XiDian News (Reporter Jianan Liu, Qiang Jiao) Recently, invited by Professor Salvador Garcia, Editor in Chief of Information Fusion, Professor Qiang Zhang's team, School of Mechano-Electronic Engineering at XiDian University and Professor Jungong Han’s team, Computer Science Department at Aberystwyth University, published a review paper ‘Deep Learning for Visible-Infrared Cross-modality Person Re-Identification: A comprehensive Review’ in Information Fusion (IF:17.564), which is the No.1 journal in the field of information fusion. The first author's affiliation of this paper is the School of Mechano-Electronic Engineering, Xidian University, Xi'an, China. Professor Qiang Zhang and his Ph.D. student Nianchang Huang are the corresponding author and the first author, respectively.
Cross-modality Person Re-Identification aims to match the query images of one modality from the gallery set with images of another modality, e.g. visible-to-infrared image matching and infrared-to-visible image matching, which has various applications in smart surveillance systems and intelligent security systems. Recently, with the construction of smart cities, there are increasing demands for surveillance and security. Accordingly, as one of the key techniques in this field, cross-modality person re-identification has gained considerable attention in recent years and has seen increasingly rapid advances in computer vision. This review provides a comprehensive and detailed review for cross-modality person re-identification, including (1) introducing its definition, challenges, datasets, evaluation metrics, etc.; (2) systematically categorizing the existing cross-modality person re-identification approaches according to their motivations and methodologies; 3) evaluating and comparing the performance of different methods; 4) providing insights for future research directions. Furthermore, this review provides a comprehensive and systematic introduction to relevant researchers and plays a positive role in the development of this task. This review also reveals that the achievements of Professor Qiang Zhang's Team and Professor Jungong Han’s team are impactful and have attracted extensive attention in the academic community.
Figure 1. Illustration of the basic framework of cross-modality person re-identification models.
Figure 2. Taxonomy of cross-modality person re-identification models.
Professor Qiang Zhang's team mainly focuses on the study of computer vision and has achieved a series of achievements in multi-modal image processing and cross-modality person re-Identification. This team has published more than 30 papers in leading journals and conferences, including IEEE TPAMI, TIP, TMM, Information Fusion, Pattern Recognition, ICCV, CVPR, and so on.
