A sparse texture representation using local affine regions

TLDR

Texture elements are represented as elliptic regions with distinctive appearance patterns. The paper proposes a texture representation that enables recognition of textured surfaces under diverse transformations, including viewpoint changes and nonrigid deformations. The method extracts a sparse set of affine Harris and Laplacian regions, normalizes their shapes, and computes two novel descriptors—the spin image and RIFT—to capture affine‑invariant texture patterns, while optionally using the original elliptical shape for discrimination. The representation was evaluated on the Brodatz database and a 1,000‑photo viewpoint collection, achieving effective retrieval and classification performance.

Abstract

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 Harris and Laplacian regions is found in the image. Each of these regions can be thought of as a texture element having a characteristic elliptic shape and a distinctive appearance pattern. This pattern is captured in an affine-invariant fashion via a process of shape normalization followed by the computation of two novel descriptors, the spin image and the RIFT descriptor. When affine invariance is not required, the original elliptical shape serves as an additional discriminative feature for texture recognition. The proposed approach is evaluated in retrieval and classification tasks using the entire Brodatz database and a publicly available collection of 1,000 photographs of textured surfaces taken from different viewpoints.