Learning deep features with adaptive triplet  loss for person reidentification

Zhiqiang Li; Nong Sang; Kezhou Chen; Changxin Gao; Ruolin Wang

doi:10.1117/12.2283478

8 March 2018 Learning deep features with adaptive triplet loss for person reidentification

Zhiqiang Li, Nong Sang, Kezhou Chen, Changxin Gao, Ruolin Wang

Proceedings Volume 10609, MIPPR 2017: Pattern Recognition and Computer Vision; 106090G (2018) https://doi.org/10.1117/12.2283478
Event: Tenth International Symposium on Multispectral Image Processing and Pattern Recognition (MIPPR2017), 2017, Xiangyang, China

Abstract

Person reidentification (re-id) aims to match a specified person across non-overlapping cameras, which remains a very challenging problem. While previous methods mostly focus on feature extraction or metric learning, this paper makes the attempt in jointly learning both the global full-body and local body-parts features of the input persons with a multichannel convolutional neural network (CNN) model, which is trained by an adaptive triplet loss function that serves to minimize the distance between the same person and maximize the distance between different persons. The experimental results show that our approach achieves very promising results on the large-scale Market-1501 and DukeMTMC-reID datasets.

Citation Download Citation

Zhiqiang Li, Nong Sang, Kezhou Chen, Changxin Gao, and Ruolin Wang "Learning deep features with adaptive triplet loss for person reidentification", Proc. SPIE 10609, MIPPR 2017: Pattern Recognition and Computer Vision, 106090G (8 March 2018); https://doi.org/10.1117/12.2283478

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available