We present a new method for multimodal conditional 3D face geometry generation that allows user-friendly control over the output identity and expression via a number of different conditioning signals. Within a single model, we demonstrate 3D faces generated from artistic sketches, portrait photos, Canny edges, FLAME face model parameters, 2D face landmarks, or text prompts. Our approach is based on a diffusion process that generates 3D geometry in a 2D parameterized UV domain. Geometry generation passes each conditioning signal through a set of cross-attention layers (IP-Adapter), one set for each user-defined conditioning signal. The result is an easy-to-use 3D face generation tool that produces topology-consistent, high-quality geometry with fine-grain user control.
@article{arxiv.2407.01074,
title = {Multimodal Conditional 3D Face Geometry Generation},
author = {Christopher Otto and Prashanth Chandran and Sebastian Weiss and Markus Gross and Gaspard Zoss and Derek Bradley},
journal= {arXiv preprint arXiv:2407.01074},
year = {2025}
}
Comments
Added more evaluation since the first version. Accepted to SMI 2025. Computers & Graphics