AI in Sports - Basketball Rule Violation Detection
summary
Enetel Solutions builds a Basketball Rule Violation Detection System (in collaboration with Professor Dr Vladimir Lučić) that detects pivot/travel violations by combining ball detection, ball-bounce heuristics and pose tracking into a player state machine. Implementation uses YOLO on a few thousand images for ball detection and AlphaPose for pose estimation, runs on static single‑camera training videos with one player, and achieves reliable lab results but struggles with occlusions in multi-player games and needs extra cameras or sensors for real‑match deployment.
AI-written summary · about
chapters (7)
- 0:00Start
- 10:04Ball detection
- 10:53Ball trajectory key moments
- 11:31Pose estimation
- 12:42Player state machine
- 16:38Final output - violation detected
- 17:06Use cases
related talks
42mDeep Learning for Automatic Basketball Video Production
20mAI-Driven Real-Time Ball Tracking for Live Sports Streaming
35mFootball video analysis using Deep Learning
41mThe evolution of basketball with data science
31mCV at the Dutch Tennis Federation: Utilizing YOLO to create insights for coaches
04mInteractive Sports Analytics
41mAI in Web: TensorFlow js
42mHow Deep Learning is Revolutionizing Sports Analytics