Principal Component Hashing: An Accelerated Approximate Nearest Neighbor Search

Authors:
Yusuke Matsushita;Toshikazu Wada
Affiliations:
Graduate School of Systems Engineering, Wakayama University, Wakayama, Japan 640-8510;Graduate School of Systems Engineering, Wakayama University, Wakayama, Japan 640-8510
Venue:
PSIVT '09 Proceedings of the 3rd Pacific Rim Symposium on Advances in Image and Video Technology
Year:
2009

Citing 9
Cited 1

An algorithm for finding nearest neighbours in (approximately) constant average time

Pattern Recognition Letters
A new version of the nearest-neighbour approximating and eliminating search algorithm (AESA) with linear preprocessing time and memory requirements

Pattern Recognition Letters
Approximate nearest neighbors: towards removing the curse of dimensionality

STOC '98 Proceedings of the thirtieth annual ACM symposium on Theory of computing
An optimal algorithm for approximate nearest neighbor searching fixed dimensions

Journal of the ACM (JACM)
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
Multidimensional binary search trees used for associative searching

Communications of the ACM
Near Neighbor Search in Large Metric Spaces

VLDB '95 Proceedings of the 21th International Conference on Very Large Data Bases
Locality-sensitive hashing scheme based on p-stable distributions

SCG '04 Proceedings of the twentieth annual symposium on Computational geometry
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

Action Recognition Based on Non-parametric Probability Density Function Estimation

ISVC '09 Proceedings of the 5th International Symposium on Advances in Visual Computing: Part II

Quantified Score

Hi-index	0.02

Visualization

Abstract

Nearest Neighbor (NN) search is a basic algorithm for data mining and machine learning applications. However, its acceleration in high dimensional space is a difficult problem. For solving this problem, approximate NN search algorithms have been investigated. Especially, LSH is getting highlighted recently, because it has a clear relationship between relative error ratio and the computational complexity. However, the p-stable LSH computes hash values independent of the data distributions, and hence, sometimes the search fails or consumes considerably long time. For solving this problem, we propose Principal Component Hashing (PCH), which exploits the distribution of the stored data. Through experiments, we confirmed that PCH is faster than ANN and LSH at the same accuracy.