A Generative Model of Natural Texture Surrogates

TL;DR

This paper introduces a generative model for natural textures based on statistical analysis of image patches, enabling realistic texture synthesis, efficient image compression, and evaluation of generative models.

Contribution

The authors develop a Gaussian-based statistical model of texture parameters that accurately reproduces natural textures with fewer components, improving synthesis and compression.

Findings

Textures generated are perceptually similar to natural images.

The model achieves high-quality image compression at 0.14 bits/pixel.

The approach provides an objective way to evaluate generative models.

Abstract

Natural images can be viewed as patchworks of different textures, where the local image statistics is roughly stationary within a small neighborhood but otherwise varies from region to region. In order to model this variability, we first applied the parametric texture algorithm of Portilla and Simoncelli to image patches of 64X64 pixels in a large database of natural images such that each image patch is then described by 655 texture parameters which specify certain statistics, such as variances and covariances of wavelet coefficients or coefficient magnitudes within that patch. To model the statistics of these texture parameters, we then developed suitable nonlinear transformations of the parameters that allowed us to fit their joint statistics with a multivariate Gaussian distribution. We find that the first 200 principal components contain more than 99% of the variance and are sufficient to generate textures that are perceptually extremely close to those generated with all 655 components. We demonstrate the usefulness of the model in several ways: (1) We sample ensembles of texture patches that can be directly compared to samples of patches from the natural image database and can to a high degree reproduce their perceptual appearance. (2) We further developed an image compression algorithm which generates surprisingly accurate images at bit rates as low as 0.14 bits/pixel. Finally, (3) We demonstrate how our approach can be used for an efficient and objective evaluation of samples generated with probabilistic models of natural images.