Abstract
Tracking a table tennis ball for umpiring purposes is a challenging task as, in real-match scenarios, the ball travels fast and can become occluded or merged with other background objects. This paper presents the design of a multi-view based tracking system that can overcome the challenges of tracking a ball in real match sequences. The system has been tested on a complete table tennis rally and the results are very promising. The system is able to continuously track the ball with only marginal variations in detection. Furthermore, the initialization of the multi-camera system means it is both a portable and cost-effective solution for umpiring purposes.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 20th International Conference on Image Processing, Computer Vision, & Pattern Recognition: IPCV 2016 |
| Editors | Hamid R. Arabnia, Leonidas Deligiannidis, Fernando G. Tinetti, Jane You |
| Publisher | CSREA Press |
| Pages | 119-125 |
| Number of pages | 7 |
| ISBN (Print) | 9781601323637, 1601323638 |
| Publication status | Published - 28 Jul 2016 |
| Event | 20th International Conference on Image Processing, Computer Vision, & Pattern Recognition - Las Vegas, United States Duration: 25 Jul 2016 → 28 Jul 2016 |
Conference
| Conference | 20th International Conference on Image Processing, Computer Vision, & Pattern Recognition |
|---|---|
| Abbreviated title | IPCV 2016 |
| Country/Territory | United States |
| City | Las Vegas |
| Period | 25/07/16 → 28/07/16 |
Keywords
- object detection
- tracking
- multi-view
- multi-agent system
- table tennis