The Chair of Photogrammetry and Remote Sensing at TUM is well known for research on models, algorithms, and methologies for mapping, analysing, and fusing data from urban areas.
Developed approaches address the processing chain ranging from signal processing and precise pose estimation by imaging sensors to surface reconstruction and change detection to semantic modeling for automated interpretation. The data considered are recorded by passiv sensors operating in the visible or in the thermal infrared domain and active sensors systems like LiDAR or synthetic aperture radar (SAR). The level of detail mapped by the sensors vary depending on the used platform operating spaceborne (satellites), airborne (aircrafts, helicopters, UAVs) or terrestrial (mobile mapping systems).
Application examples regarding urban areas consider on one side the buildings (e.g. building reconstruction from LiDAR or SAR, texuring of building models using thermal infrared image sequences, enrichment of data bases by facade interpretation, change detection of urban structures) and on the other side the traffic (e.g. detection of vehicles from visible and infrared image sequences, from dense laser point clouds or from ultra high resolution SAR imagery). Beside the mentioned applications focused on urban areas some projects extend the working space in Photogrammetry and Remote Sensing. Application examples are coming from foresty and glaciology or close range domain addressing e.g. pose estimation of robots for photogrammetric surface reconstruction (automotive industry) or patient positioning for image guided therapy (medical application).