Towards Pose-Invariant 2D Face Classification for Surveillance

Conrad Sanderson, Ting Shan and Brian C. Lovell (2007). Towards Pose-Invariant 2D Face Classification for Surveillance. In: Analysis and Modeling of Faces and Gestures Third InternationalWorkshop, AMFG 2007 proceedings. Analysis and Modeling of Faces and Gestures Third InternationalWorkshop (AMFG 2007), Rio de Janeiro, Brazil, (276-289). 20 October 2007. doi:10.1007/978-3-540-75690-3_21

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
sanderson_amfg_2007.pdf sanderson_amfg_2007.pdf application/pdf 541.01KB 0

Author Conrad Sanderson
Ting Shan
Brian C. Lovell
Title of paper Towards Pose-Invariant 2D Face Classification for Surveillance
Conference name Analysis and Modeling of Faces and Gestures Third InternationalWorkshop (AMFG 2007)
Conference location Rio de Janeiro, Brazil
Conference dates 20 October 2007
Proceedings title Analysis and Modeling of Faces and Gestures Third InternationalWorkshop, AMFG 2007 proceedings   Check publisher's open access policy
Journal name Lecture Notes in Computer Science   Check publisher's open access policy
Place of Publication Heidelberg, Germany
Publisher Springer
Publication Year 2007
Sub-type Fully published paper
DOI 10.1007/978-3-540-75690-3_21
Open Access Status File (Author Post-print)
ISBN 9783540756897
ISSN 0302-9743
Volume 4778
Start page 276
End page 289
Total pages 14
Language eng
Abstract/Summary A key problem for "face in the crowd" recognition from existing surveillance cameras in public spaces (such as mass transit centres) is the issue of pose mismatches between probe and gallery faces. In addition to accuracy, scalability is also important, necessarily limiting the complexity of face classification algorithms. In this paper we evaluate recent approaches to the recognition of faces at relatively large pose angles from a gallery of frontal images and propose novel adaptations as well as modifications. Specifically, we compare and contrast the accuracy, robustness and speed of an Active Appearance Model (AAM) based method (where realistic frontal faces are synthesized from non-frontal probe faces) against bag-of-features methods (which are local feature approaches based on block Discrete Cosine Transforms and Gaussian Mixture Models). We show a novel approach where the AAM based technique is sped up by directly obtaining pose-robust features, allowing the omission of the computationally expensive and artefact producing image synthesis step. Additionally, we adapt a histogram-based bag-of-features technique to face classification and contrast its properties to a previously proposed direct bag-of-features method. We also show that the two bag-of-features approaches can be considerably sped up, without a loss in classification accuracy, via an approximation of the exponential function. Experiments on the FERET and PIE databases suggest that the bag-of-features techniques generally attain better performance, with significantly lower computational loads. The histogram-based bag-of-features technique is capable of achieving an average recognition accuracy of 89% for pose angles of around 25 degrees.
Subjects 280200 Artificial Intelligence and Signal and Image Processing
280203 Image Processing
280207 Pattern Recognition
280208 Computer Vision
230204 Applied Statistics
810107 National Security
Q-Index Code E1
Q-Index Status Provisional Code
Institutional Status UQ

Document type: Conference Paper
Sub-type: Fully published paper
Collections: Faculty of Engineering, Architecture and Information Technology Publications
Security and Surveillance Collection
Version Filter Type
Citation counts: Scopus Citation Count Cited 15 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Thu, 31 Jul 2008, 03:12:58 EST by Conrad Sanderson on behalf of School of Information Technol and Elec Engineering