Video segmentation and stabilization for BallCam - Robotics Institute Carnegie Mellon University

Video segmentation and stabilization for BallCam

Ryohei Funakoshi, Vishnu Naresh Boddeti, Kris M. Kitani, and Hideki Koike
Conference Paper, Proceedings of 8th Augmented Human International Conference (AH '17), March, 2017

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-2017-109794,
author = {Ryohei Funakoshi and Vishnu Naresh Boddeti and Kris M. Kitani and Hideki Koike},
title = {Video segmentation and stabilization for BallCam},
booktitle = {Proceedings of 8th Augmented Human International Conference (AH '17)},
year = {2017},
month = {March},
}