RGB-T Tracking With Template-Bridged Search Interaction and Target-Preserved Template Updating
IEEE Transactions on Pattern Analysis and Machine Intelligence · 2025-01 · 47 (1) · 634-649
DOI: 10.1109/tpami.2024.3475472 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2025-01. Verfasst von Bo Li und 7 weiteren Autor:innen. Die Arbeit wurde bisher 24-mal zitiert (Crossref) und nennt 73 Literaturangaben. Gefördert von: Zhejiang Provincial Natural Science Foundation of China, Key Research and Development Program of Zhejiang Province.
Abstract (Originalsprache)
The goal of RGB-Thermal (RGB-T) tracking is to utilize the synergistic and complementary strengths of RGB and TIR modalities to enhance tracking in diverse situations, with cross-modal interaction being a crucial element. Earlier methods often simply combine the features of the RGB and TIR search frames, leading to a coarse interaction that also introduced unnecessary background noise. Many other approaches sample candidate boxes from search frames and apply different fusion techniques to individual pairs of RGB and TIR boxes, which confines cross-modal interactions to local areas and results in insufficient context modeling. Additionally, mining video temporal contexts is also under-explored in RGB-T tracking. To alleviate these limitations, we propose a novel Template-Bridged Search region Interaction (TBSI) module that exploits templates as the medium to bridge the cross-modal interaction between RGB and TIR search regions by gathering and distributing target-relevant object and environment contexts. An Illumination Guided Fusion (IGF) module is designed to adaptively fuse RGB and TIR search region tokens with a global illumination factor. Furthermore, in the inference stage, we also propose an efficient Target-Preserved Template Updating (TPTU) strategy, leveraging the temporal context within video sequences to accommodate the target's appearance change. Our proposed modules are integrated into a ViT backbone for joint feature extraction, search-template matching, and cross-modal interaction. Extensive experiments on three popular RGB-T tracking benchmarks demonstrate our method achieves new state-of-the-art performances.
Förderung
- Zhejiang Provincial Natural Science Foundation of China
- Key Research and Development Program of Zhejiang Province
Zitieren
APA
Li, B., Peng, F., Hui, T., Wei, X., Wei, X., Zhang, L., Shi, H., & Liu, S. (2025). RGB-T Tracking With Template-Bridged Search Interaction and Target-Preserved Template Updating. IEEE Transactions on Pattern Analysis and Machine Intelligence, 47(1), 634-649. https://doi.org/10.1109/tpami.2024.3475472
Vancouver
Li B, Peng F, Hui T, Wei X, Wei X, Zhang L, et al. RGB-T Tracking With Template-Bridged Search Interaction and Target-Preserved Template Updating. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2025;47(1):634-649. doi:10.1109/tpami.2024.3475472
BibTeX
@article{Li2025,
author = {Li, Bo and Peng, Fengguang and Hui, Tianrui and Wei, Xiaoming and Wei, Xiaolin and Zhang, Lijun and Shi, Hang and Liu, Si},
title = {RGB-T Tracking With Template-Bridged Search Interaction and Target-Preserved Template Updating},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2025},
volume = {47},
number = {1},
pages = {634--649},
doi = {10.1109/tpami.2024.3475472}
}
EUVIMED