3D Scene Understanding
|Title:||Master thesis proposal 3D Indoor scene understanding and layout reconstruction for a mobile robot in collaboration with UnimiB|| |
|Description:||The proposed project aims to reconstruct the 3D structural layout of an indoor environment perceived by a mobile robot. From the sensorial data, the robot should be able to reconstruct a geometrical structure of an indoor environment (e.g., an office).
Methods for indoor layout reconstruction must be significantly more tolerant to missing data than their outdoor counterparts, since environments such as offices and apartments exhibit extremely high levels of clutter, which typically results in heavy occlusions of walls and other structures of interest, large-scale artifacts, noise and missing data. The proposed work will be developed in collaboration with IRALAB, the Robotics Lab of University of Milano Bicocca.
The work will be based on an existing project, Free Your Camera (http://www.ira.disco.unimib.it/research/robotic-perception-research/free-your-camera-3d-indoor-scene-understanding-from-arbitrary-camera-motion/) and will be part of a robotic framework based on with ROS and in development at IRALAB.
|Tutor:||FrancescoAmigoni (email@example.com), MatteoLuperto (firstname.lastname@example.org)|
|Students:||1 - 2|
|CFU:||10 - 20|