1P1-F01 SIFT-Cloud-Model Generation Method for 6D Pose Estimation and its Evaluation(3D Measurement/Sensor Fusion)
Hideshi Tsubota, Satoshi Kagami, Hiroshi Mizoguchi · The Proceedings of JSME annual Conference on Robotics and Mechatronics (Robomec) · 2011
It is very important for robot to find objects and to estimate their poses, especially for a home service robot which works in human living environment. To achieve that, Nakada et al. proposed SIFT-Cloud-Model(SCM)111, that consists of SIFT features, their 3D positions and eye vector to estimate 6D pose of a target. In this paper, we propose the SIFT-Cloud-Model generation method and evaluate its accuracy, arid computational cost. Finally We developed the efficient way to create SCM and cut down about 60 % computational cost of pose estimation method.