We present a framework for texture recognition based on local affine-invariant descriptors and their spatial layout. At modelling time, a generative model of local descriptors is learned from sample images using the EM algorithm. The EM framework allows the incorporation of unsegmented multitexture images into the training set. The second modelling step consists of gathering co-occurrence statistics of neighboring descriptors. At recognition time, initial probabilities computed from the generative model are refined using a relaxation step that incorporates co-occurrence statistics. Performance is evaluated on images of an indoor scene and pictures of wild animals.
Affine-invariant local descriptors and neighborhood statistics for texture recognition
2003-01-01
1101205 byte
Conference paper
Electronic Resource
English
Affine-Invariant Local Descriptors and Neighborhood Statistics for Texture Recognition
British Library Conference Proceedings | 2003
|Illumination invariant optical flow using neighborhood descriptors
British Library Online Contents | 2016
|Illumination invariant optical flow using neighborhood descriptors
British Library Online Contents | 2016
|Affine invariant descriptors using principal components analysis
British Library Online Contents | 2008
|Ear biometric recognition using local texture descriptors
British Library Online Contents | 2014
|