Abstract | ||
---|---|---|
This paper addresses the problem of handling spatial misalignments due to camera-view changes or human-pose variations in person re-identification. We first introduce a boosting-based approach to learn a correspondence structure, which indicates the patchwise matching probabilities between images from a target camera pair. The learned correspondence structure can not only capture the spatial correspondence pattern between cameras but also handle the viewpoint or human-pose variation in individual images. We further introduce a global constraint-based matching process. It integrates a global matching constraint over the learned correspondence structure to exclude cross-view misalignments during the image patch matching process, hence achieving a more reliable matching score between images. Finally, we also extend our approach by introducing a multi-structure scheme, which learns a set of local correspondence structures to capture the spatial correspondence sub-patterns between a camera pair, so as to handle the spatial misalignments between individual images in a more precise way. Experimental results on various data sets demonstrate the effectiveness of our approach. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1109/TIP.2017.2683063 | IEEE Trans. Image Processing |
Keywords | DocType | Volume |
Cameras,Probes,Electronic mail,Correlation,Reliability,Pattern matching,Visualization | Journal | abs/1703.06931 |
Issue | ISSN | Citations |
5 | 1057-7149 | 17 |
PageRank | References | Authors |
0.58 | 39 | 7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Weiyao Lin | 1 | 732 | 68.05 |
Yang Shen | 2 | 75 | 2.52 |
Junchi Yan | 3 | 891 | 83.36 |
Mingliang Xu | 4 | 372 | 54.07 |
Jianxin Wu | 5 | 3276 | 154.17 |
Jingdong Wang | 6 | 4198 | 156.76 |
Ke Lu | 7 | 953 | 53.36 |