DocumentCode :
250743
Title :
Single image 3D object detection and pose estimation for grasping
Author :
Menglong Zhu ; Derpanis, Konstantinos G. ; Yinfei Yang ; Brahmbhatt, Samarth ; Zhang, M. ; Phillips, Chris ; Lecce, Matthieu ; Daniilidis, Kostas
Author_Institution :
Dept. of Comput. Inf. Sci., Univ. of Pennsylvania, Philadelphia, PA, USA
fYear :
2014
fDate :
May 31 2014-June 7 2014
Firstpage :
3936
Lastpage :
3943
Abstract :
We present a novel approach for detecting objects and estimating their 3D pose in single images of cluttered scenes. Objects are given in terms of 3D models without accompanying texture cues. A deformable parts-based model is trained on clusters of silhouettes of similar poses and produces hypotheses about possible object locations at test time. Objects are simultaneously segmented and verified inside each hypothesis bounding region by selecting the set of superpixels whose collective shape matches the model silhouette. A final iteration on the 6-DOF object pose minimizes the distance between the selected image contours and the actual projection of the 3D model. We demonstrate successful grasps using our detection and pose estimate with a PR2 robot. Extensive evaluation with a novel ground truth dataset shows the considerable benefit of using shape-driven cues for detecting objects in heavily cluttered scenes.
Keywords :
clutter; edge detection; grippers; image texture; object detection; pose estimation; robot vision; 3D models; 6-DOF object pose; PR2 robot; cluttered scenes; grasping; image contours; pose estimation; single image 3D object detection; texture cues; Computational modeling; Feature extraction; Image edge detection; Image segmentation; Shape; Solid modeling; Three-dimensional displays;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Robotics and Automation (ICRA), 2014 IEEE International Conference on
Conference_Location :
Hong Kong
Type :
conf
DOI :
10.1109/ICRA.2014.6907430
Filename :
6907430
Link To Document :
بازگشت