Updatable Siamese Tracker with Two-stage One-shot Learning

TL;DR

This paper introduces SiamTOL, an updatable Siamese tracker with a two-stage one-shot learner that enhances online adaptation and achieves superior performance on multiple benchmarks.

Contribution

It proposes a novel two-stage one-shot learning approach integrated into an updatable Siamese network for improved online tracking.

Findings

Achieves leading performance on OTB100, VOT2018, VOT2019, LaSOT, UAV123, and GOT10k.

Outperforms state-of-the-art trackers in accuracy and robustness.

Effectively handles complex scenes with online update capability.

Abstract

Offline Siamese networks have achieved very promising tracking performance, especially in accuracy and efficiency. However, they often fail to track an object in complex scenes due to the incapacity in online update. Traditional updaters are difficult to process the irregular variations and sampling noises of objects, so it is quite risky to adopt them to update Siamese networks. In this paper, we first present a two-stage one-shot learner, which can predict the local parameters of primary classifier with object samples from diverse stages. Then, an updatable Siamese network is proposed based on the learner (SiamTOL), which is able to complement online update by itself. Concretely, we introduce an extra inputting branch to sequentially capture the latest object features, and design a residual module to update the initial exemplar using these features. Besides, an effective multi-aspect training loss is designed for our network to avoid overfit. Extensive experimental results on several popular benchmarks including OTB100, VOT2018, VOT2019, LaSOT, UAV123 and GOT10k manifest that the proposed tracker achieves the leading performance and outperforms other state-of-the-art methods