Drucken
Univis
Search
 
FAU-Logo
Techn. Fakultät Willkommen am Institut für Informatik FAU-Logo

3D-Kameras in der Medizintechnik [SEM-3D-CAM]

Background

3-D cameras like Time-of-Flight (ToF) sensors or Microsoft's Kinect are becoming increasingly popular with potential applications in security, automobile, or consumer industry. Recent advances in 3-D camera technology have also opened new perspectives for its use in medical engineering. In particular, the resolution (40k-300k points), framerate (25-60 Hz) and metric depth accuracy (up to 1 mm) hold potential for medical applications.

The pattern recognition lab (LME) is actively involved in the establishment of 3-D cameras in the field of medical engineering. Research activities at the LME include, among others, 3-D endoscopy, augmented reality in an intra-operative environment, or patient positioning and respiratory motion management in radiotherapy.

Contents

In this seminar, we will discuss general issues, challenges and promising applications that arise with the use of 3-D cameras in the field of medical engineering. In particular, this seminar covers:

  • High performance preprocessing (GPU/CPU)
  • Multi-modal surface registration and feature descriptors
  • Object classi fication and recognition
  • Real-time visualization and data fusion

Additional topics may be provided on demand. As programming will be an important part of this seminar, participants should have established programming skills in C/C++. Furthermore, it is advantageous for participants to have attended lectures on image processing, pattern recognition or computer graphics. A tutorial on the LME framework for 3-D cameras and a refresher on image processing theory will be held at the beginning of the seminar.

Meetings

05/04/11    Kickoff / presentation of projects

Software

05/13/11 Framework

05/16/11 KinectPlugin

05/16/11 Framework Fix

06/24/11 Framework Update

Information / Dates

Credits: 5 ECTS (2/4 SWS)
Kick-off meeting: 4 May, 2011, 5-6 pm, room 09.150 [1]

[1] Early registrations are appreciated.

Contact

Supervision: Jakob Wasza, Sebastian Bauer

Contact : jakob.wasza(at)informatik.uni-erlangen.de, sebastian.bauer(at)informatik.uni-erlangen.de


Figure 1: Top left: ToF 3-D camera. Top right: Kinect 3-D camera. Bottom: 3-D grayscale (left) and depth data (right) from an anthropomorphic torso phantom.