We present MatAtlas, a method for consistent text-guided 3D model texturing. Following recent progress we leverage a large scale text-to-image generation model (e.g., Stable Diffusion) as a prior to texture a 3D model. We carefully design an RGB texturing pipeline that leverages a grid pattern diffusion, driven by depth and edges. By proposing a multi-step texture refinement process, we significantly improve the quality and 3D consistency of the texturing output. To further address the problem of baked-in lighting, we move beyond RGB colors and pursue assigning parametric materials to the assets. Given the high-quality initial RGB texture, we propose a novel material retrieval method capitalized on Large Language Models (LLM), enabling editabiliy and relightability. We evaluate our method on a wide variety of geometries and show that our method significantly outperform prior arts. We also analyze the role of each component through a detailed ablation study.
@article{arxiv.2404.02899,
title = {MatAtlas: Text-driven Consistent Geometry Texturing and Material Assignment},
author = {Duygu Ceylan and Valentin Deschaintre and Thibault Groueix and Rosalie Martin and Chun-Hao Huang and Romain Rouffet and Vladimir Kim and Gaëtan Lassagne},
journal= {arXiv preprint arXiv:2404.02899},
year = {2024}
}