@albangauthier
Research Scientist @ Solaya. Appearance modeling, rendering, novel view synthesis, generative AI. Previously postdoc at INRIA Sophia-Antipolis in GraphDeco. PhD from Telecom Paris. Interned at Adobe Research. https://albangauthier.github.io
For now, I only uploaded the weights of the SVBRDF estimation network with or without geometry buffers (depth+normals) called "UNet-HF", but I'll add more checkpoints if I see some interest, e.g. to reproduce the results from this table on the Interioverse test set:
The code allows to generate a PBR textured asset using text prompts, leveraging ControlNets, an SVBRDF estimation network, and a classical multiview texturing technique implemented in Meshlab. Here's a example of what you can achieve (visualized in Blender):
The implementation of our paper called "An evaluation of SVBRDF Prediction from Generative Image Models for Appearance Modeling of 3D Scenes" is now available at:
github.com/graphdeco-in...
Project Page:
repo-sam.inria.fr/nerphys/svbr...
The SIGGRAPH Physically Based Shading course is back this year: blog.selfshadow.com/publications...
We are very excited to announce our work "On-the-fly Reconstruction for Large-Scale Novel View Synthesis from Unposed Images", to be presented at #SIGGRAPH2025!
Learn more: repo-sam.inria.fr/nerphys/on-t...
1/3
Excited to share our paper, "Splat and Replace: 3D Reconstruction with Repetitive Elements", which I'll be presenting at #SIGGRAPH2025!
Project page: repo-sam.inria.fr/nerphys/spla...
1/3
