Current research / employment

Time-of-Flight (ToF) cameras provide a direct way to acquire 3-D surface information of objects in real-time. More recently, applications like gesture recognition or automotive passenger classification are using ToF cameras. These cameras have several advantages over other 3-D surface acquisition techniques like stereo vision or structured light techniques. The most promising advantage is that ToF is on its way to become a component of mass markets like consumer electronics and the automotive industry. Currently a high-end ToF camera is available for about USD 7,000. But in the near future, a target price for ToF cameras of a few hundred dollars can be expected. The architecture of ToF cameras enables a high portability of the system and a variety of integration prospects in existing systems. Recent ToF cameras provide data rates up to 40 frames per second with an lateral resolution of up to 200 x 200 pixels. Each of these 3-D points provides precise metric information in the camera coordinate system.

The first main contribution of this work is the introduction of a system that uses Time-of-Flight (ToF) technology to acquire a real-time multidimensional respiratory signal from a 3-D surface reconstruction of the patient's chest and abdomen without the use of markers. An advantage of ToF sensors is that their high lateral resolution (40.000 3-D points) makes it possible to define multiple regions of interest to compute an anatomy adaptive multidimensional respiratory signal. The proposed system can be used e.g. in gated radiotherapy, where a respiratory signal is used to irradiate moving tumors more precisely. We evaluated our approach by comparing a ToF based respiratory signal with the signal acquired by an commercial available external gating system and achieved an average correlation coefficient of 0.88.

Based on the same technology (ToF), a system to correct the position of a patient in respect to a prior acquired reference surface is the second main contribution of this work. ToF cameras enable to accurately compute translation and rotation of the patient in respect to a reference surface. We are using an Iterative Closest Point (ICP) algorithm to determine the 6 degree of freedom (DOF) displacement vector. Current results show that for rigid phantoms it is possible to obtain an accuracy of about 1 mm. This results is competitive to commercial available positioning systems. Potential applications for this system can be found within radiotherapy or multimodal image acquisition with different devices.

Publications J. Penne, S. Soutschek, M. Stürmer, C. Schaller, S. Placht, J. Kornhuber, J. Hornegger, Touchscreen ohne Touch - Berührungslose 3D Gesten-Interaktion für den Operationssaal (Touchscreen without Touch - Touchless 3D Gesture Interaction for the Operation Room), i-com - Zeitschrift für interaktive und kooperative Medien, 8(1), 19-23, (2009)

T. Ringbeck, M. Profittlich, C. Schaller, Kameras für die dritte Dimension, Optik & Photonik, 3/2009, 30-33, (2009)

J. Penne, C. Schaller, J. Hornegger, T. Kuwert, Robust real-time 3D respiratory motion detection using time-of-flight cameras, Computer Assisted Radiology and Surgery, 3(5), 427-431, (2008)

C. Schaller, J. Penne, J. Hornegger, Time-of-flight sensor for respiratory motion gating, Medical Phys., 35(7), 3090-3093, (2008)

J. Han, B. Berkels, M. Droske, J. Hornegger, M. Rumpf, C. Schaller, J. Scorzin, H. Urbach, Mumford-shah model for one-to-one edge matching, IEEE Transation on Image Processing, 16(11), 2720-2732, (2007)