VideoMatGen: PBR Materials through Joint Generative Modeling

Jon Hasselgren, Milos Hasan, Zheng Zeng, Jacob Munkberg; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Findings, 2026, pp. 2440-2450

We present a method for generating physically-based materials for 3D shapes based on a video diffusion transformer architecture. Our method is conditioned on input geometry and a text description, and jointly models multiple material properties (albedo, roughness, metallicity, height map) to form physically plausible materials. We further introduce a custom variational auto-encoder which encodes multiple material modalities into a compact latent space, which enables joint generation of multiple modalities without increasing the number of tokens. Our pipeline generates high-quality materials for 3D shapes given a text prompt, compatible with common content creation tools.