CoCoLoT: Combining Complementary Trackers in Long-Term Visual Tracking

TL;DR

This paper introduces CoCoLoT, a framework that combines multiple visual trackers to improve long-term object tracking by dynamically selecting and correcting tracker performance using an online deep verification model.

Contribution

It presents a novel long-term tracking framework that effectively combines complementary trackers with an online verification and decision policy.

Findings

Outperforms several existing long-term tracking methods.

Demonstrates robustness in various challenging scenarios.

Achieves competitive results on popular benchmarks.

Abstract

How to combine the complementary capabilities of an ensemble of different algorithms has been of central interest in visual object tracking. A significant progress on such a problem has been achieved, but considering short-term tracking scenarios. Instead, long-term tracking settings have been substantially ignored by the solutions. In this paper, we explicitly consider long-term tracking scenarios and provide a framework, named CoCoLoT, that combines the characteristics of complementary visual trackers to achieve enhanced long-term tracking performance. CoCoLoT perceives whether the trackers are following the target object through an online learned deep verification model, and accordingly activates a decision policy which selects the best performing tracker as well as it corrects the performance of the failing one. The proposed methodology is evaluated extensively and the comparison with several other solutions reveals that it competes favourably with the state-of-the-art on the most popular long-term visual tracking benchmarks.