MatFuse: Controllable Material Generation with Diffusion Models
arXiv:2308.11408 · doi:10.1109/CVPR52733.2024.00424
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.
References in corpus (12)
- Adam: A Method for Stochastic Optimization
- Decoupled Weight Decay Regularization
- Denoising Diffusion Probabilistic Models
- Neural Discrete Representation Learning
- Diffusion Models Beat GANs on Image Synthesis
- Classifier-Free Diffusion Guidance
- Single-Image SVBRDF Capture with a Rendering-Aware Deep Network
- Modeling Surface Appearance from a Single Photograph using Self-augmented Convolutional Neural Networks
- MaterialGAN: Reflectance Capture using a Generative SVBRDF Model
- Synthetic Datasets for Autonomous Driving: A Survey
- Composer: Creative and Controllable Image Synthesis with Composable Conditions
- MIDGARD: A Simulation Platform for Autonomous Navigation in Unstructured Environments