Real-time CNN-based Segmentation Architecture for Ball Detection in a Single View Setup
This work addresses a specific challenge in sports analytics for real-time ball detection, but it is incremental as it builds on existing CNN methods for segmentation.
The paper tackles the problem of detecting a ball in single-view videos where it is poorly contrasted and frequently interacts with players, proposing a CNN-based segmentation approach that uses consecutive images and test-time augmentation to achieve real-time inference with improved accuracy.
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.