12 May 2017 Ring-push metric learning for person reidentification
Author Affiliations +
Abstract
Person reidentification (re-id) has been widely studied because of its extensive use in video surveillance and forensics applications. It aims to search a specific person among a nonoverlapping camera network, which is highly challenging due to large variations in the cluttered background, human pose, and camera viewpoint. We present a metric learning algorithm for learning a Mahalanobis distance for re-id. Generally speaking, there exist two forces in the conventional metric learning process, one pulling force that pulls points of the same class closer and the other pushing force that pushes points of different classes as far apart as possible. We argue that, when only a limited number of training data are given, forcing interclass distances to be as large as possible may drive the metric to overfit the uninformative part of the images, such as noises and backgrounds. To alleviate overfitting, we propose the ring-push metric learning algorithm. Different from other metric learning methods that only punish too small interclass distances, in the proposed method, both too small and too large inter-class distances are punished. By introducing the generalized logistic function as the loss, we formulate the ring-push metric learning as a convex optimization problem and utilize the projected gradient descent method to solve it. The experimental results on four public datasets demonstrate the effectiveness of the proposed algorithm.
© 2017 SPIE and IS&T
Botao He, Shaohua Yu, "Ring-push metric learning for person reidentification," Journal of Electronic Imaging 26(3), 033005 (12 May 2017). https://doi.org/10.1117/1.JEI.26.3.033005 . Submission: Received: 7 December 2016; Accepted: 24 April 2017
Received: 7 December 2016; Accepted: 24 April 2017; Published: 12 May 2017
JOURNAL ARTICLE
10 PAGES


SHARE
RELATED CONTENT

Activity-based exploitation of Full Motion Video (FMV)
Proceedings of SPIE (May 25 2012)
Forensic video image analysis
Proceedings of SPIE (February 10 1997)
FIPE: a forensic image-processing environment
Proceedings of SPIE (February 04 1999)
Monitoring image quality for security applications
Proceedings of SPIE (January 24 2011)
Video coding for next-generation surveillance systems
Proceedings of SPIE (February 19 1997)
New approach to obtain height measurements from video
Proceedings of SPIE (February 04 1999)

Back to Top