Unsupervised Green Object Tracker (GOT) without Offline Pre-training
This work addresses the need for lightweight, high-performance tracking on mobile and edge devices, offering an incremental improvement over existing unsupervised methods by reducing complexity without offline pre-training.
The paper tackles the problem of high computational complexity and labeling costs in single object tracking by proposing an unsupervised Green Object Tracker (GOT) that achieves competitive accuracy with state-of-the-art unsupervised trackers at a lower computation cost, featuring a tiny model size (<3k parameters) and low inference complexity (around 58M FLOPs per frame).
Supervised trackers trained on labeled data dominate the single object tracking field for superior tracking accuracy. The labeling cost and the huge computational complexity hinder their applications on edge devices. Unsupervised learning methods have also been investigated to reduce the labeling cost but their complexity remains high. Aiming at lightweight high-performance tracking, feasibility without offline pre-training, and algorithmic transparency, we propose a new single object tracking method, called the green object tracker (GOT), in this work. GOT conducts an ensemble of three prediction branches for robust box tracking: 1) a global object-based correlator to predict the object location roughly, 2) a local patch-based correlator to build temporal correlations of small spatial units, and 3) a superpixel-based segmentator to exploit the spatial information of the target frame. GOT offers competitive tracking accuracy with state-of-the-art unsupervised trackers, which demand heavy offline pre-training, at a lower computation cost. GOT has a tiny model size (<3k parameters) and low inference complexity (around 58M FLOPs per frame). Since its inference complexity is between 0.1%-10% of DL trackers, it can be easily deployed on mobile and edge devices.