Marcel C. Buehler

PhD Graduate

Email:

marcel.buehler@inf.ethz.ch

Office:

ETH Zurich, Department of Computer Science, STD, G 28

Address:

Stampfenbachstrasse 48, 8092 Zürich, Switzerland

Biography and Research

I conducted my doctorate in the Advanced Interactive Technologies lab of the Institute for Intelligent Interactive Systems at ETH Zurich, supervised by Prof. Otmar Hilliges, Dr. Thabo Beeler, and Prof. Markus Gross. In August 2025, I defended my thesis titled Faithful 3D Avatars for Everyone: Prior-Guided Face Reconstruction for High-Quality Novel View Synthesis from Casual Few-Shot Captures to an international committee with Prof. Markus Gross (ETH Zurich), Dr. Thabo Beeler (Google), Prof. Fernando De la Torre (Carnegie Mellon University) and Prof. Gordon Wetzstein (Stanford University).

From May 2024 to June 2025, I conducted research as intern at Nvidia's Data-Driven AI for Robotics group. From July 2022 to May 2024, I was a Student Researcher at Google AR/VR. I obtained my Master's degree in Data Science from ETH Zurich in 2020 and my Bachelor's degree in Computer Science from University of Zurich in 2017.

My research enables to reconstruct personalized, high-quality 3D avatars from just a few smartphone photos. 3D avatar reconstruction from sparse inputs requires resolving fundamental ambiguities. Inspired by how humans use prior knowledge, I research how statistical and neural priors can be injected into neural models to facilitate high-quality, faithful reconstructions. Specifically, I develop neural models of 3D avatars trained on large-scale real and synthetic datasets for novel view synthesis up to 4K resolution. These models enable important applications for AR/VR like telepresence, entertainment, and fashion.

Keywords: 3D avatars; sparse inputs; generative models; differentiable rendering; NeRF; 3D Gaussian Splatting; disentanglement.

Links:

Publications

Dream, Lift, Animate: From Single Images to Animatable Gaussian Avatars

Marcel C. Buehler, Ye Yuan, Xueting Li, Yangyi Huang, Koki Nagano, Umar Iqbal

International Conference on 3D Vision (3DV), 2026

Paper

HyperGaussians: High-Dimensional Gaussian Splatting for High-Fidelity Animatable Face Avatars

Gent Serifi, Marcel C. Buehler

arxiv, 2025

Paper

Code

SurFhead: Affine Rig Blending for Geometrically Accurate 2D Gaussian Surfel-based Head Avatars

Jaeseong Lee, Taewoong Kang, Marcel C. Buehler, Minjung Kim, Sungwon Hwang, Junha Hyung, Hyojin Jang, Jaegul Choo

International Conference on Learning Representations, 2025

Paper

Code

Cafca: High-quality Novel View Synthesis of Expressive Faces from Casual Few-shot Captures

Marcel C. Buehler, Gengyan Li, Erroll Wood, Leonhard Helminger, Xu Chen, Tanmay Shah, Daoye Wang, Stephan Garbin, Sergio Orts-Escolano, Otmar Hilliges, Dmitry Lagun, Jérémy Riviere, Paulo Gotardo, Thabo Beeler, Abhimitra Meka, Kripasindhu Sarkar

Conference paper at SIGGRAPH Asia, 2024

Video

Talk

Paper

Data

MagicMirror: Fast and High-Quality Avatar Generation with a Constrained Search Space

Armand Comas, Di Qiu, Menglei Chai, Marcel C. Buehler, Amit Raj, Ruiqi Gao, Qiangeng Xu, Mark Matthews, Paulo Gotardo, Sergio Orts-Escolano, Thabo Beeler

European Conference on Computer Vision, 2024

Paper

ShellNeRF: Learning a Controllable High-resolution Model of the Eye and Periocular Region

Gengyan Li, Kripasindhu Sarkar, Abhimitra Meka, Marcel C. Buehler, Franziska Mueller, Paulo Gotardo, Otmar Hilliges, Thabo Beeler

Computer Graphics Forum (Eurographics), 2024

Paper

Preface: A Data-driven Volumetric Prior for Few-shot Ultra High-resolution Face Synthesis

Marcel C. Buehler, Kripasindhu Sarkar, Tanmay Shah, Gengyan Li, Daoye Wang, Leonhard Helminger, Sergio Orts-Escolano, Dmitry Lagun, Otmar Hilliges, Thabo Beeler, Abhimitra Meka

International Conference on Computer Vision (ICCV), 2023

Video

Paper