Presentation 2014-01-24
People Re-identification with Auxiliary Knowledge
Guanwen ZHANG, Jien KATO, Yu WANG, Kenji MASE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) There is an intrinsic issue in multiple-shot person re-identification: only a few training data for learning tasks are available in a realistic re-identification scenario. A novel adaptive metric learning method is introduced in this paper for multiple-shot people re-identification. By leveraging the auxiliary knowledge of re-identification together with the specific information of the target task, the proposed adaptive learning method is able to solve over-fitting problem caused by limited training data. We evaluated our approach on public benchmark datasets, and confirmed its superiority as compared to conventional approaches.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) People Re-identification / Auxiliary Knowledge / Adaptive Metric Learning / Local Distance Comparison
Paper # PRMU2013-114,MVE2013-55
Date of Issue

Conference Information
Committee MVE
Conference Date 2014/1/16(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Media Experience and Virtual Environment (MVE)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) People Re-identification with Auxiliary Knowledge
Sub Title (in English)
Keyword(1) People Re-identification
Keyword(2) Auxiliary Knowledge
Keyword(3) Adaptive Metric Learning
Keyword(4) Local Distance Comparison
1st Author's Name Guanwen ZHANG
1st Author's Affiliation Graduate School of Information Science, Nagoya University()
2nd Author's Name Jien KATO
2nd Author's Affiliation Graduate School of Information Science, Nagoya University
3rd Author's Name Yu WANG
3rd Author's Affiliation Graduate School of Information Science, Nagoya University
4th Author's Name Kenji MASE
4th Author's Affiliation Graduate School of Information Science, Nagoya University
Date 2014-01-24
Paper # PRMU2013-114,MVE2013-55
Volume (vol) vol.113
Number (no) 403
Page pp.pp.-
#Pages 5
Date of Issue