Object 6DOF pose estimation using CAD models -- Kinect --

asked 2014-09-03 04:07:38 -0600

Naomii gravatar image

updated 2014-09-03 04:08:38 -0600

I would like to get 6DOF-coordinate (x, y ,z, yaw, pitch, roll) of the object, ex. Lego brick, and set direction from centroid for grasping from any point of view.

i have read and tried to follow this paper below but not understood how to implement it correctly.

  • M. Ulrich, C. Wiedemann, and C. Steger. CAD-based recognition of 3d objects in monocular images. International Conference on Robotics and Automation (ICRA), 2009.

please suggest me any good tutorials or the approach that explain how everything works.

Thank you

edit retag flag offensive close merge delete

Comments

1

Altough OpenCV does have some 3D processing available its largest focus is 2D image processing for the moment. If you would like to use CAD models in a 3D space to look for the degrees of freedom, which is basically a 3D shape matching you want to perform, you will be better of using ROS or PCL ...

StevenPuttemans gravatar imageStevenPuttemans ( 2014-09-03 06:43:44 -0600 )edit
add a comment