Video face matching using subset selection and clustering of probabilistic multi-region histograms

Mau, Sandra, Chen, Shaokang, Sanderson, Conrad and Lovell, Brian C. (2010). Video face matching using subset selection and clustering of probabilistic multi-region histograms. In: 25th International Conference of Image and Vision Computing (IVCNZ). 25th International Conference of Image and Vision Computing (IVCNZ), Queenstown, New Zealand, (1-8). 8 - 9 November 2010. doi:10.1109/IVCNZ.2010.6148860

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads

Author Mau, Sandra
Chen, Shaokang
Sanderson, Conrad
Lovell, Brian C.
Title of paper Video face matching using subset selection and clustering of probabilistic multi-region histograms
Conference name 25th International Conference of Image and Vision Computing (IVCNZ)
Conference location Queenstown, New Zealand
Conference dates 8 - 9 November 2010
Proceedings title 25th International Conference of Image and Vision Computing (IVCNZ)
Journal name International Conference Image and Vision Computing New Zealand
Place of Publication Washington, DC, United States
Publisher IEEE Computer Society
Publication Year 2010
Sub-type Fully published paper
DOI 10.1109/IVCNZ.2010.6148860
ISBN 9781424496297
ISSN 2151-2191
Start page 1
End page 8
Total pages 8
Collection year 2011
Language eng
Abstract/Summary Balancing computational eciency with recognition accuracy is one of the major challenges in real-world video-based face recognition. A signicant design decision for any such system is whether to process and use all possible faces detected over the video frames, or whether to select only a few `best' faces. This paper presents a video face recognition system based on probabilistic Multi-Region Histograms to characterise performance trade-os in: (i) selecting a subset of faces compared to using all faces, and (ii) combining information from all faces via clustering. Three face selection metrics are evaluated for choosing a subset: face detection condence, random subset, and sequential selection. Experiments on the recently introduced MOBIO dataset indicate that the usage of all faces through clustering always outperformed selecting only a subset of faces. The experiments also show that the face selection metric based on face detection condence generally provides better recognition performance than random or sequential sampling. Moreover, the optimal number of faces varies drastically across selection metric and subsets of MOBIO. Given the trade-os between computational eort, recognition accuracy and robustness, it is recommended that face feature clustering would be most advantageous in batch processing (particularly for video-based watchlists), whereas face selection methods should be limited to applications with signicant computational restrictions.
Keyword Surveillance
Local features
Video processing
Face matching
Clustering
Subset selection
Efficiency
Q-Index Code E1
Q-Index Status Confirmed Code
Institutional Status UQ
Additional Notes Conference proceedings to be published by IEEE Explore. Conference series title: Image and Vision Computing Conference - ERA-ranked B - ERA ID: 43120.

 
Versions
Version Filter Type
Citation counts: Scopus Citation Count Cited 0 times in Scopus Article
Google Scholar Search Google Scholar
Created: Fri, 26 Nov 2010, 18:37:28 EST by Conrad Sanderson on behalf of School of Information Technol and Elec Engineering