Publication Details

Learning Feature Aggregation in Temporal Domain for Re-Identification

ŠPAŇHEL Jakub, SOCHOR Jakub, JURÁNEK Roman, DOBEŠ Petr, BARTL Vojtěch and HEROUT Adam. Learning Feature Aggregation in Temporal Domain for Re-Identification. Computer Vision and Image Understanding, vol. 192, no. 11, 2020, pp. 1-12. ISSN 1077-3142. Available from: https://www.sciencedirect.com/science/article/pii/S107731421830393X
Czech title
Učení agregace příznaků v časové doméně pro re-identifikaci objektů
Type
journal article
Language
english
Authors
URL
Keywords

person re-identification, vehicle re-identification, feature aggregation, temporal domain, neural network, traffic surveillance

Abstract

Person re-identification is a standard and established problem in the computer vision community. In recent years, vehicle re-identification is also getting more attention. In this paper, we focus on both these tasks and propose a method for aggregation of features in temporal domain as it is common to have multiple observations of the same object. The aggregation is based on weighting different elements of the feature vectors by different weights and it is trained in an end-to-end manner by a Siamese network. The experimental results show that our method outperforms other existing methods for feature aggregation in temporal domain on both vehicle and person re-identification tasks. Furthermore, to push research in vehicle re-identification further, we introduce a novel dataset CarsReId74k. The dataset is not limited to frontal/rear viewpoints. It contains 17,681 unique vehicles, 73,976 observed tracks, and 277,236 positive pairs. The dataset was captured by 66 cameras from various angles.

Published
2020
Pages
1-12
Journal
Computer Vision and Image Understanding, vol. 192, no. 11, ISSN 1077-3142
Book
Computer Vision and Image Understanding
Publisher
Elsevier Science
Place
Amsterdam, NL
DOI
UT WoS
000514226900003
EID Scopus
BibTeX
@ARTICLE{FITPUB12126,
   author = "Jakub \v{S}pa\v{n}hel and Jakub Sochor and Roman Jur\'{a}nek and Petr Dobe\v{s} and Vojt\v{e}ch Bartl and Adam Herout",
   title = "Learning Feature Aggregation in Temporal Domain for Re-Identification",
   pages = "1--12",
   booktitle = "Computer Vision and Image Understanding",
   journal = "Computer Vision and Image Understanding",
   volume = 192,
   number = 11,
   year = 2020,
   location = "Amsterdam, NL",
   publisher = "Elsevier Science",
   ISSN = "1077-3142",
   doi = "10.1016/j.cviu.2019.102883",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/12126"
}
Back to top