Saved in:
Bibliographic Details
Main Authors: Ling, Fuxi, Liu, Hongye, Huang, Guoqiang, Li, Jing, Wu, Hong, Tang, Zhihao
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.00665
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915133118218240
author Ling, Fuxi
Liu, Hongye
Huang, Guoqiang
Li, Jing
Wu, Hong
Tang, Zhihao
author_facet Ling, Fuxi
Liu, Hongye
Huang, Guoqiang
Li, Jing
Wu, Hong
Tang, Zhihao
contents Navigating the complexities of person re-identification (ReID) in varied surveillance scenarios, particularly when occlusions occur, poses significant challenges. We introduce an innovative Motion-Aware Fusion (MOTAR-FUSE) network that utilizes motion cues derived from static imagery to significantly enhance ReID capabilities. This network incorporates a dual-input visual adapter capable of processing both images and videos, thereby facilitating more effective feature extraction. A unique aspect of our approach is the integration of a motion consistency task, which empowers the motion-aware transformer to adeptly capture the dynamics of human motion. This technique substantially improves the recognition of features in scenarios where occlusions are prevalent, thereby advancing the ReID process. Our comprehensive evaluations across multiple ReID benchmarks, including holistic, occluded, and video-based scenarios, demonstrate that our MOTAR-FUSE network achieves superior performance compared to existing approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2502_00665
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Cross-Modal Synergies: Unveiling the Potential of Motion-Aware Fusion Networks in Handling Dynamic and Static ReID Scenarios
Ling, Fuxi
Liu, Hongye
Huang, Guoqiang
Li, Jing
Wu, Hong
Tang, Zhihao
Computer Vision and Pattern Recognition
Navigating the complexities of person re-identification (ReID) in varied surveillance scenarios, particularly when occlusions occur, poses significant challenges. We introduce an innovative Motion-Aware Fusion (MOTAR-FUSE) network that utilizes motion cues derived from static imagery to significantly enhance ReID capabilities. This network incorporates a dual-input visual adapter capable of processing both images and videos, thereby facilitating more effective feature extraction. A unique aspect of our approach is the integration of a motion consistency task, which empowers the motion-aware transformer to adeptly capture the dynamics of human motion. This technique substantially improves the recognition of features in scenarios where occlusions are prevalent, thereby advancing the ReID process. Our comprehensive evaluations across multiple ReID benchmarks, including holistic, occluded, and video-based scenarios, demonstrate that our MOTAR-FUSE network achieves superior performance compared to existing approaches.
title Cross-Modal Synergies: Unveiling the Potential of Motion-Aware Fusion Networks in Handling Dynamic and Static ReID Scenarios
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2502.00665