DMCA
Iterative quantization: A procrustean approach to learning binary codes (2011)
Cached
Download Links
| Venue: | In Proc. of the IEEE Int. Conf. on Computer Vision and Pattern Recognition (CVPR |
| Citations: | 157 - 6 self |
BibTeX
@INPROCEEDINGS{Gong11iterativequantization:,
author = {Yunchao Gong and Svetlana Lazebnik},
title = {Iterative quantization: A procrustean approach to learning binary codes},
booktitle = {In Proc. of the IEEE Int. Conf. on Computer Vision and Pattern Recognition (CVPR},
year = {2011}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
This paper addresses the problem of learning similaritypreserving binary codes for efficient retrieval in large-scale image collections. We propose a simple and efficient alternating minimization scheme for finding a rotation of zerocentered data so as to minimize the quantization error of mapping this data to the vertices of a zero-centered binary hypercube. This method, dubbed iterative quantization (ITQ), has connections to multi-class spectral clustering and to the orthogonal Procrustes problem, and it can be used both with unsupervised data embeddings such as PCA and supervised embeddings such as canonical correlation analysis (CCA). Our experiments show that the resulting binary coding schemes decisively outperform several other state-of-the-art methods. 1.
Keyphrases
binary code iterative quantization procrustean approach unsupervised data embeddings binary coding scheme canonical correlation analysis large-scale image collection several state-of-the-art method minimization scheme quantization error efficient retrieval zerocentered data zero-centered binary hypercube multi-class spectral clustering orthogonal procrustes problem
