LOCMAX SIFT - Non-Statistical Dimension Reduction on Invariant Descriptors

Dávid Losteiner, László Havasi, Tamás Szirányi

2009

Abstract

The descriptors used for image indexing - e.g. Scale Invariant Feature Transform (SIFT) - are generally parameterized in very high dimensional spaces which guarantee the invariance on different light conditions, orientation and scale. The number of dimensions limit the performance of search techniques in terms of computational speed. That is why dimension reduction of descriptors is playing an important role in real life applications. In the paper we present a modified version of the most popular algorithm, SIFT. The motivation was to speed up searching on large feature databases in video surveillance systems. Our method is based on the standard SIFT algorithm using a structural property: the local maxima of these high dimensional descriptors. The weighted local positions will be aligned with a dynamic programming algorithm (DTW) and its error is calculated as a new kind of measure between descriptors. In our approach we do not use a training set, pre-computed statistics or any parameters when finding the matches, which is very important for an online video indexing application.

References

  1. Brown, M. and Lowe, D. G., 2003. Recognizing Panorama, Proceedings of Ninth IEEE International Conference on Computer Vision Vol.2 pp. 1218- 1225 Hua, G., Brown, M., Winder, S., 2007. Discriminant Embedding for Local Image Descriptors, IEEE International Conference on Computer Vision, Rio de Janeiro, Brazil, pp. 1-8.
  2. Joliffe, I.T., 1986. Principal Component Analysis. Springer-Verlag.
  3. Ke, Y., and Sukthankar, R., 2004. PCA-SIFT: A More Distinctive Representation for Local Image Descriptors. CVPR 2004. Proceedings of the 2004 IEEE Computer Society Conference on, Vol.2 , pp. 506-513.
  4. Lowe, D., 1999. Object recognition from local scaleinvariant features. In Proceedings of International Conference on Computer Vision, pp. 1150-1157.
  5. Lowe, D., 2004. Distinctive image features from scaleinvariant keypoints. International Journal of Computer Vision. pp. 91-110.
  6. Mikolajczyk, K. and Schmid, C., 2001. Indexing based on scale invariant interest points. In Proceedings of International Conference on Computer Vision, pp. 525-531.
  7. Mikolajczyk, K. and Schmid, C., 2005. A performance evaluation of local descriptors. IEEE Transaction on Pattern Analysis and Machine Intelligence, pp. 27(10):1615-1630.
  8. Myers, C. S. and Rabiner, L. R. 1981. A comparative study of several dynamic time-warping algorithms for connected word recognition. The Bell System Technical Journal, pp. 60(7):1389-1409.
  9. Schügerl, P., Sorschag, R., Bailer, W., Thallinger, G., ,2007. Object Re-detection Using SIFT and MPEG-7 Color Descriptors, Multimedia Content Analysis and Mining, Springer-Verlag, pp. 305-314.
Download


Paper Citation


in Harvard Style

Losteiner D., Havasi L. and Szirányi T. (2009). LOCMAX SIFT - Non-Statistical Dimension Reduction on Invariant Descriptors . In Proceedings of the Fourth International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2009) ISBN 978-989-8111-69-2, pages 192-195. DOI: 10.5220/0001799401920195


in Bibtex Style

@conference{visapp09,
author={Dávid Losteiner and László Havasi and Tamás Szirányi},
title={LOCMAX SIFT - Non-Statistical Dimension Reduction on Invariant Descriptors },
booktitle={Proceedings of the Fourth International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2009)},
year={2009},
pages={192-195},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001799401920195},
isbn={978-989-8111-69-2},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Fourth International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2009)
TI - LOCMAX SIFT - Non-Statistical Dimension Reduction on Invariant Descriptors
SN - 978-989-8111-69-2
AU - Losteiner D.
AU - Havasi L.
AU - Szirányi T.
PY - 2009
SP - 192
EP - 195
DO - 10.5220/0001799401920195