Professor, Technical University of Munich
6 papers at NeurIPS 2025
We introduce a text-to-intrinsic (albedo, rough, metallic, normal) model, which can be used rerendered under arbitrary lighting condition or even distilled into room-scale PBR textures.
We introduce ROGR, a novel approach that reconstructs a relightable 3D model of an object captured from multiple views, driven by a generative relighting model.
We present the first zero-shot PBR texture super-resolution framework that outputs high-quality PBR textures from low-resoltuion PBR input.
LiteReality is an automatic pipeline that converts RGB-D scans of indoor environments into graphics-ready 3D scenes, featuring high-quality meshes with realistic PBR materials and articulated objects.
Photorealistic relightable head avatars with a new OLAT dataset of faces.
