Combining Features for Content-Based Sketch Retrieval — A Comparative Evaluation of Retrieval Performance
We study three transformation-invariant shape descriptors and evaluate their relative strengths in the context of content-based retrieval of graphical sketches. We show that the use of a combination of different shape representations may lead to a significant improvement of retrieval performance and identify an optimal combination that proves robust across different data sets and queries.
KeywordsImage Retrieval Average Precision Retrieval Performance Moment Invariant Shape Representation
Unable to display preview. Download preview PDF.
- 1.T S Rui, T S Huang, M Ortega, and S Mehrota, “Relevance feedback: a power tool for interactive content-based image retrieval,” IEEE Trans. Circuits and Systems for Video Technology, pp. 123–131, 1998.Google Scholar
- 8.M K Hu, “Visual pattern recognition by moment invariants.,” IRE Transactions on Information Theory, vol. 8, pp. 179–187, 1962.Google Scholar
- 12.J R Smith and C S Li, “Image retrieval evaluation,” in Proc. Workshop Content-Based Access of Image and Video Libraries, 1998, pp. 343–353.Google Scholar
- 13.E M Voorhees and D Harman, “Overview of the eigth text retrieval conference (TREC-8),” in Proc. TREC, 1999, pp. 1–33 and A.17-A.18.Google Scholar
- 14.A L Reno, Object recognition by stochastic model adaptation and selection, Ph.D. thesis, Imperial College, London, 1998.Google Scholar