![](bg8.png)
mines the period of a CRV. The CRVs are grouped
together in such a way that in each group, they are
all pairwise uncorrelated. The CRVs in each group
are used to compute an integer hash value that index
the data points in a one-dimensional hash table. In
the query phase, the same process is applied on the
query points and the obtained hash value is used to
fetch only the database points that can be candidate
neighbours for the query point. The proposed method
was tested on a standard dataset of real-world im-
ages and compared with LSH and FLANN Matcher.
The presented experimental results show that, in case
of a static database, our CRV matcher is faster than
FLANN for smaller databases (less than 200K fea-
tures). For a dynamic database, the CRV Matcher is
(10–20) time faster than FLANN depending on the
size of the database.
REFERENCES
Andoni, A. and Indyk, P. (2008). Near-optimal hash-
ing algorithms for approximate nearest neighbor in
high dimensions. Communications of the ACM,
51(1):117122.
Arya, S., Mount, D. M., Netanyahu, N. S., Silverman, R.,
and Wu, A. Y. (1998). An optimal algorithm for ap-
proximate nearest neighbor searching in fixed dimen-
sions. J. of the ACM, 45(6):891923.
Bawa, M., Condie, T., and Ganesan, P. (2005). Lsh forest:
Self-tuning indexes for similarity search. In Proc. In-
ternational World Wide Web Conference (WWW-05),
page 651660.
Beis, J. S. and Lowe, D. G. (1997). Shape indexing us-
ing approximate nearest-neighbour search in high-
dimensional spaces. In Proc. IEEE Conference on
Computer Vision and Pattern Recognition (CVPR-97),
page 10001006.
Bentley, L. (1975). Multidimensional binary search trees
used for associative searching. Communications of the
ACM (CACM), 18(9):509517.
Datar, M., Indyk, P., Immorlica, N., and Mirrokni, V. S.
(2004). Locality-sensitive hashing scheme based on
p-stable distributions. In Proc. of the thirtieth annual
ACM symposium on Theory of computing (STOC-04),
pages 253–262.
Friedman, J. H., Bentley, J. L., and Finkel, R. A. (1977).
An algorithm for finding best matches in logarithmic
expected time. ACM Transactions on Mathematical
Software, 3(3):209226.
Fukunaga, K. and Narendra, P. M. (1975). A branch and
bound algorithm for computing k-nearest neighbors.
IEEE Transactions on Computers (TC), C-24(7):750–
753.
Hajebi, K., Abbasi-Yadkori, Y., Shahbazi, H., and Zhang,
H. (2011). Fast approximate nearest-neighbor search
with k-nearest neighbor graph. In Proc. 22nd In-
ternational Joint Conference on Artificial Intelligence
(IJCAI-11), page 13121317.
Har-Peled, S., Indyk, P., and Motwani, R. (2012). Approxi-
mate nearest neighbor: Towards removing the curse of
dimensionality. Theory of Computing, 8(1):321–350.
Indyk, P. and Motwani, R. (1998). Approximate nearest
neighbors: Towards removing the curse of dimension-
ality. In Proc. Symposium on Computational Geome-
try (SoCG-98), pages 604–613.
Kulis, B. and Grauman, K. (2009). Kernelized locality-
sensitive hashing for scalable image search. In Proc.
IEEE 12th International Conference on Computer Vi-
sion (ICCV-09), page 21302137.
Muja, M. and Lowe, D. (2009). Fast approximate near-
est neighbors with automatic algorithm configuration.
In Proc. International Conference on Computer Vision
Theory and Applications (VISAPP-09), page 331340.
Philbin, J., Chum, O., Isard, M., Sivic, J., and Zisserman, A.
(2007). Object retrieval with large vocabularies and
fast spatial matching. In Proc. IEEE Conference on
Computer Vision and Pattern Recognition (CVPR-07).
Sebastian, B. and Kimia, B. B. (2002). Metric-based shape
retrieval in large databases. In Proc. IEEE Conference
on Computer Vision and Pattern Recognition (CVPR-
02), volume 3, pages 291–296.
Silpa-Anan, C. and Hartley, R. (2008). Optimised kd-trees
for fast image descriptor matching. In Proc. IEEE
Conference on Computer Vision and Pattern Recog-
nition (CVPR-08), pages 1–8.
Wang, J., Kumar, S., and Chang, S. F. (2010). Semi-
supervised hashing for scalable image retrieval. In
Proc. IEEE Conference on Computer Vision and Pat-
tern Recognition (CVPR-10), page 34243431.
Winder, S., Hua, G., and Brown, M. (2009). Picking the
best daisy. In Proceedings of the International Con-
ference on Computer Vision and Pattern Recognition
(CVPR09), Miami.
CRVM: Circular Random Variable-based Matcher - A Novel Hashing Method for Fast NN Search in High-dimensional Spaces
221