The R*-tree: an efficient and robust access method for points and rectangles
SIGMOD '90 Proceedings of the 1990 ACM SIGMOD international conference on Management of data
Two algorithms for nearest-neighbor search in high dimensions
STOC '97 Proceedings of the twenty-ninth annual ACM symposium on Theory of computing
The pyramid-technique: towards breaking the curse of dimensionality
SIGMOD '98 Proceedings of the 1998 ACM SIGMOD international conference on Management of data
Data structures and algorithms for nearest neighbor search in general metric spaces
SODA '93 Proceedings of the fourth annual ACM-SIAM Symposium on Discrete algorithms
The K-D-B-tree: a search structure for large multidimensional dynamic indexes
SIGMOD '81 Proceedings of the 1981 ACM SIGMOD international conference on Management of data
R-trees: a dynamic index structure for spatial searching
SIGMOD '84 Proceedings of the 1984 ACM SIGMOD international conference on Management of data
M-tree: An Efficient Access Method for Similarity Search in Metric Spaces
VLDB '97 Proceedings of the 23rd International Conference on Very Large Data Bases
VLDB '98 Proceedings of the 24rd International Conference on Very Large Data Bases
Similarity Search in High Dimensions via Hashing
VLDB '99 Proceedings of the 25th International Conference on Very Large Data Bases
The X-tree: An Index Structure for High-Dimensional Data
VLDB '96 Proceedings of the 22th International Conference on Very Large Data Bases
Indexing very high-dimensional sparse and quasi-sparse vectors for similarity searches
The VLDB Journal — The International Journal on Very Large Data Bases
Spatial indexing of high-dimensional data based on relative approximation
The VLDB Journal — The International Journal on Very Large Data Bases
NeTra: a toolbox for navigating large image databases
ICIP '97 Proceedings of the 1997 International Conference on Image Processing (ICIP '97) 3-Volume Set-Volume 1 - Volume 1
An efficient indexing method for nearest neighbor searches inhigh-dirnensional image databases
IEEE Transactions on Multimedia
MMM'07 Proceedings of the 13th international conference on Multimedia Modeling - Volume Part I
Hi-index | 0.00 |
Recently, some researches for NN (nearest neighbour) search in multimedia retrieval, for example, the VA-file [11] and the LPC-file [12], have been proposed to resolve the problem called "curse of dimensionality"[17]. They are called as filtering approach because they first filter-out the irrelevant objects by scanning all object approximations and compute the distances between the query and remaining objects to find out exact NN. In this approach, since all approximations are scanned in the filtering step, the efficiency of computation in the filtering process must be seriously considered. Otherwise, it would be very hard to be used in the real applications with a lot of multimedia objects. This paper proposes an efficient indexing mechanism for NN search to speed-up this filtering process using a novel indexing structure, called hierarchical bitmap, in which each object is represented as a bitmap of size 2ċd where d is the dimension of object's feature vector. That is, i-th feature value (1≤i≤d) of object is approximated with two bits that represent whether it is relatively high, low, or neither compared to the i-th feature value of other objects. As performing XOR operation between two bitmaps, we can calculate the lower bound of Lp-distance between two feature vectors. This mechanism can be hierarchically applied to generate multiple bitmaps of a vector in order to raise the filtering rate.