PhD in Computer Vision : 3D model registration to merge virtual reality and SLAM.
3D model registration to merge virtual reality and SLAM
embedded vision systems, real time processing, localization, mapping, 3D modeling, registration, augmented reality, SLAM, navigation
This thesis aims to address the challenge to provide a fully automated real-time fusion of data acquired by sensors (camera, lidar,...) with a virtual 3D model of indoor or outdoor buildings.
The key issue is that both acquired sensor data and 3D models contain erroneous data and moreover do not store the same information (geometric, texture, semantic,...). We will investigate the most suitable way to represent to data and its adequation to the registration method in order to reach real-time processing. Power consumption will also be taken into account, as the developed method is dedicated to be run on embedded systems.
The expected results will be presented in several domains: augmented reality, autonomous navigation, 3D modeling.
The doctorant will be a member of the IRSTV and he/she will interact with IRSTV partners in the outdoor application context.
The doctorant will also interact with the ADAPT project international partners. We also expect to collaborate with University of Auckland, a long-time partner of the Vaader Team on embedded vision systems.
Team VAADER of IETR, Rennes, on the INSA Rennes site.
Mail : email@example.com
Tel. : +33 (0)2 23 23 84 62
3-years contract on the basis of a monthly gross income of 1 760 euros approximatively. Part-time teaching can be considered. Start in automn 2018.
Research master or engineer degree in computer/ image processing /computer vision /math
(c) GdR 720 ISIS - CNRS - 2011-2018.