Navigability Graph Extraction From Large-Scale 3D Point Cloud

Imeen Ben Salah (1) , Sebastien Kramm (1) , Cédric Demonceaux (2) , Pascal Vasseur (1, 2)

1 LITIS - Laboratoire d'Informatique, de Traitement de l'Information et des Systèmes
2 VIBOT - VIsion pour la roBOTique [VIBOT CNRS ERL 6000]

This paper presents a novel method for summarizing and compression of large-scale 3D models into compact spherical representations. The information is combined into a set of optimized spheres in order to facilitate its use by systems with limited resources (smartphones, robots, UAVs, ...). This vision-based summarizing process is applied in a fully automatic way using jointly photometric, geometric and semantic information of the studied environment. The main contribution of this research is to provide a navigability graph that maximizes the significance of the contents of its nodes while maintaining the full visibility of the environment. Experimental results in summarizing large-scale 3D map demonstrate the feasibility of our approach and evaluate the performance of the algorithm.

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]

Imeen Ben Salah, Sebastien Kramm, Cédric Demonceaux, Pascal Vasseur. Navigability Graph Extraction From Large-Scale 3D Point Cloud. 21st IEEE International Conference on Intelligent Transportation Systems, Nov 2018, Maui, Hawaii, United States. ⟨10.1109/ITSC.2018.8569447⟩. ⟨hal-01929921⟩