hyunsungkim-ds/ballradar
[KDD 2023] Ball Trajectory Inference from Multi-Agent Sports Contexts Using Set Transformer and Hierarchical Bi-LSTM
This project helps sports analysts and coaches accurately understand ball movement in soccer matches. By taking player tracking data as input, it identifies which player possesses the ball and predicts its trajectory, even when the ball isn't directly observed. The output includes inferred ball paths and animated visualizations, useful for tactical analysis and player performance review.
Use this if you need to reconstruct unseen ball trajectories from player movement data in soccer, providing insights for tactical analysis or performance evaluation.
Not ideal if you already have complete and accurate ball tracking data, as this tool is designed for inference when ball data is missing or unreliable.
Stars
30
Forks
5
Language
Jupyter Notebook
License
—
Category
Last pushed
Feb 03, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/hyunsungkim-ds/ballradar"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
roboflow/sports
computer vision and sports
chonyy/AI-basketball-analysis
:basketball::robot::basketball: AI web app and API to analyze basketball shots and shooting pose.
mradovic38/football_analysis
A comprehensive tool for processing and analyzing video footage, producing detailed insights...
KieDani/UpliftingTableTennis
Official implementation of the paper "Uplifting Table Tennis: A Robust, Real-World Application...
wmcnally/deep-darts
DeepDarts is the first deep learning-based automatic scoring system for steel-tip darts. It...