An end-to-end pipeline that turns raw video into quantified, classifiable behavior: video processing, pose estimation (MediaPipe and YOLO), joint angles, derived indices, and sequence-based classification. Work through it in numbered order.
Prerequisites: Module 00B (Python Fundamentals)
Estimated time: ~12–18 hours
After this module you can extract body keypoints from video, compute kinematic features, and train a behavior classifier.
In this module¶
- Loading Data and Doing Useful Work — Video
- Frame Reduction Script — Understanding the Packages
- Video Chunker by Frame Count (OpenCV)
- Video Pair Clip Generator
- Pose Landmarks with MediaPipe — From Local Videos & Folders Using Python
- Human Pose Estimation for Sports Biomechanics using YOLO
- LS100 — Batch GPU Pose Estimation with YOLOv8-Pose (on Google Colab)
- LS100 — GPU Pose Estimation with YOLOv8-Pose (Colab)
- Training a YOLO Pose Model for Custom Keypoint Tracking
- LS100: Joint Angles & Derived Metrics
- LS100: Joint Angles and Derived Metrics
- LS100: Biomechanical Indices from Pose Angles
- Sequence-Based Classification using Machine Learning
- Sequence-Based Classification using Machine Learning