The Computational Photography Group of Brno University of Technology is a part of the Faculty of Information Technology. Our research activities focus on HDR image and video processing, image and video quality assessment, computational aesthetics, image registration, and human visual perception. We offer several courses on computational photography and advanced image processing.
Computational photography evolved from computer graphics, image
processing, computer vision, and optics to enhance or extend the capabilities of current
photography and display technology.
A typical example of computational photography technique is High Dynamic Range imaging (HDR). Using multiple exposures, HDR overcomes the issue of limited sensor contrast and enables to capture and reproduce the reality faithfully.
Visual geo-localization exploits visual information contained in multimedia to estimate camera pose. We use data from geographic information systems (GIS) for cross-domain alignment of a photo and a terrain rendered from a digital elevation model (DEM). This makes advanced computational photography techniques possible, e.g. automatic annotation of mountain panoramas, organization and management of large scale datasets of photographs, visualization and presentation of photos in the map and/or in virtual and augmented reality, historical rephotography, or fully-automated image enhancement.