This paper introduces a texture representation suitable for recognizing images of textured surfaces under a wide range of transformations, including viewpoint changes and nonrigid deformations. At the feature extraction stage, a sparse set of affine-invariant local patches is extracted from the image. This spatial selection process permits the computation of characteristic scale and neighborhood shape for every texture element. The proposed texture representation is evaluated in retrieval and classification tasks using the entire Brodatz database and a collection of photographs of textured surfaces taken from different viewpoints.
A sparse texture representation using affine-invariant regions
2003-01-01
2018297 byte
Conference paper
Electronic Resource
English
A Sparse Texture Representation Using Affine-Invariant Regions
British Library Conference Proceedings | 2003
|Affine Invariant Texture Segmentation and Shape from Texture by Variational Methods
British Library Online Contents | 1998
|A sparse curvature-based detector of affine invariant blobs
British Library Online Contents | 2012
|Efficient Texture Representation Using Multi-scale Regions
Springer Verlag | 2007
|