Nonlinear Discrete Hashing

In this paper, we propose a nonlinear discrete hashing approach to learn compact binary codes for scalable image search. Instead of seeking a single linear projection in most existing hashing methods, we pursue a multilayer network with nonlinear transformations to capture the local structure of dat...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on multimedia 2017-01, Vol.19 (1), p.123-135
Main Authors: Chen, Zhixiang, Lu, Jiwen, Feng, Jianjiang, Zhou, Jie
Format: Article
Language:English
Subjects:
Binary code
Binary codes
Data structures
discrete optimization
Errors
hashing
Manuals
Multi-layer neural network
multilayer neural network
nonlinear transformation
Nonlinearity
Optimization
Quantization (signal)
Semantics
Similarity
Transformations
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we propose a nonlinear discrete hashing approach to learn compact binary codes for scalable image search. Instead of seeking a single linear projection in most existing hashing methods, we pursue a multilayer network with nonlinear transformations to capture the local structure of data samples. Unlike most existing hashing methods that adopt an error-prone relaxation to learn the transformations, we directly solve the discrete optimization problem to eliminate the quantization error accumulation. Specifically, to leverage the similarity relationships between data samples and exploit the semantic affinities of manual labels, the binary codes are learned with the objective to: 1) minimize the quantization error between the original data samples and the learned binary codes; 2) preserve the similarity relationships in the learned binary codes; 3) maximize the information content with independent bits; and 4) maximize the accuracy of the predicted labels based on the binary codes. With an alternating optimization, the nonlinear transformation and the discrete quantization are jointly optimized in the hashing learning framework. Experimental results on four datasets including CIFAR10, MNIST, SUN397, and ILSVRC2012 demonstrate that the proposed approach is superior to several state-of-the-art hashing methods.
ISSN:1520-9210
1941-0077
DOI:10.1109/TMM.2016.2620604