My research interests are at the intersection between computer vision, computer graphics and machine learning. I am particularly interested in analyzing and modeling people from all sorts of sensor data: IMUs, 3D/4D scans, RGBD, images or video. Current computer vision algorithms can detect people in images or estimate 2D pose to a remarkable accuracy. However, people are much more complex. Humans perceive lots of information from other humans; for example we sense other people's emotional state based on facial expressions and body movements, or we make guesses about people's preferences based on what clothing they wear. In order for machines to interact with humans they have to both perceive all this information from sensory data and they have to "appear" human to us. My research revolves around perceiving and building digital generative models of people from real people. In particular research topics are:
- Analyzing people in images
- Visual inference of human pose and shape
- Human body shape, clothing and motion modeling using deep learning
- Deep learning
- Generative models of people
- Human motion capture and performance capture (3D/4D reconstruction)
- Mesh registration
Deep Inertial Poser Learning to Reconstruct Human Pose from SparseInertial Measurements in Real Time
in ACM Transactions on Graphics, (Proc. SIGGRAPH Asia), vol. 37, no. 6, 185:1-185:15 2018.
Sparse Inertial Poser: Automatic 3D Human Pose Estimation from Sparse IMUs
in Computer Graphics Forum 36(2), Proceedings of the 38th Annual Conference of the European Association for Computer Graphics (Eurographics), 349-360 2017.
Best Paper Award
PCA-enhanced stochastic optimization methods
in German Conference on Pattern Recognition (GCPR), 2012.
4D Cardiac Segmentation of the Epicardium and Left Ventricle
in World Congress of Medical Physics and Biomedical Engineering (WC), 2009.
Parametric Modeling of the Beating Heart with Respiratory Motion Extracted from Magnetic Resonance Images
in IEEE Computers in Cardiology (CINC), 2009.