Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Dual reverse attention networks for person re-identification

Abstract: 

In this paper, we enhance feature representation ability of person re-identification (Re-ID) by learning invariances to hard examples. Unlike previous works of hard examples mining and generating in image level, we propose a dual reverse attention network (DRANet) to generate hard examples in the convolutional feature space. Specifically, we use a classification branch of attention mechanism to model that ‘what’ in channel and ‘where’ in spatial dimensions are informative in the feature maps. Meanwhile, we introduce two branches of reverse attention modules in parallel way, which convert informative feature maps into hard examples of uninformative ones. In the proposed framework, both classification and dual reverse attention branches are learned in a joint way. Experimental results on three mainstream datasets demonstrate the efficacy of the proposed method.

up
0 users have voted:

Paper Details

Authors:
Shuangwei Liu, Lin Qi, Yunzhou Zhang, Weidong Shi
Submitted On:
20 September 2019 - 11:04am
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Shuangwei Liu
Paper Code:
1553
Document Year:
2019
Cite

Document Files

Dual Reverse Attention Networks for Person Re-id.pdf

(8)

Subscribe

[1] Shuangwei Liu, Lin Qi, Yunzhou Zhang, Weidong Shi, "Dual reverse attention networks for person re-identification", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4784. Accessed: Oct. 18, 2019.
@article{4784-19,
url = {http://sigport.org/4784},
author = {Shuangwei Liu; Lin Qi; Yunzhou Zhang; Weidong Shi },
publisher = {IEEE SigPort},
title = {Dual reverse attention networks for person re-identification},
year = {2019} }
TY - EJOUR
T1 - Dual reverse attention networks for person re-identification
AU - Shuangwei Liu; Lin Qi; Yunzhou Zhang; Weidong Shi
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4784
ER -
Shuangwei Liu, Lin Qi, Yunzhou Zhang, Weidong Shi. (2019). Dual reverse attention networks for person re-identification. IEEE SigPort. http://sigport.org/4784
Shuangwei Liu, Lin Qi, Yunzhou Zhang, Weidong Shi, 2019. Dual reverse attention networks for person re-identification. Available at: http://sigport.org/4784.
Shuangwei Liu, Lin Qi, Yunzhou Zhang, Weidong Shi. (2019). "Dual reverse attention networks for person re-identification." Web.
1. Shuangwei Liu, Lin Qi, Yunzhou Zhang, Weidong Shi. Dual reverse attention networks for person re-identification [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4784