Texton theory revisited: A bag-of-words approach to combine textons

The aim of this paper is to revisit an old theory of texture perception and update its computational implementation by extending it to colour. With this in mind we try to capture the optimality of perceptual systems. This is achieved in the proposed approach by sharing well-known early stages of the visual processes and extracting low-dimensional features that perfectly encode adequate properties for a large variety of textures without needing further learning stages.We propose several descriptors in a bag-of-words framework that are derived from different quantisation models on to the feature spaces. Our perceptual features are directly given by the shape and colour attributes of image blobs, which are the textons. In this way we avoid learning visual words and directly build the vocabularies on these low-dimensional texton spaces. Main differences between proposed descriptors rely on how co-occurrence of blob attributes is represented in the vocabularies. Our approach overcomes current state-of-art in colour texture description which is proved in several experiments on large texture datasets.

► We revisit the Texton theory and update its computational implementation by extending it to colour.
► We propose several colour-texture representations based on a BoW framework.
► Low-level features are obtained from early stages of visual processes.
► Vocabularies built quantising feature spaces without any learning step.
► Our descriptors overcome current state-of art in image retrieval and classification applications.