We question the role that large scale filter banks have traditionally played in texture classification. It is demonstrated that textures can be classified using the joint distribution of intensity values over extremely compact neighborhoods (starting from as small as 3 /spl times/ 3 pixels square), and that this outperforms classification using filter banks with large support. We develop a novel texton based representation, which is suited to modeling this joint neighborhood distribution for MRFs. The representation is learnt from training images, and then used to classify novel images (with unknown viewpoint and lighting) into texture classes. The power of the method is demonstrated by classifying over 2800 images of all 61 textures present in the Columbia-Utrecht database. The classification performance surpasses that of recent state-of-the-art filter bank based classifiers such as Leung & Malik, Cula & Dana, and Varma & Zisserman.
Texture classification: are filter banks necessary?
2003-01-01
383462 byte
Conference paper
Electronic Resource
English
Texture Classification: Are Filter Banks Necessary?
British Library Conference Proceedings | 2003
|Deep Filter Banks for Texture Recognition, Description, and Segmentation
British Library Online Contents | 2016
|Necessary and sufficient condition for perfect reconstruction matrix filter banks [3813-71]
British Library Conference Proceedings | 1999
|British Library Conference Proceedings | 1994
|Optimizing Gabor Filter Design for Texture Edge Detection and Classification
British Library Online Contents | 2009
|