Appeal 2007-2011
Application 09/823,272
Appellants' invention relates generally to an indexing method of a
feature vector data space. See Specification 1:3-7. Claim 1 is illustrative of
the claimed invention, and it reads as follows:
1. An indexing method of a feature vector data space in which a
plurality of feature vectors are indexed, the indexing method comprising the
steps of:
(pa-1) partitioning the feature vector data space into a plurality of
cells having a uniform size;
(a) determining whether one or more cells from said plurality of cells,
on each of which one or more of said plurality of feature vectors are
correspondingly concentrated, exist; and
(b) hierarchically indexing the feature vector data space when it is
determined that said one or more cells, on each of which said one or more of
said plurality of feature vectors are correspondingly concentrated, exist in
the step (a)
wherein, one or more feature vectors are concentrated in a cell when
the cell contains more feature vectors than a predetermined threshold.
The prior art references of record relied upon by the Examiner in
rejecting the appealed claims are:
Kothuri US 6,381,605 B1 Apr. 30, 2002
Xia Wan et al., "A New Approach to Image Retrieval with Hierarchical
Color Clustering," IEEE Transactions on Circuits and Systems for Video
Technology 8:5, September 1998, pp. 628-43. (Wan)
Roger Weber et al., "A Quantitative Analysis and Performance Study for
Similarity-Search Methods in High-Dimensional Spaces," Proceedings of
the 24th International Conference on Very Large Data Base, New York,
August 1998, pp. 194-205. (Weber)
2
Page: Previous 1 2 3 4 5 6 7 8 9 Next
Last modified: September 9, 2013