Viktor Rudnev

Hi!

My research interests lie primarily at the intersection of Computer Graphics and Computer Vision with applications in VR/AR, Visual Effects and Animation.

I study neural representations of 3D/4D scenes, with focuses

• on combining implicit and explicit elements together to achieve direct controllability of the models (Neural Actor, NeRF-OSR) or improved performance (EventHands, EventNeRF),
• on using alternative sensors, such as event cameras, to bridge the gap and push the boundaries of both traditional and event-based vision research (EventHands, EventNeRF), and
• on using exclusively self-collected and synthetic data, to minimize the amount of required externally collected data and compute resources (Neural Actor, NeRF-OSR, EventHands, EventNeRF).

Publications

Dynamic EventNeRF: Reconstructing General Dynamic Scenes from Multi-view RGB and Event Streams
V. Rudnev, G. Fox, M. Elgharib, C. Theobalt and V. Golyanik
CVPR Workshop on Event-based Vision, 2025
paper project page source code bibtex

Event-3DGS: Event-based Novel View Rendering of Large-scale Scenes using 3D Gaussians
S. Zahid, V. Rudnev, E. Ilg and V. Golyanik
International Conference on 3D Vision (3DV), 2025
paper project page source code

3D Pose Estimation of Two Interacting Hands from a Monocular Event Camera
C. Millerdurai, D. Luvizon, V. Rudnev, A. Jonas, J. Wang, C. Theobalt and V. Golyanik
International Conference on 3D Vision (3DV), 2024
paper project page source code

EventNeRF: Neural Radiance Fields from a Single Colour Event Camera
V. Rudnev, M. Elgharib, C. Theobalt and V. Golyanik
Computer Vision and Pattern Recognition (CVPR), 2023
paper project page source code bibtex

Neural Radiance Fields for Outdoor Scene Relighting
V. Rudnev, M. Elgharib, W. Smith, L. Liu, V. Golyanik and C. Theobalt
European Conference on Computer Vision (ECCV), 2022
paper project page source code bibtex

Neural Actor: Neural Free-view Synthesis of Human Actors with Pose Control
L. Liu, M. Habermann, V. Rudnev, K. Sarkar, J. Gu, C. Theobalt
SIGGRAPH Asia 2021
paper project page source code bibtex

EventHands: Real-Time Neural 3D Hand Pose Estimation from an Event Stream
V. Rudnev, V. Golyanik, J. Wang, H.-P. Seidel, F. Mueller, M. Elgharib and C. Theobalt
International Conference on Computer Vision (ICCV), 2021
Featured in "ICCV 2021 Daily" (link)
paper project page source code bibtex

Education

April 2021 – present:
Ph.D. candidate in Computer Science with Ph.D. advisors Dr. Vladislav Golyanik and Prof. Dr. Christian Theobalt at Saarland University, Saarbrücken, Germany and the Max-Planck-Institut für Informatik

October 2019 – present:
Member of the Graduate School of Computer Science at Saarland University, Saarbrücken, Germany

October 2019 – November 2021:
Doctoral Preparatory Phase/Master Studies in Computer Science at Saarland University, Saarbrücken, Germany

September 2015 – July 2019:
4 year BSc program in Computer Science at Moscow State University, Department of Mathematical Methods of Forecasting

September 2013 – July 2015:
High School student at Kolmogorov’s boarding school of Moscow State University

Recent Positions

August 2024 – October 2024:
Research Intern at Adobe Research, UK (manager: Iliyan Georgiev).

May 2020 – August 2020:
Research Immersion Lab at Graphics, Vision & Video (group leader: Prof. Dr. Christian Theobalt) at the Max-Planck-Institut für Informatik, Germany

September 2017 – July 2019:
Research Intern at the Bayesian Methods Research Group (group leader: Prof. Dr. Dmitry Vetrov), Samsung-HSE Laboratory, Moscow-Seoul

Competition Awards

Hobbies

• Playing music (guitars, bass, keys, drums, cello, violin, glockenspiel, accordion, wind), arranging/recording/mixing
• Digital drawing/painting/animation, photography, storytelling
• DIY: soldering, woodworking, 3D printing, sewing, etc.
• Computer security/hacking/CTF
• Making VR apps/games
• Longboarding, bicycling, table tennis
• Fancy coffee, baking, cooking