DeepBall: Deep Neural-Network Ball Detector
This addresses the problem of accurate ball detection in sports analytics, specifically for soccer, but is incremental as it builds on existing deep learning and hypercolumn concepts.
The paper tackles ball detection in long shot videos by introducing a fully convolutional deep network that uses hypercolumn features to incorporate larger visual context, achieving state-of-the-art results on the ISSIA-CNR Soccer Dataset.
The paper describes a deep network based object detector specialized for ball detection in long shot videos. Due to its fully convolutional design, the method operates on images of any size and produces \emph{ball confidence map} encoding the position of detected ball. The network uses hypercolumn concept, where feature maps from different hierarchy levels of the deep convolutional network are combined and jointly fed to the convolutional classification layer. This allows boosting the detection accuracy as larger visual context around the object of interest is taken into account. The method achieves state-of-the-art results when tested on publicly available ISSIA-CNR Soccer Dataset.