期刊
INTERNET OF THINGS
卷 20, 期 -, 页码 -出版社
ELSEVIER
DOI: 10.1016/j.iot.2022.100633
关键词
Video person Re-ID; 3D convolution; Attention module
资金
- Collaborative Innovation Major Project of Zhengzhou [20XTZX06013]
With the popularization of surveillance cameras, the demand for video-based person re-identification (Re-ID) applications has increased. A novel framework called MSTN is proposed to utilize spatial-temporal features for accurate person re-identification. Extensive experiments on the MARS dataset demonstrate that MSTN achieves high identification accuracy, outperforming state-of-the-art methods.
With the popularization of surveillance cameras, public-safety related applications requiring the functionality of video-based person re-identification (Re-ID) thrive. Re-ID aims at accurately identifying a person-of-interest across video sequences from multiple cameras. Existing methods usually focus on either spatially salient regions, or temporal features among frames of fixed intervals (i.e., either short- or long-term temporal features), resulting in the under-utilization of neglected features and hence moderate identification accuracy. To achieve high Re-ID accuracy, we propose a novel framework termed Multi-granular Spatial-Temporal Network (MSTN), that facilitates full utilization of spatial-temporal features for video-based person Re-ID. Within MSTN, a Temporal Kernel Attention (TKA) module is proposed to adaptively capture both short- and long-term temporal relationships; a Feature Disentanglement Spatial Attention (FDSA) module is further proposed to mine spatially salient and subtle features. Extensive experiments on the MARS dataset demonstrate that MSTN can achieve high identification accuracy, exhibiting 86.1% in terms of mAP and 91.0% in terms of Rank-1, notably higher than state-of-the-art comparison schemes.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据