MatFuse: Controllable Material Generation with Diffusion Models

TL;DR

MatFuse is a novel diffusion-based framework that enables controllable creation and editing of 3D materials in computer graphics, integrating multiple conditioning sources for enhanced flexibility and quality.

Contribution

We introduce MatFuse, a unified diffusion model that allows multi-source conditioned material generation and map-level editing with disentangled latent representations.

Findings

Effective multi-condition material generation demonstrated.

High-quality materials validated by CLIP-IQA and FID scores.

User study confirms improved control and realism.

Abstract

Creating high-quality materials in computer graphics is a challenging and time-consuming task, which requires great expertise. To simplify this process, we introduce MatFuse, a unified approach that harnesses the generative power of diffusion models for creation and editing of 3D materials. Our method integrates multiple sources of conditioning, including color palettes, sketches, text, and pictures, enhancing creative possibilities and granting fine-grained control over material synthesis. Additionally, MatFuse enables map-level material editing capabilities through latent manipulation by means of a multi-encoder compression model which learns a disentangled latent representation for each map. We demonstrate the effectiveness of MatFuse under multiple conditioning settings and explore the potential of material editing. Finally, we assess the quality of the generated materials both quantitatively in terms of CLIP-IQA and FID scores and qualitatively by conducting a user study. Source code for training MatFuse and supplemental materials are publicly available at https://gvecchio.com/matfuse.