 |
    |
Technical Demonstration on Model Based Training, Detection and Pose Estimation of Texture-Less 3D Objects in Heavily Cluttered Scenes
Stefan Hinterstoisser1, Vincent Lepetit2, Slobodan Ilic1, Stefan Holzer1, Kurt Konolige3, Gary Bradski3, and Nassir Navab1
1Department of Computer Science, CAMP, Technische Universität München (TUM), Germany
hinterst@in.tum.de
slobodan.ilic@in.tum.de
holzers@in.tum.de
navab@in.tum.de
2Ecole Polytechnique Federale de Lausanne (EPFL), Computer Vision Laboratory, Switzerland
vincent.lepetit@epfl.ch
3Industrial Perception Inc., USA
kurt@industrial-perception.com
gary@industrial-perception.com
Abstract. In this technical demonstration, we will show our framework of automatic modeling, detection, and tracking of arbitrary texture-less 3D objects with a Kinect. The detection is mainly based on the recent template-based LINEMOD approach [1] while the automatic template learning from reconstructed 3D models, the fast pose estimation and the quick and robust false positive removal is a novel addition.
In this demonstration, we will show each step of our pipeline, starting with the fast reconstruction of arbitrary 3D objects, followed by the automatic learning and the robust detection and pose estimation of the reconstructed objects in real-time. As we will show, this makes our framework suitable for object manipulation e.g. in robotics applications.
LNCS 7585, p. 593 ff.
lncs@springer.com
© Springer-Verlag Berlin Heidelberg 2012