Our CPhoto@FIT research group is exploring ways to estimate the orientation of a camera from an image and to determine the exact position of the photographer. It focuses in particular on natural environments – where researchers must contend with additional challenges: the vast scale of the area, low density of photographs, the similarity of the landscape, or the absence of recognisable landmarks, in contrast to, for example, urban environments. In this presentation, they will introduce methods of visual geolocation and camera orientation estimation, which aim to determine the position and/or orientation of the camera solely on the basis of visual information encoded in the image. Despite the high level of complexity involved, this approach yields very promising results and has a wide range of applications. You will have the opportunity to try out some of them for yourself.