The R*-tree: an efficient and robust access method for points and rectangles
SIGMOD '90 Proceedings of the 1990 ACM SIGMOD international conference on Management of data
Shape representation and image segmentation using deformable surfaces
Image and Vision Computing - Special issue: range image understanding
Surface shape and curvature scales
Image and Vision Computing
Topology matching for fully automatic similarity estimation of 3D shapes
Proceedings of the 28th annual conference on Computer graphics and interactive techniques
Mathematics for 3D game programming and computer graphics
Mathematics for 3D game programming and computer graphics
R-trees: a dynamic index structure for spatial searching
SIGMOD '84 Proceedings of the 1984 ACM SIGMOD international conference on Management of data
An overview on properties and efficacy of topological skeletons in Shape Modelling
SMI '03 Proceedings of the Shape Modeling International 2003
Content-based Retrieval of 3D Models in Distributed Web Databases by Visual Shape Information
IV '00 Proceedings of the International Conference on Information Visualisation
Matching 3D Models with Shape Distributions
SMI '01 Proceedings of the International Conference on Shape Modeling & Applications
Hi-index | 0.00 |
In this paper we propose a feature extraction method for shape-based retrieval of 3D models, which uses the mutual intersected meshes between model and growing spheres. Since the feature descriptor of 3D model should be invariant to translation, rotation and scaling, we firstly normalize the pose of 3D models using principal component analysis method. We therefore represent them in a canonical coordinate system. The proposed algorithm for feature extraction is as follows. We generate a unit-size circum-sphere around 3D model, and locate the model in the center of the circum-sphere. We produce the concentric spheres with a different radius (ri=i/n, i=1,2,...,n). After finding the intersected meshes between the concentric spheres and object, we compute the mean curvatures of the meshes for each growing spheres, and use them as the feature descriptor of 3D model. Experimental evidence shows that our algorithm outperforms other methods for 3D indexing and retrieval. To index the multi-dimensional feature vectors, we use R*-tree structure.