The campus-wide BioImage Geek Seminar Series hosts distinguished experts speaking about exciting methodological advances in the fields of microscopic imaging, image reconstruction, image analysis, computer vision and machine learning in the context of cell biology.
Our next speaker will be Björn Andres from the Computer Vision and Multimodal Computing department at the Max Planck Institute for Informatics, Saarbrücken:
Host: Carsten Rother, Computer Vision Lab Dresden.
Everybody is welcome.
Phenomenal progress in microscopy has greatly improved biologists' ability to image the structures and processes of life. Today, much biological insight lies hidden in big image data, too large to be processed entirely by hand. Mathematical models for computational and computer-assisted biological image analysis are the subject of this talk. Specifically, it is demonstrated by applications in segmentation, tracking and reconstruction how combinatorial optimization techniques facilitate the automated and assisted analysis of biological images by allowing biologists to incorporate prior knowledge and hard constraints. It is also discussed how knowledge obtained from a small amount of hand-processed data can be transferred to a large amount of unlabeled data by machine learning techniques.