Fast approximate nearest neighbors with automatic algorithm configuration (2009)
Cached
Download Links
| Venue: | In VISAPP International Conference on Computer Vision Theory and Applications |
| Citations: | 441 - 2 self |
BibTeX
@INPROCEEDINGS{Muja09fastapproximate,
author = {Marius Muja and David G. Lowe},
title = {Fast approximate nearest neighbors with automatic algorithm configuration},
booktitle = {In VISAPP International Conference on Computer Vision Theory and Applications},
year = {2009},
pages = {331--340}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
nearest-neighbors search, randomized kd-trees, hierarchical k-means tree, clustering. For many computer vision problems, the most time consuming component consists of nearest neighbor matching in high-dimensional spaces. There are no known exact algorithms for solving these high-dimensional problems that are faster than linear search. Approximate algorithms are known to provide large speedups with only minor loss in accuracy, but many such algorithms have been published with only minimal guidance on selecting an algorithm and its parameters for any given problem. In this paper, we describe a system that answers the question, “What is the fastest approximate nearest-neighbor algorithm for my data? ” Our system will take any given dataset and desired degree of precision and use these to automatically determine the best algorithm and parameter values. We also describe a new algorithm that applies priority search on hierarchical k-means trees, which we have found to provide the best known performance on many datasets. After testing a range of alternatives, we have found that multiple randomized k-d trees provide the best performance for other datasets. We are releasing public domain code that implements these approaches. This library provides about one order of magnitude improvement in query time over the best previously available software and provides fully automated parameter selection. 1
Keyphrases
automatic algorithm configuration hierarchical k-means tree approximate algorithm approximate nearest-neighbor algorithm component consists minor loss high-dimensional space known performance parameter selection query time large speedup available software high-dimensional problem minimal guidance nearest-neighbors search many datasets many algorithm many computer vision problem magnitude improvement k-d tree known exact algorithm new algorithm priority search parameter value linear search public domain code
