International Journal of Computer Vision
The pyramid-technique: towards breaking the curse of dimensionality
SIGMOD '98 Proceedings of the 1998 ACM SIGMOD international conference on Management of data
The TV-tree: an index structure for high-dimensional data
The VLDB Journal — The International Journal on Very Large Data Bases - Spatial Database Systems
Trading Quality for Time with Nearest Neighbor Search
EDBT '00 Proceedings of the 7th International Conference on Extending Database Technology: Advances in Database Technology
Approximate Nearest Neighbor Searching in Multimedia Databases
Proceedings of the 17th International Conference on Data Engineering
VLDB '98 Proceedings of the 24rd International Conference on Very Large Data Bases
An introduction to variable and feature selection
The Journal of Machine Learning Research
Distinctive Image Features from Scale-Invariant Keypoints
International Journal of Computer Vision
Foundations of Multidimensional and Metric Data Structures (The Morgan Kaufmann Series in Computer Graphics and Geometric Modeling)
Near-Optimal Hashing Algorithms for Approximate Nearest Neighbor in High Dimensions
FOCS '06 Proceedings of the 47th Annual IEEE Symposium on Foundations of Computer Science
The Many Facets of Approximate Similarity Search
SISAP '08 Proceedings of the First International Workshop on Similarity Search and Applications (sisap 2008)
HiPeR: Hierarchical Progressive Exact Retrieval in Multidimensional Spaces
SISAP '08 Proceedings of the First International Workshop on Similarity Search and Applications (sisap 2008)
KpyrRec: a recursive multidimensional indexing structure
International Journal of Parallel, Emergent and Distributed Systems
The Many Facets of Progressive Retrieval for CBIR
PCM '08 Proceedings of the 9th Pacific Rim Conference on Multimedia: Advances in Multimedia Information Processing
The GC-tree: a high-dimensional index structure for similarity search in image databases
IEEE Transactions on Multimedia
Hi-index | 0.00 |
In this article, we present an approximate technique that allows accelerating similarity search in high dimensional vector spaces. The presented approach, called HiPeR, is based on a hierarchy of subspaces and indices: it performs nearest neighbors search across spaces of different dimensions, starting with the lowest dimensions up to the highest ones, aiming at minimizing the effects of the curse of dimensionality. In this work, HiPeR has been implemented on the classical index structure VA-File, providing VA-Hierarchies. The model of precision loss defined is probabilistic and non parametric and quality of answers can be selected by user at query time. HiPeR is evaluated for range queries on 3 real data-sets of image descriptors varying from 500,000 vectors to 4 millions. The experiments show that this approximate technique improves retrieval by saving I/O access significantly.