3D shape restoration via matrix recovery

Authors:
Min Lu;Bo Zheng;Jun Takamatsu;Ko Nishino;Katsushi Ikeuchi
Affiliations:
Institute of Industrial Science, The University of Tokyo, Tokyo, Japan;Institute of Industrial Science, The University of Tokyo, Tokyo, Japan;Nara Institute of Science and Technology, Nara, Japan;Department of Computer Science, Drexel University, Philadelphia;Institute of Industrial Science, The University of Tokyo, Tokyo, Japan
Venue:
ACCV'10 Proceedings of the 2010 international conference on Computer vision - Volume part II
Year:
2010

Citing 10
Cited 1

A Method for Registration of 3-D Shapes

IEEE Transactions on Pattern Analysis and Machine Intelligence - Special issue on interpretation of 3-D scenes—part II
Simplification and Repair of Polygonal Models Using Volumetric Techniques

IEEE Transactions on Visualization and Computer Graphics
Context-based surface completion

ACM SIGGRAPH 2004 Papers
Robust repair of polygonal models

ACM SIGGRAPH 2004 Papers
Template-based mesh completion

SGP '05 Proceedings of the third Eurographics symposium on Geometry processing
Example-based 3D scan completion

SGP '05 Proceedings of the third Eurographics symposium on Geometry processing
Three-Dimensional Surface Relief Completion Via Nonparametric Techniques

IEEE Transactions on Pattern Analysis and Machine Intelligence
3D Non-rigid Surface Matching and Registration Based on Holomorphic Differentials

ECCV '08 Proceedings of the 10th European Conference on Computer Vision: Part III
Möbius voting for surface correspondence

ACM SIGGRAPH 2009 papers
Deformation-driven shape correspondence

SGP '08 Proceedings of the Symposium on Geometry Processing

Preserving the Khmer smile: classifying and restoring the faces of Bayon

VAST'11 Proceedings of the 12th International conference on Virtual Reality, Archaeology and Cultural Heritage

Quantified Score

Hi-index	0.00

Visualization

Abstract

Cultural relics are often damaged and incomplete due to various reasons. For the purpose of helping archaeological studies, we present a novel method for simultaneously restoring the original shapes of a group of similar objects. Based on the assumption that similar shapes are approximately linearly correlated, we use a matrix recovery technique to achieve the restoration. In order to represent input shapes in a matrix form, vectorization of each aligned sample is carried out by stacking coordinates of dense corresponding points that are generated by a surface matching scheme using non-rigid deformation. An experiment using 3D scans of facial sculptures from Bayon is conducted, and the result verifies the feasibility and effectiveness of our method.