STA: Spatial-Temporal Attention for Large-Scale Video-based Person Re-Identification
This work addresses the problem of identifying individuals across video frames for surveillance and security applications, representing an incremental improvement over existing methods.
The authors tackled video-based person re-identification by proposing a Spatial-Temporal Attention (STA) approach to generate robust clip-level feature representations, achieving an mAP of 87.7% on the MARS benchmark, which outperforms state-of-the-art methods by over 11.6%.
In this work, we propose a novel Spatial-Temporal Attention (STA) approach to tackle the large-scale person re-identification task in videos. Different from the most existing methods, which simply compute representations of video clips using frame-level aggregation (e.g. average pooling), the proposed STA adopts a more effective way for producing robust clip-level feature representation. Concretely, our STA fully exploits those discriminative parts of one target person in both spatial and temporal dimensions, which results in a 2-D attention score matrix via inter-frame regularization to measure the importances of spatial parts across different frames. Thus, a more robust clip-level feature representation can be generated according to a weighted sum operation guided by the mined 2-D attention score matrix. In this way, the challenging cases for video-based person re-identification such as pose variation and partial occlusion can be well tackled by the STA. We conduct extensive experiments on two large-scale benchmarks, i.e. MARS and DukeMTMC-VideoReID. In particular, the mAP reaches 87.7% on MARS, which significantly outperforms the state-of-the-arts with a large margin of more than 11.6%.