Dynamic Tracking Aggregation with Transformers for RGB-T Tracking


Xiaohu Liu, Zhiyong Lei, Journal of Information Processing Systems Vol. 19, No. 1, pp. 80-88, Feb. 2023  

10.3745/JIPS.01.0092
Keywords: Cross-modal Fusion, Dynamic Tracking Aggregation, RGB-T Tracking, Transformers
Fulltext:

Abstract

RGB-thermal (RGB-T) tracking using unmanned aerial vehicles (UAVs) involves challenges with regards to the similarity of objects, occlusion, fast motion, and motion blur, among other issues. In this study, we propose dynamic tracking aggregation (DTA) as a unified framework to perform object detection and data association. The proposed approach obtains fused features based a transformer model and an L1-norm strategy. To link the current frame with recent information, a dynamically updated embedding called dynamic tracking identification (DTID) is used to model the iterative tracking process. For object association, we designed a long short-term tracking aggregation module for dynamic feature propagation to match spatial and temporal embeddings. DTA achieved a highly competitive performance in an experimental evaluation on public benchmark datasets.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from November 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.




Cite this article
[APA Style]
Liu, X. & Lei, Z. (2023). Dynamic Tracking Aggregation with Transformers for RGB-T Tracking. Journal of Information Processing Systems, 19(1), 80-88. DOI: 10.3745/JIPS.01.0092.

[IEEE Style]
X. Liu and Z. Lei, "Dynamic Tracking Aggregation with Transformers for RGB-T Tracking," Journal of Information Processing Systems, vol. 19, no. 1, pp. 80-88, 2023. DOI: 10.3745/JIPS.01.0092.

[ACM Style]
Xiaohu Liu and Zhiyong Lei. 2023. Dynamic Tracking Aggregation with Transformers for RGB-T Tracking. Journal of Information Processing Systems, 19, 1, (2023), 80-88. DOI: 10.3745/JIPS.01.0092.