A parametric texture model based on joint statistics of complex wavelet coefficients

We present a universal statistical model for texture images in the context of an overcomplete complex wavelet transform. The model is parameterized by a set of statistics computed on pairs of coefficients corresponding to bas is functions at adjacent spatial locations, orientations, and scales. We de velop an efficient algorithm for synthesizing random images subject to thes e constraints, by iteratively projecting onto the set of images satisfying each constraint, and we use this to test the perceptual validity of the mod el. In particular, we demonstrate the necessity of subgroups of the paramet er set by showing examples of texture synthesis that fail when those parame ters are removed from the set. We also demonstrate the power of our model b y successfully synthesizing examples drawn from a diverse collection of art ificial and natural textures.