3D point pattern matching based on spatial geometric flexibility

We propose a new method for matching two 3D point sets of identical cardinality with global similarity but local non-rigid deformations and distribution errors. This problem arises from marker based optical motion capture (Mocap) systems for facial Mocap data. To establish one-to-one identifications, we introduce a forward 3D point pattern matching (PPM) method based on spatial geometric flexibility, which considers a non-rigid deformation between the two point-sets. First, a model normalization algorithm based on simple rules is presented to normalize the two point-sets into a fixed space. Second, a facial topological structure model is constructed, which is used to preserve spatial information for each FP. Finally, we introduce a Local Deformation Matrix (LDM) to rectify local searching vector to meet the local deformation. Experimental results confirm that this method is applicable for robust 3D point pattern matching of sparse point sets with underlying non-rigid deformation and similar distribution.

[1]  Robert T. Schultz,et al.  Registration of Cortical Anatomical Structures via Robust 3D Point Matching , 1999, IPMI.

[2]  Michael Gleicher,et al.  Animation from observation: Motion capture and motion editing , 1999, COMG.

[3]  Dinesh Manocha,et al.  OBBTree: a hierarchical structure for rapid interference detection , 1996, SIGGRAPH.

[4]  Gino van den Bergen.  Efficient Collision Detection of Complex Deformable Models using AABB Trees , 1997, J. Graphics, GPU, & Game Tools.

[5]  David Doermann,et al.  Robust Point Matching for Non-Rigid Shapes: A Relaxation Labeling Based Approach , 2004 .

[6]  James T. Kajiya,et al.  Ray tracing complex scenes , 1986, SIGGRAPH.

[7]  Anand Rangarajan,et al.  A new point matching algorithm for non-rigid registration , 2003, Comput. Vis. Image Underst..

[8]  E. H. Furnée,et al.  Early detection of Parkinson's disease through automatic movement evaluation. , 1998, IEEE engineering in medicine and biology magazine : the quarterly magazine of the Engineering in Medicine & Biology Society.

[9]  Jürgen Weese,et al.  Point-Based Elastic Registration of Medical Image Data Using Approximating Thin-Plate Splines , 1996, VBC.

[10]  J. Richards,et al.  The measurement of human motion: A comparison of commercially available systems , 1999 .

[11]  Sariel Har-Peled,et al.  Efficiently approximating the minimum-volume bounding box of a point set in three dimensions , 1999, SODA '99.

[12]  Dinesh Manocha,et al.  Incremental Algorithms for Collision Detection Between Polygonal Models , 1997, IEEE Trans. Vis. Comput. Graph..

[13]  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).

[14]  Adrian Hilton,et al.  A survey of advances in vision-based human motion capture and analysis , 2006, Comput. Vis. Image Underst..

[15]  Philip M. Hubbard,et al.  Collision Detection for Interactive Graphics Applications , 1995, IEEE Trans. Vis. Comput. Graph..

[16]  Thomas B. Moeslund,et al.  A Survey of Computer Vision-Based Human Motion Capture , 2001, Comput. Vis. Image Underst..

[17]  Kenichi Kanatani,et al.  Analysis of 3-D Rotation Fitting , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

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

[19]  Berthold K. P. Horn,et al.  Closed-form solution of absolute orientation using unit quaternions , 1987 .

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

[21]  Jessica K. Hodgins,et al.  Capturing and animating skin deformation in human motion , 2006, SIGGRAPH 2006.

[22]  David S. Doermann,et al.  Robust point matching for nonrigid shapes by preserving local neighborhood structures , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[23]  Baihua Li,et al.  Using k-d trees for robust 3D point pattern matching , 2003, Fourth International Conference on 3-D Digital Imaging and Modeling, 2003. 3DIM 2003. Proceedings..

[24]  Horace Ho-Shing Ip,et al.  Robust point correspondence matching and similarity measuring for 3D models by relative angle-context distributions , 2008, Image Vis. Comput..

[25]  Eric Mjolsness,et al.  New Algorithms for 2D and 3D Point Matching: Pose Estimation and Correspondence , 1998, NIPS.