Spatial temporal and channel aware network for video-based person re-identification

作者：

Highlights：

• A feature aggregation module, CSGS, is proposed to capture temporal and channel relations.

• An attention mechanism, SAE, is introduced to enhance spatial dependencies.

• A Fusion mechanism is proposed to combine the output of SAE and CSGS

摘要

•A feature aggregation module, CSGS, is proposed to capture temporal and channel relations.•An attention mechanism, SAE, is introduced to enhance spatial dependencies.•A Fusion mechanism is proposed to combine the output of SAE and CSGS

论文关键词：Video-based Re-ID,Spatial temporal feature,Channel segmentation,Group shuffle convolution

论文评审过程：Received 9 July 2021, Revised 9 December 2021, Accepted 11 December 2021, Available online 16 December 2021, Version of Record 24 December 2021.

论文官网地址：https://doi.org/10.1016/j.imavis.2021.104356