- Classification
- Dimensionality
- 2D
 - 3D
 
 - Subject Count
- Single Person
 - Multiple People
 
 - Human Body Representation
- Skeleton
 - Contour
 - Mesh
 
 - Input Modalities
- RGB
 - RGB-D
 - Infrared
 - Multi-View Systems
 - WiFi Signals
 
 - Learning Paradigms
- Supervised Learning
 - Unsupervised and Self-Supervised Learning
 - Unsupervised Domain Adaptation (UDA)
 
 
 - Dimensionality
 
- Architectural Evolution
- Pre-deep learning era
- Pictorial Structures Model
 
 - Dawn of deep learning era
- CNN based regression with DeepPose
 
 - Heatmap Representation
- Stacked Hourglass Networks
 - HRNet
 
 - Bottom-up architectures
- OpenPose
 
 - The Transformer Era
- PoseFormer
 - ViTPose
 
 - Hybrid Archutectures
- Fusing CNNs
 - Transformer-GCNs
 
 
 - Pre-deep learning era
 
- 
Human Pose Estimation Technology in Fitness & Rehab Therapy Apps
 - 
Human pose estimation and its application to action recognition: A survey (PDF)
 - 
HP-YOLO: A Lightweight Real-Time Human Pose Estimation Method - MDPI
 - 
Human Pose Estimation - Everything You Need to Know - viso.ai
 - 
A Comparative Study of Human Pose Estimation - ResearchGate (PDF)
 - 
Vision-based Human Pose Estimation via Deep Learning: A Survey - arXiv (PDF)
 - 
CNN and Transformer-Based Human Pose Estimation - Encyclopedia
 - 
Leveraging Temporal Context in Human Pose Estimation: A Survey - SciTePress (PDF)
 - 
State of the Art in Monocular 3D Human Pose Estimation - MDPI
 - 
Advancing Human Pose Estimation with Transformer Models - ResearchGate
 - 
Human Pose Estimation Based on 2D and 3D Classification - SciTePress
 - 
Benchmarking 3D Human Pose Estimation Models under Occlusions - arXiv
 - 
Occlusion-Aware Networks for 3D Human Pose Estimation in Video - ICCV
 - 
3D Human Pose Estimation with Occlusions: BlendMimic3D - CVPRW 2024