Claudia Nieuwenhuis has joined our Vision team on a postdoctoral fellowship within ICSI's German visiting agreement, which is funded through the FITweltweit program, sponsored by the BMBF and administered by the DAAD, the German Academic Exchange Program.

Claudia received a master's degree in computer science from TU Ilmenau, Germany, and a master's degree in mathematics from the University of Hagen, Germany. Her PhD, also in computer science, is from the University of Heidelberg. For her thesis, she developed methods for optical flow estimation and validation. The problem was to analyze multiple images in a sequence, as in a video, to estimate motion - what is moving, in what direction, and how fast. Applications include automatic car brakes, which engage when an object, such as a pedestrian, is detected and its motion is estimated to put it in the way of the car.

After receiving her PhD in 2009, she took a postdoctoral fellowship in Munich, where she worked on problems in computer vision using continuous optimization methods. While discrete optimization is widely used in computer vision problems, continuous optimization has received less attention. In continuous optimization problems, images are analyzed continuously, and attention moves to the level of the pixel only at the end of the calculation. Claudia used these methods for image segmentation (separating an image into its constituent parts); object detection (identifying what objects are); and scene understanding (analyzing the general content of an image, for example, the type of scene). Here at ICSI, she's interested in whether her work in this area could be applied to video analysis.

Claudia is interested in modern dance, ballet, choir, and the violin. She also plays squash and tennis and enjoys yoga and climbing.