Shape-Colour Histograms for matching 3D video sequences

Most 3D object retrieval and matching methods only consider geometric similarity. This paper introduces a novel descriptor, Shape-Colour Histograms, to match objects with similar shape and appearance. This is motivated by the requirement to concatenate captured 3D video sequences for animation production. A quantitative evaluation based on the Receiver-Operator Characteristic (ROC) curve is presented to compare the performance of conventional 3D shape descriptors and new shape-colour descriptors with temporal filtering in the task to match 3D video sequences. 3D shape descriptors including Shape Histogram, LightField are considered. Evaluation shows that filtered Shape-Colour Histograms outperform descriptors using shape only. Finally, temporal Shape-Colour Histograms are applied to a publically available database of 3D video of people to identify optimal transitions and synthesize 3D character animation. Results demonstrate the accurate matching of surface shape, appearance and motion at transition points.

[1]  Szymon Rusinkiewicz,et al.  Rotation Invariant Spherical Harmonic Representation of 3D Shape Descriptors , 2003, Symposium on Geometry Processing.

[2]  Andrew E. Johnson,et al.  Using Spin Images for Efficient Object Recognition in Cluttered 3D Scenes , 1999, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Jiri Matas,et al.  On representation and matching of multi-coloured objects , 1995, Proceedings of IEEE International Conference on Computer Vision.

[4]  Adrian Hilton,et al.  Correspondence labelling for wide-timeframe free-form surface matching , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[5]  Christian Rössl,et al.  Dense correspondence finding for parametrization-free animation reconstruction from video , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[6]  Bernard Chazelle,et al.  Shape distributions , 2002, TOGS.

[7]  Adrian Hilton,et al.  A Study of Shape Similarity for Temporal Surface Sequences of People , 2007, Sixth International Conference on 3-D Digital Imaging and Modeling (3DIM 2007).

[8]  Adrian Hilton,et al.  Human motion synthesis from 3D video , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[9]  Barbara Caputo,et al.  How to Combine Color and Shape Information for 3D Object Recognition: Kernels do the Trick , 2002, NIPS.

[10]  Glenn Healey,et al.  Combining color and geometric information for the illumination invariant recognition of 3-D objects , 1995, Proceedings of IEEE International Conference on Computer Vision.

[11]  Adrian Hilton,et al.  Temporal 3D shape matching , 2007 .

[12]  Adrian Hilton,et al.  Surface Capture for Performance-Based Animation , 2007, IEEE Computer Graphics and Applications.

[13]  Bobby Bodenheimer,et al.  Synthesis and evaluation of linear motion transitions , 2008, TOGS.

[14]  Hans-Peter Kriegel,et al.  3D Shape Histograms for Similarity Search and Classification in Spatial Databases , 1999, SSD.

[15]  Hans-Peter Seidel,et al.  Performance capture from sparse multi-view video , 2008, ACM Trans. Graph..

[16]  Lucas Kovar,et al.  Motion graphs , 2002, SIGGRAPH '08.

[17]  Radu Horaud,et al.  Temporal Surface Tracking Using Mesh Evolution , 2008, ECCV.

[18]  Ming Ouhyoung,et al.  On Visual Similarity Based 3D Model Retrieval , 2003, Comput. Graph. Forum.

[19]  Kiyoharu Aizawa,et al.  Motion Editing in 3D Video Database , 2006, Third International Symposium on 3D Data Processing, Visualization, and Transmission (3DPVT'06).