Abstract
In this paper, we tackle the problem of performing efficient co-localization in images and videos. Co-localization is the problem of si- multaneously localizing (with bounding boxes) ob jects of the same class across a set of distinct images or videos. Building upon recent state- of-the-art methods, we show how we are able to naturally incorporate temporal terms and constraints for video co-localization into a quadratic programming framework. Furthermore, by leveraging the Frank-Wolfe al- gorithm (or conditional gradient), we show how our optimization formu- lations for both images and videos can be reduced to solving a succession of simple integer programs, leading to increased efficiency in both mem- ory and speed. To validate our method, we present experimental results on the PASCAL VOC 2007 dataset for images and the YouTube-Ob jects dataset for videos, as well as a joint combination of the two.