SiamTHN: Siamese Target Highlight Network for Visual Tracking

TL;DR

SiamTHN introduces a novel Siamese network-based visual tracker with a Target Highlight Module and corrective loss, improving focus on targets and aligning classification and regression branches for more accurate and efficient tracking.

Contribution

The paper proposes a Target Highlight Module and a corrective loss to enhance focus and alignment in Siamese trackers, leading to improved accuracy and efficiency.

Findings

Outperforms current models on 5 benchmark datasets.

Operates at 38 fps, demonstrating efficiency.

Achieves more accurate tracking by focusing on target regions.

Abstract

Siamese network based trackers develop rapidly in the field of visual object tracking in recent years. The majority of siamese network based trackers now in use treat each channel in the feature maps generated by the backbone network equally, making the similarity response map sensitive to background influence and hence challenging to focus on the target region. Additionally, there are no structural links between the classification and regression branches in these trackers, and the two branches are optimized separately during training. Therefore, there is a misalignment between the classification and regression branches, which results in less accurate tracking results. In this paper, a Target Highlight Module is proposed to help the generated similarity response maps to be more focused on the target region. To reduce the misalignment and produce more precise tracking results, we propose a corrective loss to train the model. The two branches of the model are jointly tuned with the use of corrective loss to produce more reliable prediction results. Experiments on 5 challenging benchmark datasets reveal that the method outperforms current models in terms of performance, and runs at 38 fps, proving its effectiveness and efficiency.