Accelerations of a 3D Tracking Method for Non-Rigid Objects

This study aims to track a non-rigidly deformable object in real-time using successive range images. Our approach assumes that the object is globally rigid and partially non-rigid during a short interval between 3D snapshots. The tracking process consists of 3 steps; estimating the rigid transformation by ICP algorithm, extracting the non-rigidly deformed areas, and estimating their deformation vectors by Thin Plate Spline Robust Point Matching (TPS-RPM) algorithm. In our past research, it took mach time for estimating the deformation vectors. Then, in this paper, techniques for accelerating the tracking process are described.

[1]  Francisco Sandoval Hernández,et al.  Real-time template-based tracking of non-rigid objects using bounded irregular pyramids , 2004, 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566).

[2]  H. Tsui,et al.  Geodesic closest point constrained inter-subject non-rigid registration , 2004, ICPR 2004.

[3]  Marc Levoy,et al.  Efficient variants of the ICP algorithm , 2001, Proceedings Third International Conference on 3-D Digital Imaging and Modeling.

[4]  Gérard G. Medioni,et al.  Object modeling by registration of multiple range images , 1991, Proceedings. 1991 IEEE International Conference on Robotics and Automation.

[5]  Paul J. Besl,et al.  A Method for Registration of 3-D Shapes , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Robert T. Schultz,et al.  A unified non-rigid feature registration method for brain mapping , 2003, Medical Image Anal..

[7]  G. Wahba.  Spline models for observational data , 1990 .

[8]  Anand Rangarajan,et al.  A new algorithm for non-rigid point matching , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[9]  Martial Hebert,et al.  Fast 3D tracking of non-rigid objects , 2003, 2003 IEEE International Conference on Robotics and Automation (Cat. No.03CH37422).

[10]  Jorge Rivera-Rovelo,et al.  Non-rigid registration and geometric approach for tracking in neurosurgery , 2004, ICPR 2004.

[11]  Peter Johannes Neugebauer,et al.  Geometrical cloning of 3D objects via simultaneous registration of multiple range images , 1997, Proceedings of 1997 International Conference on Shape Modeling and Applications.