Lu, Chris Xiaoxuan , Kan , Xuan , Du, Bowen, Chen , Changhao , Wen , Hongkai, Markham , Andrew , Trigoni , Niki and Stankovic , John (2019) Autonomous learning for face recognition in the wild via ambient wireless cues. In: The Web Conference 2019, San Francisco, 13-17 May 2019. Published in: Proceedings of WWW '19 The World Wide Web Conference 1175-1186 . doi:10.1145/3308558.3313398

Preview	PDF WRAP-autonomous-learning-face-recognition-ambient-cues-Wen-2019.pdf - Published Version - Requires a PDF viewer. Available under License Creative Commons Attribution 4.0. Download (1846Kb) | Preview
	PDF WRAP-autonomous-learning-face-recognition-wild-ambient-cues-Wen-2019.pdf - Accepted Version Embargoed item. Restricted access to Repository staff only - Requires a PDF viewer. Download (2191Kb)

Request Changes to record.

Abstract

Facial recognition is a key enabling component for emerging Internet of Things (IoT) services such as smart homes or responsive offices. Through the use of deep neural networks, facial recognition has achieved excellent performance. However, this is only possibly when trained with hundreds of images of each user in different viewing and lighting conditions. Clearly, this level of effort in enrolment and labelling is impossible for wide-spread deployment and adoption. Inspired by the fact that most people carry smart wireless devices with them, e.g. smartphones, we propose to use this wireless identifier as a supervisory label. This allows us to curate a dataset of facial images that are unique to a certain domain e.g. a set of people in a particular office. This custom corpus can then be used to finetune existing pre-trained models e.g. FaceNet. However, due to the vagaries of wireless propagation in buildings, the supervisory labels are noisy and weak. We propose a novel technique, AutoTune, which learns and refines the association between a face and wireless identifier over time, by increasing the inter-cluster separation and minimizing the intra-cluster distance. Through extensive experiments with multiple users on two sites, we demonstrate the ability of AutoTune to design an environment-specific, continually evolving facial recognition system with entirely no user effort.

Item Type:	Conference Item (Paper)
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions:	Faculty of Science, Engineering and Medicine > Science > Computer Science
Library of Congress Subject Headings (LCSH):	Human face recognition (Computer science), Internet of things, FaceNet
Journal or Publication Title:	Proceedings of WWW '19 The World Wide Web Conference
Publisher:	ACM
Official Date:	21 January 2019
Dates:	Date Event 21 January 2019 Accepted
Page Range:	1175-1186
DOI:	10.1145/3308558.3313398
Status:	Peer Reviewed
Publication Status:	Published
Access rights to Published version:	Restricted or Subscription Access
Date of first compliant deposit:	11 February 2019
Date of first compliant Open Access:	11 February 2019
Conference Paper Type:	Paper
Title of Event:	The Web Conference 2019
Type of Event:	Conference
Location of Event:	San Francisco
Date(s) of Event:	13-17 May 2019
Related URLs:	Publisher

Request changes or add full text files to a record

Repository staff actions (login required)

View Item

Downloads