The Portilla-Simoncelli Texture Model: Towards the Understanding of the Early Visual Cortex
Jonathan Vacher, Thibaud Briand
⚠ This is a preprint. It may change before it is accepted for publication.

Abstract

Texture synthesis is a prolific subarea in computer vision where statistic-based methods are often successful. The Portilla and Simoncelli (PS) pyramid-based texture analysis/synthesis al- gorithm is such a method that is still considered as a state-of-the art texture synthesis algorithm. Inspired by the functional properties of the primary visual cortex and by tex- ture perception theories, the PS algorithm has had a large impact in vision studies. In the following article we briefly review the influence of this algorithm. We bring a detailed explana- tion of the mathematical framework and provide a fast C++ implementation. Finally, we test different parameters of the algorithm that we illustrate with many synthesis examples.

Download