The Computer Vision group conducts research on theory and algorithms for the analysis of image-based data. This includes imaging, computer vision, visual-data analytics, and image processing, with a focus on discrete data on multidimensional grids. This can be usual 2d images of standard cameras, volumetric data (3d), video (2d+t), hyper-spectral images (2d+s)) but also include surfaces in 3d (2.5d) or point clouds. Computer Vision and Image Processing include methods from pixel-based “low-level vision”, over object-based “mid-level vision” to scene-based “high-level vision”.
Deep Learning (DL) is one of the most prominent facets of Machine Learning (ML), which in turn refers to data-driven, statistical Artificial Intelligence (AI) methods. DL refers to learning methods implemented using so-called deep neural networks (DNN). Data-driven methods work particularly well when very large amounts of data (Big Data) are available to describe the problem to be solved. Very large data sets in turn require large computational resources (High Performance Computing, HPC).
The Computer Vision group makes use of machine learning methods, develops them further and brings them into application. Methodologically, however, it addresses more than ML for gridded data. Computer vision, visual-data analytics, and image processing also include data modeling and the associated inference of model parameters, use simulation methods to generate data, and can be applied to even the smallest data sets if domain knowledge is available for data modeling. Thus, they are not only "Big Data" but "All Data", not only "Learning" but also "Inference", not only "HPC", but can also target even the smallest computing resources, currently often smartphones but mostly desktop PCs or workstations. However, even if solutions are to be found for the smallest computers, scalable learning and simulation methods on supercomputers are usually necessary for their development.