View based approach to forensic face recognition

A. Dutta, R.T.A. van Rootseler, Raymond N.J. Veldhuis, Lieuwe Jan Spreeuwers

Research output: Book/ReportReport

Abstract

Face recognition is a challenging problem for surveillance view images commonly encountered in a forensic face recognition case. One approach to deal with a non-frontal test image is to synthesize the corresponding frontal view image and compare it with frontal view reference images. However, it is often difficult to synthesize a good quality frontal view image from a surveillance video because the test image is usually of low quality. In this paper, we investigate if it is useful to instead transform the reference images so that it matches the pose, illumination and camera of the surveillance view test image. This approach, also called the view based approach, ensures that a face recognition system always gets to compare images having a similar, not necessarily the frontal, view. Our results with surveillance view images captured 6 months apart (taken from the MultiPIE data set) and using five different face recognition systems show that improved recognition performance under surveillance conditions can be attained by exactly matching the pose, illumination and camera between the test and reference images.
LanguageUndefined
Place of PublicationEnschede
PublisherCentre for Telematics and Information Technology (CTIT)
Number of pages5
StatePublished - Sep 2012

Publication series

NameCTIT Technical Report Series
PublisherCentre for Telematics and Information Technology, University of Twente
No.TR-CTIT-12-21
ISSN (Print)1381-3625

Keywords

  • EWI-22236
  • METIS-289687
  • IR-84353
  • SCS-Safety

Cite this

Dutta, A., van Rootseler, R. T. A., Veldhuis, R. N. J., & Spreeuwers, L. J. (2012). View based approach to forensic face recognition. (CTIT Technical Report Series; No. TR-CTIT-12-21). Enschede: Centre for Telematics and Information Technology (CTIT).
Dutta, A. ; van Rootseler, R.T.A. ; Veldhuis, Raymond N.J. ; Spreeuwers, Lieuwe Jan. / View based approach to forensic face recognition. Enschede : Centre for Telematics and Information Technology (CTIT), 2012. 5 p. (CTIT Technical Report Series; TR-CTIT-12-21).
@book{3ee568fb01f141c2afea955fe9bd0781,
title = "View based approach to forensic face recognition",
abstract = "Face recognition is a challenging problem for surveillance view images commonly encountered in a forensic face recognition case. One approach to deal with a non-frontal test image is to synthesize the corresponding frontal view image and compare it with frontal view reference images. However, it is often difficult to synthesize a good quality frontal view image from a surveillance video because the test image is usually of low quality. In this paper, we investigate if it is useful to instead transform the reference images so that it matches the pose, illumination and camera of the surveillance view test image. This approach, also called the view based approach, ensures that a face recognition system always gets to compare images having a similar, not necessarily the frontal, view. Our results with surveillance view images captured 6 months apart (taken from the MultiPIE data set) and using five different face recognition systems show that improved recognition performance under surveillance conditions can be attained by exactly matching the pose, illumination and camera between the test and reference images.",
keywords = "EWI-22236, METIS-289687, IR-84353, SCS-Safety",
author = "A. Dutta and {van Rootseler}, R.T.A. and Veldhuis, {Raymond N.J.} and Spreeuwers, {Lieuwe Jan}",
year = "2012",
month = "9",
language = "Undefined",
series = "CTIT Technical Report Series",
publisher = "Centre for Telematics and Information Technology (CTIT)",
number = "TR-CTIT-12-21",
address = "Netherlands",

}

Dutta, A, van Rootseler, RTA, Veldhuis, RNJ & Spreeuwers, LJ 2012, View based approach to forensic face recognition. CTIT Technical Report Series, no. TR-CTIT-12-21, Centre for Telematics and Information Technology (CTIT), Enschede.

View based approach to forensic face recognition. / Dutta, A.; van Rootseler, R.T.A.; Veldhuis, Raymond N.J.; Spreeuwers, Lieuwe Jan.

Enschede : Centre for Telematics and Information Technology (CTIT), 2012. 5 p. (CTIT Technical Report Series; No. TR-CTIT-12-21).

Research output: Book/ReportReport

TY - BOOK

T1 - View based approach to forensic face recognition

AU - Dutta,A.

AU - van Rootseler,R.T.A.

AU - Veldhuis,Raymond N.J.

AU - Spreeuwers,Lieuwe Jan

PY - 2012/9

Y1 - 2012/9

N2 - Face recognition is a challenging problem for surveillance view images commonly encountered in a forensic face recognition case. One approach to deal with a non-frontal test image is to synthesize the corresponding frontal view image and compare it with frontal view reference images. However, it is often difficult to synthesize a good quality frontal view image from a surveillance video because the test image is usually of low quality. In this paper, we investigate if it is useful to instead transform the reference images so that it matches the pose, illumination and camera of the surveillance view test image. This approach, also called the view based approach, ensures that a face recognition system always gets to compare images having a similar, not necessarily the frontal, view. Our results with surveillance view images captured 6 months apart (taken from the MultiPIE data set) and using five different face recognition systems show that improved recognition performance under surveillance conditions can be attained by exactly matching the pose, illumination and camera between the test and reference images.

AB - Face recognition is a challenging problem for surveillance view images commonly encountered in a forensic face recognition case. One approach to deal with a non-frontal test image is to synthesize the corresponding frontal view image and compare it with frontal view reference images. However, it is often difficult to synthesize a good quality frontal view image from a surveillance video because the test image is usually of low quality. In this paper, we investigate if it is useful to instead transform the reference images so that it matches the pose, illumination and camera of the surveillance view test image. This approach, also called the view based approach, ensures that a face recognition system always gets to compare images having a similar, not necessarily the frontal, view. Our results with surveillance view images captured 6 months apart (taken from the MultiPIE data set) and using five different face recognition systems show that improved recognition performance under surveillance conditions can be attained by exactly matching the pose, illumination and camera between the test and reference images.

KW - EWI-22236

KW - METIS-289687

KW - IR-84353

KW - SCS-Safety

M3 - Report

T3 - CTIT Technical Report Series

BT - View based approach to forensic face recognition

PB - Centre for Telematics and Information Technology (CTIT)

CY - Enschede

ER -

Dutta A, van Rootseler RTA, Veldhuis RNJ, Spreeuwers LJ. View based approach to forensic face recognition. Enschede: Centre for Telematics and Information Technology (CTIT), 2012. 5 p. (CTIT Technical Report Series; TR-CTIT-12-21).