SMILEtrack: SiMIlarity LEarning for Occlusion-Aware Multiple Object Tracking
Proceedings of the AAAI Conference on Artificial Intelligence · 24.03.2024 · 38 (6) · 5740-5748
DOI: 10.1609/aaai.v38i6.28386 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 24.03.2024. Verfasst von Yu-Hsiang Wang und 5 weiteren Autor:innen. Die Arbeit wurde bisher 88-mal zitiert (Crossref).
Abstract (Originalsprache)
Despite recent progress in Multiple Object Tracking (MOT), several obstacles such as occlusions, similar objects, and complex scenes remain an open challenge. Meanwhile, a systematic study of the cost-performance tradeoff for the popular tracking-by-detection paradigm is still lacking. This paper introduces SMILEtrack, an innovative object tracker that effectively addresses these challenges by integrating an efficient object detector with a Siamese network-based Similarity Learning Module (SLM). The technical contributions of SMILETrack are twofold. First, we propose an SLM that calculates the appearance similarity between two objects, overcoming the limitations of feature descriptors in Separate Detection and Embedding (SDE) models. The SLM incorporates a Patch Self-Attention (PSA) block inspired by the vision Transformer, which generates reliable features for accurate similarity matching. Second, we develop a Similarity Matching Cascade (SMC) module with a novel GATE function for robust object matching across consecutive video frames, further enhancing MOT performance. Together, these innovations help SMILETrack achieve an improved trade-off between the cost (e.g., running speed) and performance (e.g., tracking accuracy) over several existing state-of-the-art benchmarks, including the popular BYTETrack method. SMILETrack outperforms BYTETrack by 0.4-0.8 MOTA and 2.1-2.2 HOTA points on MOT17 and MOT20 datasets. Code is available at http://github.com/pingyang1117/SMILEtrack_official.
Zitieren
APA
Wang, Y., Hsieh, J., Chen, P., Chang, M., So, H., & Li, X. (2024). SMILEtrack: SiMIlarity LEarning for Occlusion-Aware Multiple Object Tracking. Proceedings of the AAAI Conference on Artificial Intelligence, 38(6), 5740-5748. https://doi.org/10.1609/aaai.v38i6.28386
Vancouver
Wang Y, Hsieh J, Chen P, Chang M, So H, Li X. SMILEtrack: SiMIlarity LEarning for Occlusion-Aware Multiple Object Tracking. Proceedings of the AAAI Conference on Artificial Intelligence. 2024;38(6):5740-5748. doi:10.1609/aaai.v38i6.28386
BibTeX
@article{Wang2024,
author = {Wang, Yu-Hsiang and Hsieh, Jun-Wei and Chen, Ping-Yang and Chang, Ming-Ching and So, Hung-Hin and Li, Xin},
title = {SMILEtrack: SiMIlarity LEarning for Occlusion-Aware Multiple Object Tracking},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2024},
volume = {38},
number = {6},
pages = {5740--5748},
doi = {10.1609/aaai.v38i6.28386}
}
EUVIMED