Paper Info
Title
Toward facial re-identification: Experiments with data from an operational surveillance camera plant
Abstract
Person re-identification (ReID) is a popular topic of research. Almost all existing ReID approaches employ local and global body features (e.g., clothing color and pattern, body symmetry, etc.). These `body ReID' methods implicitly assume that facial resolution is too low to aid in the ReID process. We assert that faces, even when captured in low resolution environments, may contain unique and stable features for ReID. Such `facial ReID' approaches are relatively unexplored in the literature. In this work, we explore facial ReID using a new dataset that was collected from a real surveillance network in a municipal rapid transit system. It is a challenging ReID dataset, as it includes intentional changes in persons' appearances over time. We conduct multiple experiments on this dataset, exploiting deep neural networks to extract dense, low resolution facial features to boost matching stability. We conclude that in cases where pedestrian appearance changes, low resolution faces can be utilized to improve ReID matching performance.
Year
DOI
Venue
2016
10.1109/BTAS.2016.7791204
2016 IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS)
Keywords
Field
DocType
facial reidentification,operational surveillance camera plant,person reidentification,body ReID method,facial resolution,municipal rapid transit system,low resolution facial features
Computer vision,Pedestrian,Computer science,Feature extraction,Surveillance camera,Artificial intelligence,Artificial neural network,Image resolution,Deep neural networks,Principal component analysis
Conference
ISSN
ISBN
Citations 
2474-9680
978-1-4673-9734-6
0
PageRank 
References 
Authors
0.34
0
3
Name
Order
Citations
PageRank
Pei Li181.85
Joel Brogan2142.93
Patrick J. Flynn34405307.04