The Mathematical Imaging Group is a research group within the Centre for Mathematical Sciences. We are developing new mathematical methods for problems arising in imaging and computer vision. In terms of mathematical tools, we are currently focused on optimization methods, polynomial equations, inverse problems, large scale computing and graph theory. We address problems in geometric computer vision, medical image analysis, cognition and image analysis. We are frequently collaborating with academic partners, industry and society at large.
Software release: STRUDLWe recently released STRUDL, a software program for automatically creating world coordinate trajectories of road users from surveillance videos, based on a neural network object detector and tracking. It...
In november we will give a new undergraduate course in Machine Learning at the Centre for Mathematical Sciences. The interest from students is overwhelming - 180 students have signed up for the course.
A new project "Semantic mapping and visual navigation for smart robots" was recently granted 31 MSEK by the Swedish Foundation for strategi research.