Real-time CNN-based segmentation architecture for ball detection in a single view setup
(2019) ACM Multimedia 2019, 2nd International Workshop on Multimedia Content Analysis in Sports — Location: Nice (21.October.2019)
Files
ballseg-ACM-compressed.pdf
Anonymous version, compressed
Open Access - Adobe PDF
- 1.43 MB
Details
- Authors
Van Zandycke, Gabriel 
- Abstract
- This paper considers the task of detecting the ball from a single viewpoint in the challenging but common case where the ball interacts frequently with players while being poorly contrasted with respect to the background. We propose a novel approach by formulating the problem as a segmentation task solved by an efficient CNN architecture. To take advantage of the ball dynamics, the network is fed with a pair of consecutive images. Our inference model can run in real time without the delay induced by a temporal analysis. We also show that test-time data augmentation allows for a significant increase the detection accuracy. As an additional contribution, we publicly release the dataset on which this work is based.
- Affiliations
Citations
Van Zandycke, G., & De Vleeschouwer, C. (2019). Real-time CNN-based segmentation architecture for ball detection in a single view setup. Proceedings of ACM Multimedia 2019, 1(1), 8. https://hdl.handle.net/2078.5/126785 (Original work published 2019)