TOTNet: Occlusion-Aware Temporal Tracking for Robust Ball Detection in Sports Videos

TL;DR

TOTNet is a novel occlusion-aware temporal tracking network that significantly improves ball detection accuracy in sports videos, especially under occlusion, by leveraging 3D convolutions, visibility-weighted loss, and occlusion augmentation.

Contribution

The paper introduces TOTNet, a new deep learning model for robust ball tracking under occlusion, and a new occlusion-rich table tennis dataset for sports analytics.

Findings

Outperforms prior methods with RMSE reduced from 37.30 to 7.19

Achieves higher accuracy on fully occluded frames, from 0.63 to 0.80

Demonstrates effectiveness across tennis, badminton, and table tennis datasets

Abstract

Robust ball tracking under occlusion remains a key challenge in sports video analysis, affecting tasks like event detection and officiating. We present TOTNet, a Temporal Occlusion Tracking Network that leverages 3D convolutions, visibility-weighted loss, and occlusion augmentation to improve performance under partial and full occlusions. Developed in collaboration with Paralympics Australia, TOTNet is designed for real-world sports analytics. We introduce TTA, a new occlusion-rich table tennis dataset collected from professional-level Paralympic matches, comprising 9,159 samples with 1,996 occlusion cases. Evaluated on four datasets across tennis, badminton, and table tennis, TOTNet significantly outperforms prior state-of-the-art methods, reducing RMSE from 37.30 to 7.19 and improving accuracy on fully occluded frames from 0.63 to 0.80. These results demonstrate TOTNets effectiveness for offline sports analytics in fast-paced scenarios. Code and data access:\href{https://github.com/AugustRushG/TOTNet}{AugustRushG/TOTNet}.