Generation

1. motion transfer

Liquid Warping GAN with Attention: A Unified Framework for Human Image Synthesis
FOMM
MRAA
Thin

2. audio to pose

speech to pose

Speech2Gesture: Learning Individual Styles of Conversational Gesture
Speech2Video: Synthesis with 3D Skeleton Regularization and Expressive Body Poses

music to pose

Dancing to Music
ChoreoNet: Towards Music to Dance Synthesis with
LISTEN TO DANCE
Music2Dance
Audio to Body Dynamics
Multi-Instrumentalist Net Unsupervised Generation of Music from Body Movements
Dance with Melody: An LSTM-autoencoder Approach to

3. text to pose

Text2Action
Language2Pose: Natural Language Grounded Pose Forecasting

4. multi-model to pose

TriModel

Human-Pose Reconstruction and Rendering (Mesh or Appearance)

1. parametric model

Human Mesh Recovery from Monocular Images via a Skeleton-disentangled Representation
VIBE: Video Inference for Human Body Pose and Shape Estimation
METRO

2. NeRF model

Neural Body: Implicit Neural Representations With Structured Latent Codes for Novel View Synthesis of Dynamic Humans (CVPR 2021) [Project] [Code] [Paper]
Neural actor: neural free-view synthesis of human actors with pose control (SIGGRAPH Asia 2021) [Project] [Code] [Paper]
Animatable Neural Radiance Fields for Modeling Dynamic Human Bodies (ICCV 2021) [Project] [Code] [Paper]
A-NeRF: Articulated Neural Radiance Fields for Learning Human Shape, Appearance, and Pose (NeurIPS 2021) [Project] [Code] [Paper]
Neural Articulated Radiance Field (ICCV 2021) [Project] [Code] [Paper]
HumanNeRF: Free-viewpoint Rendering of Moving People from Monocular Video (CVPR 2022) [Project] [Code] [Paper]
HumanNeRF: Efficiently Generated Human Radiance Field from Sparse Inputs (CVPR 2022) [Project] [Code] [Paper]

From https://github.com/jintaiWang/NeRF-About-Human-Pose-Reconstruction-and-Rendering

WinstonDeng / Human-Pose-DeepLearning-List