Person re-identification handles pedestrian matching and ranking across non-overlapping camera views. It has many important applications in video surveillance by saving a lot of human efforts on exhaustively searching for a person from large amounts of video sequences.
We focus on feature design, attribute and salience learning, metric learning, and manifold modeling.
Highlights
Locally Aligned Feature Transforms across Views
W. Li and X. Wang. In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (Poster, CVPR 2013)
PDF
Unsupervised Salience Learning for Person Re-identification
R. Zhao, W. Ouyang and X. Wang. In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (Poster, CVPR 2013)
PDF Project Page
Person Re-Identification: What Features are Important?
C. Liu, S. Gong, C. C. Loy, and X. Lin. In Proceedings of European Conference on Computer Vision, International Workshop on Re-Identification (Poster, ECCV 2013)
PDF Data partition and CMC curves
Human Reidentification with Transferred Metric Learning
W. Li, R. Zhao and X. Wang. In Proceedings of Asian Conference on Computer Vision (Poster, ACCV 2012)
PDF
Shape and Appearance Context Modeling
X. Wang, G. Doretto, T. Sebastian, J. Rittscher, and P. Tu. in Proceedings of IEEE International Conference on Computer Vision (ICCV) 2007.
PDF