GUSOT: Green and Unsupervised Single Object Tracking for Long Video Sequences

TL;DR

GUSOT is a lightweight, unsupervised single-object tracker designed for long videos in resource-constrained environments, featuring novel modules for lost object recovery and shape proposal to improve long-term accuracy.

Contribution

It introduces GUSOT, a new unsupervised tracker with two modules that enhance long-term tracking performance while maintaining low computational and memory costs.

Findings

GUSOT outperforms baseline trackers on LaSOT dataset.

It achieves higher accuracy in long video sequences.

It is suitable for mobile and edge computing applications.

Abstract

Supervised and unsupervised deep trackers that rely on deep learning technologies are popular in recent years. Yet, they demand high computational complexity and a high memory cost. A green unsupervised single-object tracker, called GUSOT, that aims at object tracking for long videos under a resource-constrained environment is proposed in this work. Built upon a baseline tracker, UHP-SOT++, which works well for short-term tracking, GUSOT contains two additional new modules: 1) lost object recovery, and 2) color-saliency-based shape proposal. They help resolve the tracking loss problem and offer a more flexible object proposal, respectively. Thus, they enable GUSOT to achieve higher tracking accuracy in the long run. We conduct experiments on the large-scale dataset LaSOT with long video sequences, and show that GUSOT offers a lightweight high-performance tracking solution that finds applications in mobile and edge computing platforms.