ICPIK: Inverse Kinematics Based Articulated-ICP

Shachar Fleishman, Mark Kliger, Alon Lerner, Gershom Kutliroff; Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2015, pp. 28-35

In this paper we address the problem of matching a kinematic model of an articulated body to a point cloud obtained from a consumer grade 3D sensor. We present the ICPIK algorithm - an Articulated Iterative Closest Point algorithm based on a solution to the Inverse Kinematic problem. The main virtue of the presented algorithm is its computational efficiency, achieved by relying on inverse-kinematics framework for analytical derivation of the Jacobian matrix, and the enforcement of kinematic constraints. We demonstrate the performance of the ICPIK algorithm by integrating it into a real-time hand tracking system. The presented algorithm achieves similar accuracy as state of the art methods, while significantly reducing computation time.