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

Ziang Zhang, Bin Yang, Mang Ye
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:159954-159967, 2026.

Abstract

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-zhang26hs, title = {Cross-Modal Semantic Decoupling and Transfer for Text-to-Visible-Infrared Person Re-Identification}, author = {Zhang, Ziang and Yang, Bin and Ye, Mang}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {159954--159967}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/zhang26hs/zhang26hs.pdf}, url = {https://proceedings.mlr.press/v306/zhang26hs.html}, abstract = {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.} }
Endnote
%0 Conference Paper %T Cross-Modal Semantic Decoupling and Transfer for Text-to-Visible-Infrared Person Re-Identification %A Ziang Zhang %A Bin Yang %A Mang Ye %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-zhang26hs %I PMLR %P 159954--159967 %U https://proceedings.mlr.press/v306/zhang26hs.html %V 306 %X 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.
APA
Zhang, Z., Yang, B. & Ye, M.. (2026). Cross-Modal Semantic Decoupling and Transfer for Text-to-Visible-Infrared Person Re-Identification. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:159954-159967 Available from https://proceedings.mlr.press/v306/zhang26hs.html.

Related Material