SiamReID: Confuser Aware Siamese Tracker with Re-identification Feature

TL;DR

SiamReID enhances Siamese deep-network trackers by integrating a re-identification framework that effectively rejects confusers during prolonged occlusions, significantly improving aerial object tracking performance.

Contribution

The paper introduces SiamReID, a novel re-identification framework for Siamese trackers that addresses confuser issues during occlusions, tailored for aerial imagery.

Findings

Achieves state-of-the-art results on UAVDT benchmark.

Effectively rejects confusers during prolonged occlusions.

Improves robustness of Siamese trackers in aerial scenarios.

Abstract

Siamese deep-network trackers have received significant attention in recent years due to their real-time speed and state-of-the-art performance. However, Siamese trackers suffer from similar looking confusers, that are prevalent in aerial imagery and create challenging conditions due to prolonged occlusions where the tracker object re-appears under different pose and illumination. Our work proposes SiamReID, a novel re-identification framework for Siamese trackers, that incorporates confuser rejection during prolonged occlusions and is well-suited for aerial tracking. The re-identification feature is trained using both triplet loss and a class balanced loss. Our approach achieves state-of-the-art performance in the UAVDT single object tracking benchmark.