Texture Classification with Minimal Training Images
2008 (English)In: 19TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOLS 1-6, 2008, 717-720 p.Conference paper (Refereed)
The objective of this work is classifying texture from a single image under unknown lighting conditions. The current and successful approach to this task is to treat it as a statistical learning problem and learn a classifier from a set of training images, but this requires a sufficient number and variety of training images. We show that the number of training images required can be drastically reduced (to as few as three) by synthesizing additional training data using photometric stereo. We demonstrate the method on the PhoTex and ALOT texture databases. Despite the limitations of photometric stereo, the resulting classification performance surpasses the state of the art results.
Place, publisher, year, edition, pages
2008. 717-720 p.
, INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, ISSN 1051-4651
Classification performance, Lighting conditions, Minimal training, Photometric stereo, Single images, State of the art, Statistical learning, Texture classification, Training data, Training image, Face recognition, Photometry
Computer and Information Science
IdentifiersURN: urn:nbn:se:kth:diva-31230DOI: 10.1109/ICPR.2008.4761388ISI: 000264729000176ScopusID: 2-s2.0-77957936844ISBN: 978-1-4244-2174-9OAI: oai:DiVA.org:kth-31230DiVA: diva2:405911
19th International Conference on Pattern Recognition (ICPR 2008), Tampa, FL, DEC 08-11, 2008
QC 201103242011-03-242011-03-112011-03-24Bibliographically approved