Deep Attention Aware Feature Learning for Person Re-Identification

1 Mar 2020  ·  Yifan Chen, Han Wang, Xiaolu Sun, Bin Fan, Chu Tang ·

Visual attention has proven to be effective in improving the performance of person re-identification. Most existing methods apply visual attention heuristically by learning an additional attention map to re-weight the feature maps for person re-identification. However, this kind of methods inevitably increase the model complexity and inference time. In this paper, we propose to incorporate the attention learning as additional objectives in a person ReID network without changing the original structure, thus maintain the same inference time and model size. Two kinds of attentions have been considered to make the learned feature maps being aware of the person and related body parts respectively. Globally, a holistic attention branch (HAB) makes the feature maps obtained by backbone focus on persons so as to alleviate the influence of background. Locally, a partial attention branch (PAB) makes the extracted features be decoupled into several groups and be separately responsible for different body parts (i.e., keypoints), thus increasing the robustness to pose variation and partial occlusion. These two kinds of attentions are universal and can be incorporated into existing ReID networks. We have tested its performance on two typical networks (TriNet and Bag of Tricks) and observed significant performance improvement on five widely used datasets.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Person Re-Identification DukeMTMC-reID DAAF-BoT(RK) Rank-1 91.7 # 11
mAP 89.6 # 10
Person Re-Identification DukeMTMC-reID DAAF-BoT Rank-1 87.9 # 41
mAP 77.9 # 45
Person Re-Identification Market-1501 DAAF-BoT(RK) Rank-1 96.4 # 9
mAP 95 # 6
Person Re-Identification Market-1501 DAAF-BoT Rank-1 95.1 # 46
mAP 87.9 # 48


No methods listed for this paper. Add relevant methods here