Camera Localization and 3D Reconstruction

Contact Persons

Patrick Ruhkamp (p.ruhkamp@tum.de), Mahdi Hamad (mahdi.hamad@tum.de), Benjamin Busam (b.busam@tum.de)

Project Coordinator: Benjamin Busam (b.busam@tum.de)

Abstract

The goal of the project is to build a 3D dense surface reconstruction of the operating room using a sparse point cloud extracted with the help of feature-based SLAM such as ORB-SLAM. This is done in an active and dynamic fashion, where we detect changes in the environments and possible defects (i.e. holes, missing parts...), then rescan the scene where we detect these issues (active SLAM) and add it to the reconstruction pipeline. We use multiple cameras in both setups inside-out and outside-in tracking fused with IMU data to collaboratively provide the highest precision possible in an operating room.

Keywords: 3D Computer Vision, 3D Reconstruction, Active SLAM, Collaborative SLAM in the OR

Research Partners

Huawei Technologies, Munich

Siemens AG, Munich

Project Members

Patrick Ruhkamp, p.ruhkamp@tum.de

Mahdi Hamad, mahdi.hamad@tum.de

Benjamin Busam, b.busam@tum.de