Graphcut textures

TLDR

The paper proposes a graph‑cut based algorithm that automatically selects optimal patch sizes for image and video texture synthesis. The method extracts and transforms patch regions from a source texture, then uses a graph‑cut seam optimization to stitch them together, with approximative offset search enabling efficient 2D and 3D synthesis. Experiments demonstrate high‑quality synthesis of regular, random, and natural images and videos, and show that the algorithm can be used interactively to merge disparate images into new scenes.

Abstract

In this paper we introduce a new algorithm for image and video texture synthesis. In our approach, patch regions from a sample image or video are transformed and copied to the output and then stitched together along optimal seams to generate a new (and typically larger) output. In contrast to other techniques, the size of the patch is not chosen a-priori , but instead a graph cut technique is used to determine the optimal patch region for any given offset between the input and output texture. Unlike dynamic programming, our graph cut technique for seam optimization is applicable in any dimension. We specifically explore it in 2D and 3D to perform video texture synthesis in addition to regular image synthesis. We present approximative offset search techniques that work well in conjunction with the presented patch size optimization. We show results for synthesizing regular, random, and natural images and videos. We also demonstrate how this method can be used to interactively merge different images to generate new scenes.