Recognition from appearance subspaces across image sets of variable scale Arandjelovic, Ognjen 2010, Recognition from appearance subspaces across image sets of variable scale, in BMVC 2010 : Proceedings of the 21st British machine vision association conference 2010, BMVA Press, Manchester, England, pp. 1-11, doi: 10.5244/C.24.79. Attached Files Name Description MIMEType Size Downloads arandjelovic-recognitionfrom-2010.pdf Published version application/pdf 543.98KB 190 Title Recognition from appearance subspaces across image sets of variable scale Author(s) Arandjelovic, Ognjen Conference name British Machine Vision. Conference (21st : 2010 : Aberystwyth, Wales) Conference location Aberystwyth, Wales Conference dates 31 Aug.-3 Sept. 2010 Title of proceedings BMVC 2010 : Proceedings of the 21st British machine vision association conference 2010 Editor(s) [Unknown] Publication date 2010 Conference series British Machine Vision Conference Start page 1 End page 11 Total pages 11 Publisher BMVA Press Place of publication Manchester, England Summary Linear subspace representations of appearance variation are pervasive in computer vision. In this paper we address the problem of robustly matching them (computing the similarity between them) when they correspond to sets of images of different (possibly greatly so) scales. We show that the naïve solution of projecting the low-scale subspace into the high-scale image space is inadequate, especially at large scale discrepancies. A successful approach is proposed instead. It consists of (i) an interpolated projection of the low-scale subspace into the high-scale space, which is followed by (ii) a rotation of this initial estimate within the bounds of the imposed “downsampling constraint”. The optimal rotation is found in the closed-form which best aligns the high-scale reconstruction of the low-scale subspace with the reference it is compared to. The proposed method is evaluated on the problem of matching sets of face appearances under varying illumination. In comparison to the naïve matching, our algorithm is shown to greatly increase the separation of between-class and within-class similarities, as well as produce far more meaningful modes of common appearance on which the match score is based. ISBN 01901725405 Language eng DOI 10.5244/C.24.79 Field of Research 080104 Computer Vision 080106 Image Processing 080109 Pattern Recognition and Data Mining Socio Economic Objective 890205 Information Processing Services (incl. Data Entry and Capture) HERDC Research category E1.1 Full written paper - refereed Copyright notice ©2010, BMVA Free to Read? Yes Persistent URL http://hdl.handle.net/10536/DRO/DU:30058424 Document type: Conference Paper Collections: Faculty of Science, Engineering and Built Environment Centre for Pattern Recognition and Data Analytics (PRADA) Open Access Collection Related Links Link Description Connect to published version (restricted access) Go to link with your DU access privileges     Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved. Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au. Versions Version Filter Type Tue, 26 Nov 2013, 12:51:16 EST Tue, 26 Nov 2013, 13:12:14 EST Wed, 27 Nov 2013, 15:18:29 EST Mon, 03 Feb 2014, 11:09:38 EST Tue, 17 Nov 2015, 13:59:34 EST Wed, 18 Nov 2015, 06:38:33 EST Tue, 05 Dec 2017, 12:44:49 EST Fri, 07 Sep 2018, 16:48:26 EST Filtered Full Citation counts:   Cited 0 times in TR Web of Science Cited 0 times in Scopus Search Google Scholar Access Statistics: 180 Abstract Views, 190 File Downloads  -  Detailed Statistics Created: Tue, 26 Nov 2013, 12:51:16 EST