Wang, Foqin, Zhang, Xuehan, Ma, Jixin, Tang, Jin and Zheng, Aihua (2017) Manifold ranking weighted local maximal occurrence descriptor for person re-identification. In: 2017 IEEE 15th International Conference on Software Engineering Research, Management and Applications (SERA). IEEE, pp. 111-114. ISBN 978-1509057566 (doi:10.1109/SERA.2017.7965715)

Preview

PDF (Author's Accepted Manuscript) 24322 MA_Manifold_Weighted_Occurrence_Descriptor_Re-identification_(AAM)_2017.pdf - Accepted Version Download (178kB) | Preview

Abstract

Person re-identification is an important task of matching pedestrians across non-overlapping camera views. In this paper, we exploit a weighted feature descriptor for person re-identification.We firstly compute the weights on the superpixel level via graph-based manifold ranking algorithm, then integrate the computed weights into a patch-based feature descriptor, named local maximal occurrence. Finally, the weighted descriptors are fed into a top-push distance learning to mitigate the cross-view gaps. We evaluate the proposed method on three benchmark datasets iLIDS-VID, PRID 450S and VIPeR. The promising experimental results demonstrate the effectiveness of the proposed method comparing with the state-of-the-arts.

Item Type:	Conference Proceedings
Title of Proceedings:	2017 IEEE 15th International Conference on Software Engineering Research, Management and Applications (SERA)
Uncontrolled Keywords:	person re-identification, manifold ranking, local maximal occurrence, weighted descriptor
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Faculty / School / Research Centre / Research Group:	Faculty of Liberal Arts & Sciences > Computational Science & Engineering Group (CSEH) Faculty of Engineering & Science > School of Computing & Mathematical Sciences (CMS) Faculty of Engineering & Science
Last Modified:	04 Mar 2022 13:07
URI:	http://gala.gre.ac.uk/id/eprint/24322

Actions (login required)

View Item

Downloads