2015/11/19 by Andrii Maksai, Xinchao Wang, Maksai, Andrii +3 · 2 citations
Computer Science · Engineering · #Video Analysis and Summarization #Human Pose and Action Recognition #Human Motion and Animation
paper · pdf · doi:10.48550/arxiv.1511.06181
Tracking the ball is critical for video-based analysis of team sports.\nHowever, it is difficult, especially in low-resolution images, due to the small\nsize of the ball, its speed that creates motion blur, and its often being\noccluded by players. In this paper, we propose a generic and principled\napproach to modeling the interaction between the ball and the players while\nalso imposing appropriate physical constraints on the ball's trajectory. We\nshow that our approach, formulated in terms of a Mixed Integer Program, is more\nrobust and more accurate than several state-of-the-art approaches on real-life\nvolleyball, basketball, and soccer sequences.\n