Picture of me :))
Univ.-Prof. Dr.-Ing.  Darius Burschka

Technische Universität München
Lab for Robotics and Embedded Systems (I6)
Boltzmannstr. 3
D-85748 Garching bei München

Phone: (+49 89) 289-18133
Fax
:      (+49 89) 289-18107
e-mail: burschka@cs.tum.edu          
                    My public PGP-key

NavigaTUM
Vision-Based Navigation Group

Joint appointments:

German Aerospace Center (DLR), Institute for Robotics and Mechatronics,

News:

Doktorandenstellen verfügbar im Bereich
vision-based navigation
dynamic scene segmentation
image classification
Diplomarbeiten auf Anfrage im Bereich
Medizinische 3D Rekonstruktion
Shape-from-Shading
Image Retrieval
VSLAM

Research interests:

My principle areas of research are sensors for mobile systems and Human Computer Interfaces.
The focus of my research is on vision-based navigation and three-dimensional reconstruction from sensor data.

Teaching:
Wintersemmester 2006/2007
Vorlesung
Robotics
Principles of Computer Vision
Sommersemester 2007
Vorlesung
Robot Motion Planning

Hauptseminar
Sensorbasierte Generierung von 3D-Umgebungsmodellen


Curriculum Vitae

my publications


3D Exploration of Indoor Environments (32MB- MPEG)

Human Computer Interfaces (8MB MPEG)  


Quick links

 Research projects             Online Documents            Publications           Useful links
 Private travel page
 

Download Page



Research projects:
                     Human Computer Interaction based on XVision
(left) interactive "pong" game, where the user plays with his hands (right) gesture recognition

(see example MPEG movie for the gesture recognition- it starts with a teaching step, where the system requests specific gestures followed by the recognition step showing the currently grabbed image in the upper small window and the matched template in the lower one)
 

System presentation: see ZDTV report

               Examples (Applications) of Human Computer Interaction
 
  • Manipulation of Artifical Objects (MPEG) (grasping of objects on the screen) - after an initial teaching phase the system allows manipulations of artifical objects
  • Vision-Based Interfaces(MPEG)(calculator application)
  •                             
  • Color based feature selection + tracking 
    (MPEG from ICRA99 in Detroit)
    (MPEG of color tracking based on XVision)
    Laser-Based Navigation based on  XVision and 
     
     

     HTML-slides about Laser-Based Map Generation

    (see example(MPEG)  of Visual Servoing method for mobile navigation)

    Vision-based control
    work together with UIUC group

    Positive (objects) and negative (gaps) obstacle detection from stereo vision.
    One of the original images (left), calculated disparity image (right) and detected obstacle (bottom).
    In the MPEG movie  the box is recognized as an obstacle, but not the newspaper presented later.
    Vision-based obstacle avoidance used for safe navigation of the mobile robot ( MPEG movie )
                         
    Omnidirectional navigation for mobile systems. 
       - Vision Based Navigation
       - Virtual Joystick ( MPEG movie )

               
    Vision-Based Control for Robocup
    A few Robocup impressions from Seattle 2001


    PhD Thesis: "Videobasierte Exploration von Innenräumen am Beispiel eines binokularen Stereo-Kamerasystems",
                             Reihe 10, Nr. 573, VDI Verlag, 1999
                             (engl. "Vision-Based Exploration of Indoor Environments at an Example of a Binocular Stereo System")

             HTML-slides about my Thesis (engl.)

    Pages about Vision-Based Exploration


    Dario's  US-Travel Page

     Sailing in Greece ('97)


    Useful links: