# DeepBall: Deep Neural-Network Ball Detector

## TL;DR

DeepBall is a fully convolutional neural network designed for accurate ball detection in long shot videos, leveraging hypercolumn features to incorporate extensive visual context and achieve state-of-the-art results.

## Contribution

It introduces a novel deep network architecture with hypercolumn features for improved ball detection accuracy in sports videos.

## Key findings

- Achieves state-of-the-art detection accuracy on ISSIA-CNR Soccer Dataset.
- Operates on images of any size due to fully convolutional design.
- Utilizes hypercolumn features to incorporate multi-level visual context.

## Abstract

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.