Autonomy starts with awareness of the environment. Robots are given autonomy using sensors that endow them with perceptual capabilities, such as cameras . Recently, a new type of camera working under the Time-of-Flight principle has been developed, capable of acquiring dense depth maps at high frame rates. Its small size and weight make it suitable for its use on-board a flying vehicle for indoor localization and mapping . This document outlines the first approaches taken in the use of a ToF camera for such tasks constrained by real-time requirements. The camera has been mounted on a flying vehicle that uses the open source Paparazzi autopilot system developed by the ENAC ( Ecole National de l'Aviation Civile ) french team. Since indoor environments are predominantly planar, planar patches have been favoured to model the environment and detect motion of the UAV. A Region Growing segmentation algorithm identifies and extracts planes from the scene in real-time. Planes are tracked and registered across a sequence of frames to estimate the camera's ego-motion . Initial results of plane-based visual odometry are presented and confirm the device suitability.