My name is Frederik Hagelskjaer. I'm a PostDoc at University of Southern Denmark (SDU).
My main interests are computer vision and deep learning. I work with both 2D and point cloud data, and my main interest is pose estimation. However, I also work with many other data processing methods.
I am currently looking for different ways to process very large point cloud data using neural networks. I you have any interesting ideas just contact me at:
frhag AT mmmi DOT sdu DOT dk
One of my current projects is training deep neural networks for pose estimation using only synthetic data. Check out the paper:
Bridging the Reality Gap for Pose Estimation Networks using Sensor-Based Domain Randomization
A visualization of the method in action is shown below: