Activity-Aware Video Stabilization for BallCam - Robotics Institute Carnegie Mellon University

Activity-Aware Video Stabilization for BallCam

Ryohei Funakoshi, Vishnu Naresh Boddeti, Kris Kitani, and Hideki Koike
Conference Paper, Proceedings of 29th Annual Symposium on User Interface Software and Technology (UIST '16), pp. 197 - 198, October, 2016

Abstract

We present a video stabilization algorithm for ball camera systems that undergo extreme egomotion during sports play. In particular, we focus on the BallCam system which is an American football embedded with an action camera at the tip of the ball. We propose an activity-aware video stabilization algorithm which is able to understand the current activity of the BallCam, which uses estimated activity labels to inform a robust video stabilization algorithm. Activity recognition is performed with a deep convolutional neural network, which uses optical flow.

BibTeX

@conference{Funakoshi-2016-109799,
author = {Ryohei Funakoshi and Vishnu Naresh Boddeti and Kris Kitani and Hideki Koike},
title = {Activity-Aware Video Stabilization for BallCam},
booktitle = {Proceedings of 29th Annual Symposium on User Interface Software and Technology (UIST '16)},
year = {2016},
month = {October},
pages = {197 - 198},
}