Structure-Preserving Bilateral Texture Filtering

Abstract

Extracting meaningful structures from images with complicated texture patterns is challenging since it is hard to separate structure from texture of similar scale or intensity contrast. In this paper, we propose a structure-preserving bilateral texture filtering algorithm to flatten texture while preserving dominant structures. We design a new scheme, dual-scale patch toggle. That is, patches of two scales are used to represent pixels, the smaller for pixels located at structure edges and the bigger for pixels in texture regions, and then DASM (Directional Anisotropic Structure Measurement) on each pixel is estimated to determine which type of patch to represent it. The algorithm is based on the joint bilateral filtering framework, so it is fast, easy to implement, yet effective for adaptive image smoothing. In particular, our approach outperforms previous methods in terms of preserving small structures. The proposed method achieves excellent results that illustrate its effectiveness and efficiency.