Fast content-based image retrieval using convolutional neural network and hash function

Domonkos Varga, T. Szirányi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Citations (Scopus)

Abstract

Due to the explosive increase of online images, content-based image retrieval has gained a lot of attention. The success of deep learning techniques such as convolutional neural networks have motivated us to explore its applications in our context. The main contribution of our work is a novel end-to-end supervised learning framework that learns probability-based semantic-level similarity and feature-level similarity simultaneously. The main advantage of our novel hashing scheme that it is able to reduce the computational cost of retrieval significantly at the state-of-the-art efficiency level. We report on comprehensive experiments using public available datasets such as Oxford, Holidays and ImageNet 2012 retrieval datasets.

Original languageEnglish
Title of host publication2016 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2016 - Conference Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages2636-2640
Number of pages5
ISBN (Electronic)9781509018970
DOIs
Publication statusPublished - Feb 6 2017
Event2016 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2016 - Budapest, Hungary
Duration: Oct 9 2016Oct 12 2016

Other

Other2016 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2016
CountryHungary
CityBudapest
Period10/9/1610/12/16

Fingerprint

Hash functions
Content-based Image Retrieval
Supervised learning
Image retrieval
Hash Function
Retrieval
Semantics
Neural Networks
Neural networks
Hashing
Supervised Learning
Computational Cost
Costs
Experiments
Experiment
Similarity
Deep learning
Framework
Learning
Context

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Artificial Intelligence
  • Control and Optimization
  • Human-Computer Interaction

Cite this

Varga, D., & Szirányi, T. (2017). Fast content-based image retrieval using convolutional neural network and hash function. In 2016 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2016 - Conference Proceedings (pp. 2636-2640). [7844637] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SMC.2016.7844637

Fast content-based image retrieval using convolutional neural network and hash function. / Varga, Domonkos; Szirányi, T.

2016 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2016 - Conference Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. p. 2636-2640 7844637.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Varga, D & Szirányi, T 2017, Fast content-based image retrieval using convolutional neural network and hash function. in 2016 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2016 - Conference Proceedings., 7844637, Institute of Electrical and Electronics Engineers Inc., pp. 2636-2640, 2016 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2016, Budapest, Hungary, 10/9/16. https://doi.org/10.1109/SMC.2016.7844637
Varga D, Szirányi T. Fast content-based image retrieval using convolutional neural network and hash function. In 2016 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2016 - Conference Proceedings. Institute of Electrical and Electronics Engineers Inc. 2017. p. 2636-2640. 7844637 https://doi.org/10.1109/SMC.2016.7844637
Varga, Domonkos ; Szirányi, T. / Fast content-based image retrieval using convolutional neural network and hash function. 2016 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2016 - Conference Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 2636-2640
@inproceedings{c513123a77a24f4db9f095c0debd0a56,
title = "Fast content-based image retrieval using convolutional neural network and hash function",
abstract = "Due to the explosive increase of online images, content-based image retrieval has gained a lot of attention. The success of deep learning techniques such as convolutional neural networks have motivated us to explore its applications in our context. The main contribution of our work is a novel end-to-end supervised learning framework that learns probability-based semantic-level similarity and feature-level similarity simultaneously. The main advantage of our novel hashing scheme that it is able to reduce the computational cost of retrieval significantly at the state-of-the-art efficiency level. We report on comprehensive experiments using public available datasets such as Oxford, Holidays and ImageNet 2012 retrieval datasets.",
author = "Domonkos Varga and T. Szir{\'a}nyi",
year = "2017",
month = "2",
day = "6",
doi = "10.1109/SMC.2016.7844637",
language = "English",
pages = "2636--2640",
booktitle = "2016 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2016 - Conference Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Fast content-based image retrieval using convolutional neural network and hash function

AU - Varga, Domonkos

AU - Szirányi, T.

PY - 2017/2/6

Y1 - 2017/2/6

N2 - Due to the explosive increase of online images, content-based image retrieval has gained a lot of attention. The success of deep learning techniques such as convolutional neural networks have motivated us to explore its applications in our context. The main contribution of our work is a novel end-to-end supervised learning framework that learns probability-based semantic-level similarity and feature-level similarity simultaneously. The main advantage of our novel hashing scheme that it is able to reduce the computational cost of retrieval significantly at the state-of-the-art efficiency level. We report on comprehensive experiments using public available datasets such as Oxford, Holidays and ImageNet 2012 retrieval datasets.

AB - Due to the explosive increase of online images, content-based image retrieval has gained a lot of attention. The success of deep learning techniques such as convolutional neural networks have motivated us to explore its applications in our context. The main contribution of our work is a novel end-to-end supervised learning framework that learns probability-based semantic-level similarity and feature-level similarity simultaneously. The main advantage of our novel hashing scheme that it is able to reduce the computational cost of retrieval significantly at the state-of-the-art efficiency level. We report on comprehensive experiments using public available datasets such as Oxford, Holidays and ImageNet 2012 retrieval datasets.

UR - http://www.scopus.com/inward/record.url?scp=85015726916&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85015726916&partnerID=8YFLogxK

U2 - 10.1109/SMC.2016.7844637

DO - 10.1109/SMC.2016.7844637

M3 - Conference contribution

AN - SCOPUS:85015726916

SP - 2636

EP - 2640

BT - 2016 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2016 - Conference Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -