A General Framework for Approximate Nearest Subspace Search

Ronen Basri1    Tal Hassner2    Lihi Zelnik-Manor3

1. Dept. of Computer Science and Applied Math.
Weizmann Institute of Science

2. Computer Science Division
The Open University of Israel

3. Dept. of Electrical Engineering
California Institute of Technology


Abstract: Subspaces offer convenient means of representing information in many Pattern Recognition, Machine Vision, and Statistical Learning applications. Contrary to the growing popularity of subspace representations, the problem of efficiently searching through large subspace databases has received little attention in the past. In this paper we present a general solution to the Approximate Nearest Subspace search problem. Our solution uniformly handles cases where both query and database elements may differ in dimensionality, where the database contains subspaces of different dimensions, and where the queries themselves may be subspaces. To this end we present a simple mapping from subspaces to points, thus reducing the problem to the well studied Approximate Nearest Neighbor problem on points. We provide theoretical proofs of correctness and error bounds of our construction and demonstrate its capabilities on synthetic and real data. Our experiments indicate that an approximate nearest subspace can be located significantly faster than the nearest subspace, with little loss of accuracy.

Reference:  R. Basri, T.Hassner and L. Zelnik-Manor, A General Framework for Approximate Nearest Subspace Search2nd IEEE International Workshop on Subspace Methods at the IEEE International Conference on Computer Vision (ICCV), Kyoto, Sept 2009.

Click here for the PDF (833kb)

Click here for the BibTex

Older tech-report Reference:  R. Basri, T.Hassner and L. Zelnik-Manor, A General Framework for Approximate Nearest Subspace Search, Technical Report CCIT #699 , Department of Electrical Engineering, Technion, June 2008.


 

Some results

Please see paper for a full description of our experiments and more results.

Synthetic data tests

Varying database size Varying query dimension Varying ambient space
Varying database size Varying query dimension Varying ambient space dimension

log-scale run times compared for an exact linear search, linear search with a preprocessed database and our ANS method with ANN epsilon=100. The following tests were performed. Varying number of database subspaces (database size): Database subspaces of dimension 30 embedded in a 60 dimensional  space, tested with 1000 queries of dimension 10. Varying query dimension: Database contains 1000 subspaces with dimension 30, tested with 1000 queries. Varying ambient dimension: Database contains 1000 subspaces with dimension 30, tested with 1000 queries of dimension 10. The error rate for our method remained at an almost constant 0.01 in all three experiments, through all values tested.

Scene classification results

Method Run-time Correct
ANS (Our result) 2.3 sec. 63%
Exact subspace search (with preprocessing) 13.8 sec. 63%
Exact nearest patch 135 sec 55%


Copyright: no material is allowed to be copied or used in any way without written permission of the authors.

Last update 21st of Sept, 2009