A hierarchical indexing structure for 3D model retrieval based on the hierarchical self organizing map (HSOM) is proposed. The proposed approach organizes the database into a hierarchy so that head models are partitioned by coarse features initially and finer scale features are used in lower levels. The aim is to traverse a small subset of the database during retrieval. This is made possible by exploiting the multi-resolution capability of spherical wavelet features to successively approximate the salient characteristics of the head models, which are encoded in the form of weight vectors associated with the nodes at different levels (from coarse to fine) of the HSOM. To avoid premature commitment to a possibly erroneous model class, search is propagated from a subset of nodes at each level, which is selected based on a fuzzy membership measure between the query feature vector and weight vector, instead of taking the winner-take-all approach. Experiments show that, in addition to efficiency improvement, model retrieval based on the HSOM approach is able to achieve a much higher accuracy compared with the case where no indexing is performed.
[1]
Jouko Lampinen,et al.
Clustering properties of hierarchical self-organizing maps
,
1992,
Journal of Mathematical Imaging and Vision.
[2]
Laurent Jacques,et al.
Wavelets on the sphere: implementation and approximations
,
2002
.
[3]
Dietmar Saupe,et al.
Tools for 3D-object retrieval: Karhunen-Loeve transform and spherical harmonics
,
2001,
2001 IEEE Fourth Workshop on Multimedia Signal Processing (Cat. No.01TH8564).
[4]
José Miguel Espadero,et al.
3D wavelet-based multiresolution object representation
,
2001,
Pattern Recognit..
[5]
Taku Komura,et al.
Topology matching for fully automatic similarity estimation of 3D shapes
,
2001,
SIGGRAPH.
[6]
Daniel A. Keim,et al.
Efficient geometry-based similarity search of 3D spatial databases
,
1999,
SIGMOD '99.
[7]
Marc Rioux,et al.
Description of shape information for 2-D and 3-D objects
,
2000,
Signal Process. Image Commun..