Learning Occlusion-Robust Pedestrian Representations via Uncertainty-Guided Feature Pruning
Frontiers in Applied Physics and Mathematics · 24.02.2026 · 3 (1) · 100-108
DOI: 10.71465/fapm658 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Applied Physics and Mathematics, online erschienen am 24.02.2026. Verfasst von Lukas M. Schneider und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Occlusion and background interference remain major challenges for pedestrian re-identification in urban traffic environments. Inspired by uncertainty-aware CLIP-based frameworks, this paper introduces an uncertainty-guided feature selection mechanism that adjusts the contribution of local visual regions and semantic cues according to their estimated reliability. The proposed method is evaluated on two autonomous driving datasets with both real-world and synthetic occlusion patterns, covering occlusion ratios from 20% to 60%. Comparisons are conducted against attention-based and part-based ReID methods, including PCB, OSNet, and transformer-based attention models. The proposed approach achieves mAP improvements ranging from 4.5% to 6.2% under severe occlusion conditions, while maintaining comparable performance in fully visible settings.
Zitieren
APA
Schneider, L. M., Vogel, A. K., & Weber, T. R. (2026). Learning Occlusion-Robust Pedestrian Representations via Uncertainty-Guided Feature Pruning. Frontiers in Applied Physics and Mathematics, 3(1), 100-108. https://doi.org/10.71465/fapm658
Vancouver
Schneider LM, Vogel AK, Weber TR. Learning Occlusion-Robust Pedestrian Representations via Uncertainty-Guided Feature Pruning. Frontiers in Applied Physics and Mathematics. 2026;3(1):100-108. doi:10.71465/fapm658
BibTeX
@article{Schneider2026,
author = {Schneider, Lukas M. and Vogel, Anna K. and Weber, Tobias R.},
title = {Learning Occlusion-Robust Pedestrian Representations via Uncertainty-Guided Feature Pruning},
journal = {Frontiers in Applied Physics and Mathematics},
year = {2026},
volume = {3},
number = {1},
pages = {100--108},
doi = {10.71465/fapm658}
}
EUVIMED