A Compound Moving Average Bidirectional Texture Function Model

Abstract: 
This paper describes a simple novel compound random field model capable of realistic modelling the most advanced recent representation of visual properties of surface materials - the bidirectional texture function. The presented compound random field model combines a non-parametric control random field with local multispectral models for single regions and thus allows to avoid demanding iterative methods for both parameters estimation and the compound random field synthesis. The local texture regions (not necessarily continuous) are represented by an analytical bidirectional texture function model which consists of single scale factors modeled by the three-dimensional moving average random field model which can be analytically estimated as well as synthesized.

From the left to the right: original texture (lichen), its control field (five classes), enlarged (2x) control field (by Roller), mapped synthesized classes (each class analyzed/synthesized by extended moving average random field model) resulting in enlarged synthetic texture.