Efficient Approximate Nearest Neighbors for General Metric Spaces
This is a github fork of the original SourceForge site: http://sourceforge.net/projects/proximityforest/ More information is available: https://sites.google.com/site/svohara/proximity-forest
Categories
- Spatial Algorithms
- Clustering
- Indexing
- Machine Vision
- Machine Learning
License
- GNU General Public License version 3.0 (GPLv3)
Features
- Approximate Nearest Neighbors of General Metric Spaces
- Simple Python code base
- Includes Code to Reproduce CVPR 2012 Publication's Results
- Includes action recognition examples using a Subspace Forest
- Includes Code to Reproduce WACV 2013 Publication's Results
Description
A proximity forest is a data structure that allows for efficient computation of approximate nearest neighbors of arbitrary data elements in a metric space.
See: O'Hara and Draper, "Are You Using the Right Approximate Nearest Neighbor Algorithm?", WACV 2013 (best student paper award).
One application of a ProximityForest is given in the following CVPR publication: Stephen O'Hara and Bruce A. Draper, "Scalable Action Recognition with a Subspace Forest," IEEE Conference on Computer Vision and Pattern Recognition, 2012.
This source code is provided without warranty and is available under the GPL license. More commercially-friendly licenses may be available. Please contact Stephen O'Hara for license options.
Please view the wiki on this site for installation instructions and examples on reproducing the results of the papers.