PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Combining local feature histograms of different granularities
Ville Viitaniemi and Jorma Laaksonen
In: 16th Scandinavian Conference on Image Analysis, 15-18 Jun 2009, Oslo, Norway.

Abstract

Histograms of local features have proven to be powerful representations in image category detection. Histograms with different numbers of bins encode the visual information with different granularities. In this paper we experimentally compare techniques for combining different granularities in a way that the resulting descriptors can be used as feature vectors in conventional vector space learning algorithms. In particular, we consider two main approaches: fusing the granularities on SVM kernel level and moving away from binary or hard to soft histograms. We find soft histograms to be a more effective approach, resulting in substantial performance improvement over single-granularity histograms.

EPrint Type:Conference or Workshop Item (Poster)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Computational, Information-Theoretic Learning with Statistics
Machine Vision
Theory & Algorithms
Information Retrieval & Textual Information Access
ID Code:6590
Deposited By:Jorma Laaksonen
Deposited On:08 March 2010