Texture synthesis and unsupervised recognition with a nonparametric multiscale Markov random field model

We present noncausal, nonparametric, multiscale, Markov random field model for synthesising and recognising texture. The model has the ability to capture the characteristics of a wide variety of textures. For texture synthesis, we use our own novel multiscale approach, incorporating local annealing, allowing one to use large neighbourhood systems to model some complex textures. We show how one is able to manipulate the statistical order of our high dimensional model without over compromising the integrity of the representation. Also, by varying the statistical order of our model we are able to optimise it for the unsupervised recognition of textures with respect to textures that have not been modelled