Using very deep autoencoders for content-based image retrieval.

Published 1 January 2011

Alex Krizhevsky, Geoffrey E. Hinton

Citations375

TL;DR

This work shows how to learn many layers of features on color images and how these features are used to initialize deep autoencoders, which are then used to map images to short binary codes.

Abstract

Abstract. We show how to learn many layers of features on color images and we use these features to initialize deep autoencoders. We then use the autoencoders to map images to short binary codes. Using semantic hashing [6], 28-bit codes can be used to retrieve images that are similar to a query image in a time that is independent of the size of the database. This extremely fast retrieval makes it possible to search using multiple di erent transformations of the query image. 256-bit binary codes allow much more accurate matching and can be used to prune the set of images found using the 28-bit codes. 1

Keywords

Computer Science

Learning Multiple Layers of Features from Tiny Images

25,432 Citations2024Alex Krizhevsky

View PDF

ScienceReducing the Dimensionality of Data with Neural Networks

20,953 Citations2006Geoffrey E. Hinton, Ruslan Salakhutdinov

Neural ComputationA Fast Learning Algorithm for Deep Belief Nets

16,404 Citations2006Geoffrey E. Hinton, Simon Osindero +1 more

A fast, greedy algorithm is derived that can learn deep, directed belief networks one layer at a time, provided the top two layers form an undirected associative memory.

Lecture notes in computer scienceA Practical Guide to Training Restricted Boltzmann Machines

2,802 Citations2012Geoffrey E. Hinton

This guide is an attempt to share expertise at training restricted Boltzmann machines with other machine learning researchers.

neural information processing systemsSpectral Hashing

2,083 Citations2008Yair Weiss, Antonio Torralba +1 more

This paper shows that the problem of finding a best code for a given dataset is closely related to the problem of graph partitioning and can be shown to be NP hard and obtains a spectral method whose solutions are simply a subset of thresholded eigenvectors of the graph Laplacian.

International Journal of Approximate ReasoningSemantic hashing

1,272 Citations2008Ruslan Salakhutdinov, Geoffrey E. Hinton

A deep graphical model of the word-count vectors obtained from a large set of documents is shown, which achieves higher accuracy than applying TF-IDF to the entire document set.

Small codes and large image databases for recognition

739 Citations2008Antonio Torralba, Rob Fergus +1 more

The goal is to develop efficient image search and scene matching techniques that are not only fast, but also require very little memory, enabling their use on standard hardware or even on handheld devices.

Learning a Nonlinear Embedding by Preserving Class Neighbourhood Structure

438 Citations2007Ruslan Salakhutdinov, Geoffrey E. Hinton