Marcel Bühler
Research Assistant at ETH Zürich
Zurich, Zurich, Switzerland
Summary
👤
Senior🎓
Top SchoolMarcel Bühler is a research-focused computer scientist and data scientist with 11 years of experience developing neural models for high-fidelity 3D avatar reconstruction from sparse smartphone photos. Based at ETH Zürich and currently pursuing a Doctor of Science in Computer Science, he combines statistical and neural priors to resolve fundamental ambiguities in sparse-input reconstruction and achieves novel-view synthesis up to 4K resolution. His internships and research stints at Google and NVIDIA reflect strong industry collaboration and an emphasis on scalable datasets—both real and synthetic—for AR/VR applications like telepresence, entertainment, and fashion. Beyond research, Marcel has entrepreneurial and product experience as a co-founder and has worked across roles from system administration to web development, giving him a pragmatic edge in deploying complex models. A detail that sets him apart: he explicitly draws inspiration from human perceptual priors to design model inductive biases, bridging cognitive insight with state-of-the-art neural methods.
11 years of coding experience
8 years of employment as a software developer
High School Spanish Language and Literature, High School Spanish Language and Literature at Schule BaldeggBachelor’s Degree Computer Science, Bachelor’s Degree Computer Science at University of ZurichDoctor of Science Computer Science, Doctor of Science Computer Science at ETH ZürichUndergraduate Exchange Program Information Technology, Undergraduate Exchange Program Information Technology at University of Technology Sydney
English, Spanish, French, German, Italian
Github Skills (62)
image-synthesis10
neural-rendering10
popup9
generative-adversarial-network9
cyclegan9
generative-model9
deepfake8
computer-vision8
tooltips8
dropdown8
deep-learning8
font-size8
image-manipulation8
computer-graphics8
image-translation8
Programming languages (7)
Github contributions (5)
mcbuehler/VariTex
Mar 2021 - Oct 2021
VariTex: Variational Neural Face Textures, ICCV 2021.
Contributions:14 commits, 4 PRs, 12 pushes in 7 months
aicomputer-visiondeep-learningfacesgan
mcbuehler/DeepSEE
Apr 2020 - Mar 2021
DeepSEE: A novel framework for Deep Disentangled Semantic Explorative Extreme Super-Resolution, ACCV 2020 (oral)
Contributions:23 commits, 2 PRs, 29 pushes in 11 months
pytorchextremedeep-learningdeepseeresolution
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.