A Deep Learning approach to identify and localize room assets using handheld RGB-D sensors

With the availability of low-cost edge devices equipped with color and depth sensors, such as iPhone or iPad, 3D data capturing is becoming more accessible and convenient. Asset managers and building owners seek benefiting from this potential to accurately and automatically create 3D indoor models of buildings. In particular, having 3D indoor models containing objects of interest enables several applications, such as asset inventory and maintenance management. However, the automatic and accurate generation of such 3D models introduced many challenges, and the current methods exposed several limitations. This research project aims to propose and implement a framework, which employs various state-of-the-art computer vision techniques, to overcome the current limitations.

Majid Seydgar
Superviseur universitaire: 
Ali Motamedi;Érik Andrew Poirier
Partner University: