Graphics pipelines require physically-based rendering (PBR) materials, yet current 3D content generation approaches are built on RGB models. We propose to model the PBR image distribution directly, avoiding photometric inaccuracies in RGB generation and the inherent ambiguity in extracting PBR from RGB. As existing paradigms for cross-modal fine-tuning are not suited for PBR generation due to both a lack of data and the high dimensionality of the output modalities, we propose to train a new PBR model that is tightly linked to a frozen RGB model using a novel cross-network communication paradigm. As the base RGB model is fully frozen, the proposed method retains its general performance and remains compatible with e.g. IPAdapters for that base model.
@article{arxiv.2402.05919,
title = {Collaborative Control for Geometry-Conditioned PBR Image Generation},
author = {Shimon Vainer and Mark Boss and Mathias Parger and Konstantin Kutsy and Dante De Nigris and Ciara Rowles and Nicolas Perony and Simon Donné},
journal= {arXiv preprint arXiv:2402.05919},
year = {2024}
}