RoCoTex: A Robust Method for Consistent Texture Synthesis with Diffusion Models

TL;DR

This paper introduces RoCoTex, a robust texture synthesis method using diffusion models that produces consistent, seamless, and well-aligned textures with meshes, overcoming view inconsistency and seam issues.

Contribution

RoCoTex combines diffusion models, view synthesis, regional prompts, and novel blending techniques to improve texture consistency and seam reduction in texture generation.

Findings

Outperforms existing methods in view consistency and seam reduction

Produces highly detailed and well-aligned textures

Demonstrates robustness across various textures and meshes

Abstract

Text-to-texture generation has recently attracted increasing attention, but existing methods often suffer from the problems of view inconsistencies, apparent seams, and misalignment between textures and the underlying mesh. In this paper, we propose a robust text-to-texture method for generating consistent and seamless textures that are well aligned with the mesh. Our method leverages state-of-the-art 2D diffusion models, including SDXL and multiple ControlNets, to capture structural features and intricate details in the generated textures. The method also employs a symmetrical view synthesis strategy combined with regional prompts for enhancing view consistency. Additionally, it introduces novel texture blending and soft-inpainting techniques, which significantly reduce the seam regions. Extensive experiments demonstrate that our method outperforms existing state-of-the-art methods.