Education

Max Planck Institute for Intelligent Systems / University of Tübingen
Ph.D. Candidate in the Perceiving Systems Department, advised by Michael J. Black and Timo Bolkart.
2020 - present
ETH Zurich
M.Sc. in Visual Computing.
2014 - 2016
Czech Technical University in Prague
B.Sc. in Information and Computer Science with a minor in Mathematics. Graduated with honors.
2011 - 2014

Employment

Amazon
Part-time Research Intern at Amazon Tübingen. Extension a research project from a previous internship. Led by Raja Bala, Hiro Takeda, and Brandon Smith.
Dec 2024 - Present
Tübingen, Germany
Amazon
Research Intern in the field of 3D speech-driven avatars. Led by Raja Bala, Hiro Takeda, and Brandon Smith.
Nov 2023 - Aug 2024
Sunnyvale, CA, USA
Max Planck Institute for Intelligent Systems
Researcher / Ph.D. Candidate in the Perceiving Systems Department. Research in computer vision, machine learning, and computer graphics, with emphasis on human faces. Advised by Michael J. Black and Timo Bolkart.
Nov 2020 - Present
Tübingen, Germany
ETH Zurich / Arbrea Labs
Research Assistant in a joint program between ETH Zurich and Arbrea Labs. Developed augmented reality software for cosmetic surgery preview.
May 2019 - Aug 2020
Zurich, Switzerland
Disney Research
Research Associate working on high-fidelity face capture using deep learning. Led by Thabo Beeler and Derek Bradley.
Mar 2018 - Mar 2019
Zurich, Switzerland
Vizrt Switzerland
Software Engineer working on object tracking and deep learning-based systems. Also previously an intern focused on player detection and tracking for sports broadcasting.
Jan 2017 - Aug 2017
Zurich, Switzerland
ETH Zurich
Research Assistant working on publication and extension of Master’s thesis.
Sep 2016 - Nov 2016
Zurich, Switzerland
Vizrt Switzerland
Software Engineering Intern. Worked on real-time player detection and tracking.
Aug 2015 - Feb 2016
Zurich, Switzerland
CTU Prague - Center for Machine Perception
Research Intern. Worked on feature detector implementation and evaluation in C/C++ and Matlab. Supervised by Prof. Jiří Matas.
Jul 2012 - Jun 2014
Prague, Czech Republic

Publications

THUNDER: Supervising 3D Talking Head Avatars with Analysis-by-Audio-Synthesis Introduces a new form of audio-based self-supervision for 3D talking head avatars.
arXiv, 2025
SMIRK: 3D Facial Expressions through Analysis-by-Neural-Synthesis Uses neural rendering to reconstruct highly accurate 3D facial expressions.
CVPR, 2024
Emotional Speech-driven 3D Body Animation via Disentangled Latent Diffusion Generates emotional co-speech 3D gestures using diffusion models.
CVPR, 2024
EMOTE: Emotional Speech-Driven Animation with Content-Emotion Disentanglement Animates 3D faces with speech while controlling the emotion the animated face communicates.
SIGGRAPH Asia, 2023
EMOCA: Emotion Driven Monocular Face Capture and Animation
R. Daněček, M. Black, T. Bolkart
Reconstructs highly emotional faces from in-the-wild monocular images.
CVPR, 2022
DeepGarment: 3D Garment Shape Estimation from a Single Image Uses CNNs to estimate 3D shape of garments from a single image.
EuroGraphics, 2017