An effective feature representation called Local Maximal Occurrence (LOMO), and a subspace and metric learning method called Cross-view Quadratic Discriminant Analysis (XQDA) are proposed for person re-identification. The LOMO feature analyzes the horizontal occurrence of local features, and maximizes the occurrence to make a stable representation against viewpoint changes.
Besides, to handle illumination variations, we apply the Retinex transform and the Scale Invariant Local Ternary Pattern (SILTP). To learn a discriminant metric, we propose to learn a discriminant low dimensional subspace by cross-view quadratic discriminant analysis, and simultaneously, a QDA metric is learned on the derived subspace. We also present a practical computation method for XQDA, as well as its regularization. Experiments on four challenging person re-identification databases, VIPeR, QMUL GRID, CUHK Campus, and CUHK03, show that the proposed method improves the state-of-the-art rank-1 identification rates by 2.2\%, 4.88\%, 28.91\%, and 31.55\% on the four databases, respectively.
Download:
- Source code: LOMO_XQDA.zip
- Extracted features: lomo_viper+grid.zip cuhk01_lomo.mat cuhk03_labeled_lomo.mat cuhk03_detected_lomo.mat
Results:
CMC curves on the VIPeR database
CMC curves on the CUHK01 (CUHK Campus) database
Table 1. Summary of results (%) for the proposed LOMO+XQDA algorithm
Database | Rank 1 | Rank 5 | Rank 10 | Rank 15 | Rank 20 |
VIPeR | 40.00 | 68.13 | 80.51 | 87.37 | 91.08 |
QMUL Grid | 16.56 | 33.84 | 41.84 | 47.68 | 52.40 |
CUHK01 | 63.21 | 83.89 | 90.04 | 92.59 | 94.16 |
CUHK03 Labeled | 52.20 | 82.23 | 92.14 | 94.74 | 96.25 |
CUHK03 Detected | 46.25 | 78.90 | 88.55 | 92.30 | 94.25 |
Note: the source code package contains the CMC curves for performance plot.
Contact:
National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences.
References:
[1]Shengcai Liao, Yang Hu, Xiangyu Zhu, and Stan Z. Li, “Person Re-identification by Local Maximal Occurrence Representation and Metric Learning.” In IEEE International Conference on Computer Vision and Pattern Recognition, June 7-12, Boston, Massachusetts, USA, 2015. [pdf]
Last updated: May 7, 2015