Product Quantization for Vector Retrieval with No Error

Andrzej Wichert

Abstract

We propose a coding mechanism for less costly exact vector retrieval for data bases representing vectors. The search starts at the subspace with the lowest dimension. In this subspace, the set of all possible similar vectors is determined. In the next subspace, additional metric information corresponding to a higher dimension is used to reduce this set. We demonstrate the method performing experiments on image retrieval on one thousand gray images of the size 128×96. Our model is twelve times less complex than a list matching.

References

  1. Andoni, A., Dater, M., Indyk, P., Immorlica, N., and Mirrokni, V. (2006). Locality-sensitive hashing using stable distributions. In MIT-Press, editor, Nearest Neighbor Methods in Learning and Vision: Theory and Practice, chapter 4. T. Darrell and P. Indyk and G.
  2. Ciaccia, P. and Patella, M. (2002). Searching in metric spaces with user-defined and approximate distances. ACM Transactions on Database Systems, 27(4).
  3. Faloutsos, C. (1999). Modern information retrieval. In Baeza-Yates, R. and Ribeiro-Neto, B., editors, Modern Information Retrieval, chapter 12, pages 345-365. Addison-Wesley.
  4. Faloutsos, C., Barber, R., Flickner, M., Hafner, J., Niblack, W., Petkovic, D., and Equitz, W. (1994). Efficient and effective querying by image content. Journal of Intelligent Information Systems, 3(3/4):231-262.
  5. Jegou, H., Douze, M., and Schmid, S. (2011). Product quantization for nearest neighbor search. IEEE Transactions on Pattern Analysis and Machine Intelligence, 33(1):117-128.
  6. Olafsson, A., Jonsson, B., and Amsaleg, L. (2008). Dynamic behavior of balanced nv-trees. In International Workshop on Content-Based Multimedia Indexing Conference Proceedings, IEEE, pages 174-183.
  7. Paolo Ciaccia, Marco Patella, P. Z. (1997). M-tree: An efficient access method for similarity search in metric spaces. In VLDB, pages 426-435.
  8. Sakurai, Y., Yoshikawa, M., Uemura, S., and Kojima, H. (2002). Spatial indexing of high-dimensional data based on relative approximation. VLDB Journal, 11(2):93-108.
  9. Wang, J., Li, J., and Wiederhold, G. (2001). Simplicity: Semantics-sensitive integrated matching for picture libraries. IEEE Transactions on Pattern Analysis and Machine Intelligence, 23(9):947-963.
  10. Wichert, A. (2008). Content-based image retrieval by hierarchical linear subspace method. Journal of Intelligent Information Systems, 31(1):85-107.
  11. Wichert, A. (2009). Image categorization and retrieval. In Proceedings of the 11th Neural Computation and Psychology Workshop. World Scientific.
  12. Wichert, A., Teixeira, P., Santos, P., and Galhardas, H. (2010). Subspace tree: High dimensional multimedia indexing with logarithmic temporal complexity. Journal of Intelligent Information Systems, 35(3):495- 516.
Download


Paper Citation


in Harvard Style

Wichert A. (2012). Product Quantization for Vector Retrieval with No Error . In Proceedings of the 14th International Conference on Enterprise Information Systems - Volume 1: ICEIS, ISBN 978-989-8565-10-5, pages 87-92. DOI: 10.5220/0003952800870092


in Bibtex Style

@conference{iceis12,
author={Andrzej Wichert},
title={Product Quantization for Vector Retrieval with No Error},
booktitle={Proceedings of the 14th International Conference on Enterprise Information Systems - Volume 1: ICEIS,},
year={2012},
pages={87-92},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003952800870092},
isbn={978-989-8565-10-5},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 14th International Conference on Enterprise Information Systems - Volume 1: ICEIS,
TI - Product Quantization for Vector Retrieval with No Error
SN - 978-989-8565-10-5
AU - Wichert A.
PY - 2012
SP - 87
EP - 92
DO - 10.5220/0003952800870092