» Download Article

Authors
Georg Biegelbauer; Matthias J. Schlemmer; Markus Vincze

Identifier
10.3182/20060906-3-IT-2910.00032

Index Terms
vision systems

Abstract
The most natural approach for a robot to learn about a new object is the short presentation of the object either by hand or on a table. The robot should learn a model of the object and use it to later find the object again and track it. All these steps should be executed autonomously. One of our long-term goals is the usage of our system for robot grasping tasks. Following this approach we developed a method of extracting the object model using a depth image acquired through the scan of the object. The model extracted is subsequently exploited for detecting the object in the environment. After detection the approach automatically initializes a tracking method that follows the object motion to enable grasping or navigation tasks. The autonomous execution is made possible by the integration of depth and appearance data using a laser-based depth camera and a color camera. The use of both depth and colour images makes the approach robust to illumination changes and different conditions during learning the object model and then later re-detecting it. Experiments show the feasibility of the concept even in situations where the object is partially occluded and the scene is cluttered.

References

[1] Biegelbauer, G. and M. Vincze (2005). Fast and
    robust 3D object detection using a simplified
    superquadric model description. Proceedings
    of the 7th Conference on Optical 3-D Measurement
     Techniques 2, 222-230.
[2] Fischler, M.A. and R.C. Bolles (1981). Random
    sample consensus: A paradigm for model fitting
     with applications to image analysis and
    automated cartography. Communications of
    the ACM 24, 381-395.
[3] Jang, H.Y., H. Moradi, S. Lee and J. Han (2005).
    A visibility-based accessibility analysis of the
    grasp points for real-time manipulation. In:
    Proceedings of the IEEE/RSJ International
    Conference on Intelligent Robots and Systems
    (IROS 2005). Edmonton, Alberta, Canada.
[4] Kim, S., I. Kweon and I. Kim (2003). Robust
    model-based 3d object recognition by combining
     feature matching with tracking. Proceedings
     of the IEEE International Conference
     on Robotics and Automation 2, 2123-
    2128.
[5] Kragic, D. and H.I. Christensen (2002). Model
    based techniques for robotic servoing and
    grasping. Proceedings of the IEEE/RSJ International
     Conference on Intelligent Robots
    and System 1, 299-304.
[6] Leonardis, A. and A. Jaklic (1997). Superquadrics
    for segmenting and modelling range data.
    IEEE Transactions on Pattern Analysis and
    Machine Intelligence 19(11), 1289-1295.
[7] Lu, C.P., G.D. Hager and E. Mjolsness (2000).
    Fast and globally convergent pose estimation
    from video images. IEEE Transactions on
    Pattern Analysis and Machine Intelligence
    22(6), 610-622.
[8] Mikolajczyk, K. and C. Schmid (2002). An affine
    invariant interest point detector. Proceedings
    of the 7th European Conference of Computer
    Vision 1, 128-142.
[9] Moré, J.J. (1977). The Levenberg-Marquardt Algorithm:
     Implementation and Theory. Numerical
     Analysis - Lecture Notes in Mathematics,
     Springer Verlag 630, 105-116.
[10] Mukherjee, S. and S.K. Nayar (1995). Automatic
     generation of grbf networks for visual learning.
      Proceedings of the IEEE International
     Conference on Computer Vision pp. 794-800.
[11] Parhami, B. (1994). Voting algorithms. Machine
     Learning (IEEE Transactions on Reliability)
     43(4), 617-629.
[12] Salganicoff, M., L.H. Ungar and R. Bajcsy (1996).
     Active learning for vision-based robot grasping.
      Machine Learning (Kluwer) 23(2), 251-
     278.
[13] Solina, F. and R. Bajcsy (1990). Recovery of parametric
      models from range images: The case
     for superquadrics with global deformation.
     IEEE Transactions on Pattern Analysis and
     Machine Intelligence 12(2), 131-147.
[14] Taylor, G. and L. Kleeman (2004). Integration
     of robust visual perception and control for
     a domestic humanoid robot. Proceedings of
     the IEEE/RSJ International Conference on
     Intelligent Robots and Systems 1, 1010-1015.
[15] Yoon, Y., A. Kosaka, J.B. Park and A.C.
     Kak (2005). A new approach to the use
     of edge extremities for model-based object
     tracking. Proceedings of the IEEE International
      Conference on Robotics and Automation
      pp. 1883-1889.