This paper presents a framework for texture recognition based on local affine-invariant descriptors and their spatial layout. At modeling time, a generative model of local descriptors is learned from sample images using the EM algorithm. The EM framework allows the incorporation of unsegmented multi-texture images into the training set. The second modeling 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.