ABSTRACT
Commonly used content-based image retrieval systems focus on the problem of finding similar images for a given single query object out of a database of media objects. We consider a similarity-based image join of two image tables, where the image data components are represented by their respective feature vectors. For each image of the first table, similar images are looked up in the second table. Matching tuples are combined. We consider multiple joins which allows one to join a previous join result to another image table, and so on. Thus, each multiple join result tuple contains n images, if n tables are joined. An image of the result tuple is therefore not only similar to the image from its join partner, but also to the image similar to it. In this context, the paper presents processing and optimizing strategies for multiple similarity-based image joins and a cost model for integrating them in a multimedia database. The cost model is validated by an X-tree reference implementation. The presented strategies in this paper are currently been implemented in Oracle Multimedia.
- M. S. Lew, N. Sebe, C. Djerba, and R. Jain. Content-based multimedia information retrieval: State-of-the-art and challenges. ACM Transactions on Multimedia Computing, Communications, and Applications, 2(1):1--19, 2006. Google ScholarDigital Library
- R. Datta, D. Joshi, J. Li, and J. Z. Wang. Image retrieval: Ideas, influences, and trends of the new age. ACM Computing Surveys, 40(2):1--60, 2008. Google ScholarDigital Library
- V. Oria, M. T. Özsu, S. Lin, and P. Iglinski. Similarity queries in the DISIMA image DBMS. In ACM Multimedia 2001, pages 475--478, 2001. Google ScholarDigital Library
- Harald Kosch and Solomon Atnafu. Processing a multimedia join through the method of nearest neighbor search. Inf. Process. Lett., 82(5):269--276, 2002. Google ScholarDigital Library
- C. Böhm. A cost model for query processing in high dimensional data spaces. ACM Transactions on Database Systems, 25, 2000. Google ScholarDigital Library
- H. Ferhatosmanoglu, E. Tuncel, D. Agrawal, and A. El Abbadi. High dimensional nearest neighbor searching. Information Systems, 31(6):512--540, 2006. Google ScholarDigital Library
- H. Samet. Foundations of Multidimensional and Metric Data Structures. Morgan Kaufmann, 2006. Google ScholarDigital Library
- T. Neumann and G. Moerkotte. An efficient framework for order optimization. In International Conference on Data Engineering, ICDE 2004, pages 461--472, 2004. Google ScholarDigital Library
- J. M. Hellerstein. Encyclopedia of Database Systems, chapter Generalized Search Tree, pages 1222--1224. Springer US, 2009.Google Scholar
Index Terms
- Optimizing similarity-based image joins in a multimedia database
Recommendations
Foundations of multimedia database systems
Though numerous multimedia systems exist in the commercial market today, relatively little work has been done on developing the mathematical foundation of multimedia technology. We attempt to take some initial steps towards the development of a ...
Similarity Joins
Similarity Joins are extensively used in multiple application domains and are recognized among the most useful data processing and analysis operations. They retrieve all data pairs whose distances are smaller than a predefined threshold ε. While several ...
Optimizing Top-k Selection Queries over Multimedia Repositories
Repositories of multimedia objects having multiple types of attributes (e.g., image, text) are becoming increasingly common. A query on these attributes will typically request not just a set of objects, as in the traditional relational query model (...
Comments