CARTOGRAPHIE ET INTERPRÉTATION DE L'ENVIRONNEMENT PAR DRONE.

This article presents the processing chain for geometric and semantic mapping of a UAV environment that we developed at ONERA/DTIM. A precise 3D modelling of the environment is computed using video and (if available) Lidar data captured during the UAV flight. Then semantic mapping is performed by interactive learning on the model, thus allowing generic object detection. Finally, event detection and moving-object tracking are performed in the video stream and localized in the 3D model, thus giving a global view of the situation. [ABSTRACT FROM AUTHOR]