Shape Indexing of Image Databases (SIID)

The main goal of this project is to create an image retrieval system that is primarily based on shape. The use of shape as a cue for indexing into pictorial databases has been traditionally based on global invariant statistics and deformable templates, on the one hand, and local edge correlation on the other. We have proposed an intermediate approach based on a characterization of the symmetry in edge maps.

Shape Representation

Representation of Symmetry as Shocks
Graph Representation of Shock Structure

Shape Similarity

Graph Matching Approach (Graduated Assignment)
Morphing Approach (Edit Distance)

Selected Publications

Symmetry-based Indexing of Image Databases. Postscript
Shape matching using edit-distance: an implementation. SODA2001. Postscript
Recognition of Shapes by Editing their Shock Graphs. ICCV2001. Postscript

Database

Download shape database (small) Paper to cite: Postscript, Bibtex

Download latest shape databases: 99 shapes , 216 shapes Paper to cite: Postscript, Bibtex

Note: The shapes for the 216 shape database was selected from those used for MPEG-7 test

Acknowledgements

The support of NSF Grants CCR-9700146, IRI-9700497, IRI-0083231 and BCS-9980091 is gratefully acknowledged