Event Detection in Broadcasting Video for Halfpipe Sports
In recorded broadcasts, there are the best tricks performed by elite players of halfpipe sports.
These videos can inspire players to learn moves, create new tricks, or improve their skills.
However, finding a particular trick in a video is time-consuming because certain tricks may be sparse in the video and the timestamps of them are unknown to a video watcher.
Recognizing the trick type in halfpipe videos is challenging because:
- there are often more than one shot taken at different locations by shoulder-mount camcorders in a single trick segment, and
- the scenes in a halfpipe video are motion blurred since players move fast.
Fortunately, with the help of the HP court color coverage ratio and salient object detection mechanisms, we overcome the challenge.
In this work, a low-cost and efficient system is proposed to automatically analyze the halfpipe sports videos. Besides, a novel and efficient method for detecting the spin event is proposed on the basis of native motion vectors existing in a compressed video.
Hao-Kai Wen, Wei-Che Chang, Chia-Hu Chang, Yin-Tzu Lin, Ja-Ling Wu
- J. Harding, C. Mackintosh, A. Hahn, and D. James. Classification of aerial acrobatics in elite half-pipe snowboarding using body mounted inertial sensors. In The Engineering of Sport 7, pages 447-456. Springer Paris, 2008.
- T. Liu, Z. Yuan, J. Sun, J. Wang, N. Zheng, X. Tang, and H.-Y. Shum. Learning to detect a salient object. IEEE trans. PAMI, 33(2), 2011.
- J. Ning, L. Zhang, D. Zhang, and C. Wu. Robust mean-shift tracking with corrected background-weighted histogram. Computer Vision, IET, 6(1):62-69, January 2012.