EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

DW-ReID: Vision–Language Learning for Person Re-Identification Under Diverse Weather Conditions

Lei Cai iD, Yuying Liang iD, Bin Wang iD, Hexi Li, Jinquan Yang, Tao Zhu

Sensors · 06.04.2026 · 26 (7) · 2263

DOI: 10.3390/s26072263 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 06.04.2026. Verfasst von Lei Cai und 5 weiteren Autor:innen. Beteiligte Einrichtung: College of Engineering, Huaqiao University, Quanzhou 362021, China. Die Arbeit nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of Fujian Province, Natural Science Foundation of Xiamen, China, Scientific Research Funds of Huaqiao University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Person re-identification (ReID) under diverse weather conditions remains a critical yet insufficiently explored problem. Most existing ReID approaches are developed and benchmarked on clear-weather datasets, resulting in significant performance degradation when deployed in rainy, snowy, or hazy environments. Conventional image restoration methods, typically optimized for low-level image quality metrics, are often misaligned with the objectives of high-level identity discrimination and thus fail to improve the person ReID performance. To address these limitations, we propose DW-ReID, a unified framework that integrates weather-degraded image restoration with person re-identification tasks. The proposed DW-ReID is built upon a large-scale Contrastive Language-Image Pre-training (CLIP) model and achieved by a two-stage training paradigm. In the first stage, a set of learnable text prompts is optimized to construct identity-specific ambiguous descriptions for each person’s identity. In the second stage, the optimized text descriptions, together with a frozen text encoder, provide language supervision to jointly train a weather encoder, an image restorer, and a ReID encoder in an end-to-end manner. The experimental results on two our contributed synthetic datasets consistently demonstrate the effectiveness and superior performance of the proposed DW-ReID method.

Schlagwörter

Stichwörter der Autor:innen

adverse weatherperson re-identificationvision–language learning

Autor:innen und Institutionen

  • College of Engineering, Huaqiao University, Quanzhou 362021, China.
  • School of Computing, Guangzhou College of Applied Science and Technology, Zhaoqing 526072, China.
  • School of Computer Science and Cyber Engineering, Guangzhou University, Guangzhou 510006, China.
  • School of Electronic Engineering and Automation, Guilin University of Electronic Technology, Guilin 541004, China.

ORCID-Profile: Lei Cai, Yuying Liang, Bin Wang

Förderung

  • Natural Science Foundation of Fujian Province
  • Natural Science Foundation of Xiamen, China
  • Scientific Research Funds of Huaqiao University
  • National Natural Science Foundation of China
  • Natural Science Foundation of Guangxi Province

Zitieren

APA

Cai, L., Liang, Y., Wang, B., Li, H., Yang, J., & Zhu, T. (2026). DW-ReID: Vision–Language Learning for Person Re-Identification Under Diverse Weather Conditions. Sensors, 26(7), 2263. https://doi.org/10.3390/s26072263

Vancouver

Cai L, Liang Y, Wang B, Li H, Yang J, Zhu T. DW-ReID: Vision–Language Learning for Person Re-Identification Under Diverse Weather Conditions. Sensors. 2026;26(7):2263. doi:10.3390/s26072263

BibTeX
@article{Cai2026,
  author = {Cai, Lei and Liang, Yuying and Wang, Bin and Li, Hexi and Yang, Jinquan and Zhu, Tao},
  title = {DW-ReID: Vision–Language Learning for Person Re-Identification Under Diverse Weather Conditions},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {7},
  pages = {2263},
  doi = {10.3390/s26072263}
}
Kostenlos starten →