New Variants of Frank-Wolfe Algorithm for Video Co-localization Problem

TL;DR

This paper introduces new variants of the Frank-Wolfe algorithm tailored for video co-localization, demonstrating improved efficiency through numerical experiments on the YouTube-Objects dataset.

Contribution

The paper proposes novel Frank-Wolfe algorithm variants inspired by the conditional gradient sliding method for more efficient video co-localization.

Findings

Improved convergence rate in Wolfe gap reduction

Enhanced efficiency on YouTube-Objects dataset

Effective in localizing objects across video frames

Abstract

The co-localization problem is a model that simultaneously localizes objects of the same class within a series of images or videos. In \cite{joulin2014efficient}, authors present new variants of the Frank-Wolfe algorithm (aka conditional gradient) that increase the efficiency in solving the image and video co-localization problems. The authors show the efficiency of their methods with the rate of decrease in a value called the Wolfe gap in each iteration of the algorithm. In this project, inspired by the conditional gradient sliding algorithm (CGS) \cite{CGS:Lan}, We propose algorithms for solving such problems and demonstrate the efficiency of the proposed algorithms through numerical experiments. The efficiency of these methods with respect to the Wolfe gap is compared with implementing them on the YouTube-Objects dataset for videos.