Abstract
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.
Abstract (translated)
在各种监控场景中，特别是在遮挡发生的情况下，处理人员重新识别（ReID）的复杂性带来了很多挑战。我们提出了一种创新的运动感知融合（MOTAR-FUSE）网络，该网络利用从静态图像中提取的运动线索来显著增强ReID的能力。此网络包括一个双输入视觉适配器，能够同时处理图像和视频，从而实现更有效的特征提取。我们的方法的独特之处在于整合了一个运动一致性任务，这使得运动感知变压器能够有效地捕捉人类动作的动力学特性。这种方法大大提高了在遮挡普遍存在的情况下识别特征的能力，从而推进了ReID过程的进步。我们在多个ReID基准测试上进行了全面的评估，包括整体、遮挡和基于视频的场景，结果显示我们的MOTAR-FUSE网络相比现有方法表现出色。
URL
https://arxiv.org/abs/2502.00665
