Fast texture synthesis using tree-structured vector quantization

TLDR

Texture synthesis is important for many applications in computer graphics, vision, and image processing, yet designing an efficient algorithm that produces high‑quality results remains challenging. The paper proposes an efficient algorithm for realistic texture synthesis. The method uses a Markov Random Field model and deterministic search, accelerated by tree‑structured vector quantization, requiring only a sample texture as input. It produces textures of equal or superior perceived quality to prior methods while running two orders of magnitude faster, enabling applications such as constrained image editing and temporal texture generation that were previously impractical.

Abstract

Texture synthesis is important for many applications in computer graphics, vision, and image processing. However, it remains difficult to design an algorithm that is both efficient and capable of generating high quality results. In this paper, we present an efficient algorithm for realistic texture synthesis. The algorithm is easy to use and requires only a sample texture as input. It generates textures with perceived quality equal to or better than those produced by previous techniques, but runs two orders of magnitude faster. This permits us to apply texture synthesis to problems where it has traditionally been considered impractical. In particular, we have applied it to constrained synthesis for image editing and temporal texture generation. Our algorithm is derived from Markov Random Field texture models and generates textures through a deterministic searching process. We accelerate this synthesis process using tree-structured vector quantization.