Abstract | ||
---|---|---|
In this work, we tackle the problem of person search, which is a challenging task consisted of pedestrian detection and person re-identification (re-ID). Instead of sharing representations in a single joint model, we find that separating detector and re-ID feature extraction yields better performance. In order to extract more representative features for each identity, we segment out the foreground person from the original image patch. We propose a simple yet effective re-ID method, which models foreground person and original image patches individually, and obtains enriched representations from two separate CNN streams. We also propose a Confidence Weighted Stream Attention method which further re-adjusts the relative importance of the two streams by incorporating the detection confidence. Furthermore, we simplify the whole pipeline by incorporating semantic segmentation into the re-ID network, which is trained by bounding boxes as weakly-annotated masks and identification labels simultaneously. From the experiments on two standard person search benchmarks i.e. CUHK-SYSU and PRW, we achieve mAP of 83.3% and 32.8% respectively, surpassing the state of the art by a large margin. The extensive ablation study and model inspection further justifies our motivation. |
Year | DOI | Venue |
---|---|---|
2020 | 10.1109/TIP.2020.2973513 | IEEE TRANSACTIONS ON IMAGE PROCESSING |
Keywords | DocType | Volume |
Feature extraction, Streaming media, Task analysis, Image segmentation, Training, Detectors, Search problems, Person search, pedestrian detection, person re-identification | Journal | 29 |
Issue | ISSN | Citations |
1 | 1057-7149 | 0 |
PageRank | References | Authors |
0.34 | 23 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Di Chen | 1 | 21 | 4.07 |
Shanshan Zhang | 2 | 231 | 21.21 |
Wanli Ouyang | 3 | 2371 | 105.17 |
Jian Yang | 4 | 6102 | 339.77 |
Ying Tai | 5 | 213 | 25.74 |