MaPoseEstimation

Chair for Computer Aided Medical Procedures & Augmented Reality
Lehrstuhl für Informatikanwendungen in der Medizin & Augmented Reality

THIS WEBPAGE IS DEPRECATED - please visit our new website

Siemens AG: X-ray PoseNet - Recovering the Poses of Portable X-Ray Device with Deep Learning

Thesis by: Mai Bui
Advisor: Prof. Nassir Navab
Supervision by: Shadi Albarqouni, Slobodan Ilic

Abstract

For most CT setups normally the systems geometric parameters are known. This is necessary to compute an accurate reconstruction of the scanned object. Unfortunately for a mobile CT this might not be the case. However to enable the reconstruction of an object given its projections from unknown geometric parameters, we explored the possibility of using Convolutional Neural Networks to train a model and estimate the necessary geometric parameters needed for tomographic reconstruction. We use the object's CAD model as prior information to create X-ray projections for training the model and introduce an optimized loss function for reconstruction to obtain pose estimates suitable for tomographic reconstruction.

6DoFposeEstimation.png

Location

Siemens AG
Corporate Technology Research & Technology Center
Otto-Hahn-Ring 6
81739 München, Deutschland

Literature

2018
M. Bui, S. Zakharov, S. Albarqouni, S. Ilic, N. Navab
When Regression meets Manifold Learning for Object Recognition and Pose Estimation
Proceedings of IEEE International Conference on Robotics and Automation (ICRA), 21-25 May, 2018, Brisbane, Australia (bib)

Resultant Paper

2017
M. Bui, S. Albarqouni, M. Schrapp, N. Navab, S. Ilic
X-ray PoseNet?: 6 DoF? Pose Estimation for Mobile X-ray Devices
Proceedings of IEEE Winter Conference on Applications of Computer Vision (WACV), Mar 24, 2017 - Mar 31, 2017, Santa Rosa, USA
The first two authors contribute equally to this paper.
(bib)

ProjectForm
Title: Siemens AG: X-ray PoseNet - Recovering the Poses of Portable X-Ray Device with Deep Learning
Abstract: For most CT setups usually the systems geometric parameters are known. This is necessary to compute an accurate reconstruction of the scanned object. Unfortunately for a Mobile CT this might not be the case. However to enable the reconstruction of an object given its projections from unknown geometric parameters, this master thesis explores the possibility of using Convolutional Neural Networks to train a model and estimate the necessary geometric parameters needed for tomographic reconstruction.
Student: Mai Bui
Director: Prof. Nassir Navab
Supervisor: Shadi Albarqouni, Slobodan Ilic
Type: Master Thesis
Area: Machine Learning
Status: finished
Start:  
Finish:  
Thesis (optional):  
Picture:  


Edit | Attach | Refresh | Diffs | More | Revision r1.13 - 29 Mar 2017 - 23:14 - MaiBui