Cross-Modal Semantic Decoupling and Transfer for Text-to-Visible-Infrared Person Re-Identification

Text-to-Image Person Re-Identification (TI-ReID) retrieves visible pedestrian images using text queries. Yet in low-light or nighttime settings, visible images lack sufficient identity details, while infrared images effectively capture pedestrian contours and textures. To enable all-day surveillance, we propose a dual cross-modal retrieval task called Text-to-Visible-Infrared Re-Identification (TVI-ReID) and construct corresponding tri-modal datasets. Compared to TI-ReID, TVI-ReID faces two key challenges: (1) complex hybrid discrepancies in dual cross-modal retrieval from three modalities, and (2) semantic inconsistency between pretraining and downstream tasks. To address these issues, we propose a Cross-Modal Semantic Decoupling and Transfer (CSDT) framework. CSDT constructs color-related and color-irrelevant feature subspaces via Semantic Decoupling Learning (SDL) to align shared semantics across text and dual image modalities, reducing hybrid discrepancies. Moreover, Semantic Distribution Transfer (SDT) adapts pretrained text-visible alignment to text-infrared matching. Extensive experiments on tri-modal datasets show our approach outperforms existing state-of-the-art TI-ReID methods.

This work centers on all-day text-to-image person re-identification. Current methods for retrieving pedestrian images from textual descriptions perform poorly at night and in low-light environments, whereas infrared images can clearly capture pedestrians' contours and features. We explore whether we can develop a new retrieval framework applicable to both visible and infrared images for daytime and nighttime scenarios. To tackle this challenge, we build dedicated datasets and propose the CSDT framework. It achieves strong overall performance on both text-to-visible and text-to-infrared retrieval tasks, outperforming existing mainstream methods. We have released our code and datasets to facilitate further research by the community.