Object categorization via local kernels
2004 (English)In: PROCEEDINGS OF THE 17TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL 2 / [ed] Kittler, J; Petrou, M; Nixon, M, 2004, 132-135 p.Conference paper (Refereed)
This paper considers the problem of multi-object categorization. We present an algorithm that combines support vector machines with local features via a new class of Mercer kernels. This class of kernels allows us to perform scalar products on feature vectors consisting of local descriptors, computed around interest points (like corners); these feature vectors are generally of different lengths for different images. The resulting framework is able to recognize multi-object categories in different settings, from lab-controlled to real-world scenes. We present several experiments, on different databases, and we benchmark our results with state-of-the-art algorithms for categorization, achieving excellent results.
Place, publisher, year, edition, pages
2004. 132-135 p.
, INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, ISSN 1051-4651
Computer and Information Science
IdentifiersURN: urn:nbn:se:kth:diva-44312DOI: 10.1109/ICPR.2004.1334079ISI: 000223877400032ScopusID: 2-s2.0-10044240377ISBN: 0-7695-2128-2OAI: oai:DiVA.org:kth-44312DiVA: diva2:451225
17th International Conference on Pattern Recognition (ICPR) Location: British Machine Vis Assoc, Cambridge, ENGLAND Date: AUG 23-26, 2004
QC 201110252011-10-252011-10-202011-10-25Bibliographically approved