A new vector space model based on the deep learning

Hanen Karamti, Mohamed Tmar, Faiez Gargouri

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

Deep learning has become one of the top performing methods for many computer vision tasks such as images retrieval. It has been deployed so far to bring improvements to learning feature representations and similarity measures. In this article, we present a new search method to represent and to retrieve images based on the vector space method, called vectorization. This method transforms any matching model of images to a vector space model providing a score using the Convolutional Neural Networks (CNN). The results obtained by this model are illustrated through some experiments and compared with several state-of-art methods.

Original languageEnglish
Title of host publicationNeural Information Processing - 24th International Conference, ICONIP 2017, Proceedings
PublisherSpringer Verlag
Pages750-758
Number of pages9
ISBN (Print)9783319701356
DOIs
Publication statusPublished - 1 Jan 2017
Event24th International Conference on Neural Information Processing, ICONIP 2017 - Guangzhou, China
Duration: 14 Nov 201718 Nov 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10639 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other24th International Conference on Neural Information Processing, ICONIP 2017
CountryChina
CityGuangzhou
Period14/11/1718/11/17

Fingerprint

Vector Space Model
Vector spaces
Model-based
Image retrieval
Vectorization
Model Matching
Computer vision
Image Retrieval
Similarity Measure
Search Methods
Computer Vision
Neural networks
Vector space
Neural Networks
Transform
Learning
Deep learning
Experiments
Experiment

Keywords

  • CNN
  • Convolutional neural networks
  • Deep learning
  • Image
  • Logistic regression
  • Vector space model

Cite this

Karamti, H., Tmar, M., & Gargouri, F. (2017). A new vector space model based on the deep learning. In Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings (pp. 750-758). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10639 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-70136-3_79
Karamti, Hanen ; Tmar, Mohamed ; Gargouri, Faiez. / A new vector space model based on the deep learning. Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings. Springer Verlag, 2017. pp. 750-758 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{b420e55bd16141e5a132cc0d75888857,
title = "A new vector space model based on the deep learning",
abstract = "Deep learning has become one of the top performing methods for many computer vision tasks such as images retrieval. It has been deployed so far to bring improvements to learning feature representations and similarity measures. In this article, we present a new search method to represent and to retrieve images based on the vector space method, called vectorization. This method transforms any matching model of images to a vector space model providing a score using the Convolutional Neural Networks (CNN). The results obtained by this model are illustrated through some experiments and compared with several state-of-art methods.",
keywords = "CNN, Convolutional neural networks, Deep learning, Image, Logistic regression, Vector space model",
author = "Hanen Karamti and Mohamed Tmar and Faiez Gargouri",
year = "2017",
month = "1",
day = "1",
doi = "10.1007/978-3-319-70136-3_79",
language = "English",
isbn = "9783319701356",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "750--758",
booktitle = "Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings",
address = "Germany",

}

Karamti, H, Tmar, M & Gargouri, F 2017, A new vector space model based on the deep learning. in Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10639 LNCS, Springer Verlag, pp. 750-758, 24th International Conference on Neural Information Processing, ICONIP 2017, Guangzhou, China, 14/11/17. https://doi.org/10.1007/978-3-319-70136-3_79

A new vector space model based on the deep learning. / Karamti, Hanen; Tmar, Mohamed; Gargouri, Faiez.

Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings. Springer Verlag, 2017. p. 750-758 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10639 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

TY - GEN

T1 - A new vector space model based on the deep learning

AU - Karamti, Hanen

AU - Tmar, Mohamed

AU - Gargouri, Faiez

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Deep learning has become one of the top performing methods for many computer vision tasks such as images retrieval. It has been deployed so far to bring improvements to learning feature representations and similarity measures. In this article, we present a new search method to represent and to retrieve images based on the vector space method, called vectorization. This method transforms any matching model of images to a vector space model providing a score using the Convolutional Neural Networks (CNN). The results obtained by this model are illustrated through some experiments and compared with several state-of-art methods.

AB - Deep learning has become one of the top performing methods for many computer vision tasks such as images retrieval. It has been deployed so far to bring improvements to learning feature representations and similarity measures. In this article, we present a new search method to represent and to retrieve images based on the vector space method, called vectorization. This method transforms any matching model of images to a vector space model providing a score using the Convolutional Neural Networks (CNN). The results obtained by this model are illustrated through some experiments and compared with several state-of-art methods.

KW - CNN

KW - Convolutional neural networks

KW - Deep learning

KW - Image

KW - Logistic regression

KW - Vector space model

UR - http://www.scopus.com/inward/record.url?scp=85035125376&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-70136-3_79

DO - 10.1007/978-3-319-70136-3_79

M3 - Conference contribution

SN - 9783319701356

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 750

EP - 758

BT - Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings

PB - Springer Verlag

ER -

Karamti H, Tmar M, Gargouri F. A new vector space model based on the deep learning. In Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings. Springer Verlag. 2017. p. 750-758. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-70136-3_79