TZI Literaturverzeichnis

WIKINDX Resources

Journal Article: ID no. (ISBN etc.):  ISSN: 2005-4254 BibTeX citation key:  Stommel2010
M. Stommel, "Binarising SIFT-Descriptors to Reduce the Curse of Dimensionality in Histogram-Based Object Recognition", International Journal of Signal Processing, vol. 3, iss. 1, pp. 25–36, Mar. 2010.
Added by: Martin Stommel 2010-08-27 16:47:46
Categories: AG-KI
Keywords: binarisation, clustering, curse of dimensionality, Object Recognition, SIFT
Creators: Stommel
Collection: International Journal of Signal Processing

Peer reviewed
Number of views:  637
Popularity index:  26.9%
Number of attachment views:  1349

Attached file:    

It is shown that distance computations between SIFT-descriptors using the Euclidean distance suffer from the curse of dimensionality. The search for exact matches is less affected than the generalisation of image patterns, e.g. by clustering methods. Experimental results indicate that for the case of generalisation, the Hamming distance on binarised SIFTdescriptors is a much better choice. It is shown that the binary feature representation is visually plausible, numerically stable and information preserving. In an histogram-based object recognition system, the binary representation allows for the quick matching, compact storage and fast training of a code-book of features. A time-consuming clustering of the input data is redundant.
Added by: Martin Stommel

Publisher: Science & Engineering Research Support Center SERSC
Added by: Martin Stommel

wikindx      |     Total Resources:  3135     |     Database queries:  34     |     Script execution:  0.47644 secs