Helge
Rhodin,
Computer
Vision
Laboratory
Ecole
polytechnique
fédérale
de
Lausanne
With the aim of enabling rich, yet natural and non-intrusive human-computer interaction, I work at the intersection of computer graphics and machine learning-based computer vision. My work is based on processing plain video input to form dynamic reconstructions of our everyday environment. I apply these reconstructions to interactive animation and mobile augmented reality, thus improving graphical output and enriching the information flow from computer to human. In turn, the real-time human motion capture methods provide a natural input modality for augmented reality (AR). Both research directions will pave the way for a worthwhile pervasive AR future that assists our everyday life. Furthermore, they have applications in sports injury prevention, performance analysis, and distant fields such as neuroscience.
I will be talking about my fundamental algorithmic contributions (presented at SIGGRAPH, SIG Asia, TOG, ICCV, ECCV, and CVPR) and their impact on gesture-driven character animation (EG and SIG Asia), ski performance analysis (ICSS), and telepresence in virtual reality (IEEE VR). Moreover, I will outline my goal of replacing hand-crafted CG and CV models with learned ones and how I plan to address this challenge methodologically through self-supervised training of deep neural networks and my concept of World Representation Learning.
Bio: Helge Rhodin is a lecturer at EPFL and postdoctoral researcher in Pascal Fua's Computer Vision Lab. His research interests range from computer graphics and augmented reality, over 3D computer vision, to machine learning.
Helge received the BSc and MSc degree in computer science from Saarland University, respectively supervised by Philipp Slusallek and Sebastian Hack, and Michael Breuß and Joachim Weickert. He graduated with a PhD in 2016, with his thesis on Motion Capture for Interactive Virtual Worlds, mentored by Christian Theobalt and Hans-Peter Seidel at the Max-Planck Institute for Informatics.