Tampere University of Technology

TUTCRIS Research Portal

Simultenious binary hash and features learning for image retrieval

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

Standard

Simultenious binary hash and features learning for image retrieval. / Frantc, V. A.; Makov, S. V.; Voronin, V. V.; Marchuk, V. I.; Semenishchev, E. A.; Egiazarian, K. O.; Agaian, S.

Mobile Multimedia/Image Processing, Security, and Applications 2016. SPIE, 2016. 986902 (SPIE Conference Proceedings; Vol. 9869).

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

Harvard

Frantc, VA, Makov, SV, Voronin, VV, Marchuk, VI, Semenishchev, EA, Egiazarian, KO & Agaian, S 2016, Simultenious binary hash and features learning for image retrieval. in Mobile Multimedia/Image Processing, Security, and Applications 2016., 986902, SPIE Conference Proceedings, vol. 9869, SPIE, Mobile Multimedia/Image Processing, Security, and Applications, 1/01/00. https://doi.org/10.1117/12.2223605

APA

Frantc, V. A., Makov, S. V., Voronin, V. V., Marchuk, V. I., Semenishchev, E. A., Egiazarian, K. O., & Agaian, S. (2016). Simultenious binary hash and features learning for image retrieval. In Mobile Multimedia/Image Processing, Security, and Applications 2016 [986902] (SPIE Conference Proceedings; Vol. 9869). SPIE. https://doi.org/10.1117/12.2223605

Vancouver

Frantc VA, Makov SV, Voronin VV, Marchuk VI, Semenishchev EA, Egiazarian KO et al. Simultenious binary hash and features learning for image retrieval. In Mobile Multimedia/Image Processing, Security, and Applications 2016. SPIE. 2016. 986902. (SPIE Conference Proceedings). https://doi.org/10.1117/12.2223605

Author

Frantc, V. A. ; Makov, S. V. ; Voronin, V. V. ; Marchuk, V. I. ; Semenishchev, E. A. ; Egiazarian, K. O. ; Agaian, S. / Simultenious binary hash and features learning for image retrieval. Mobile Multimedia/Image Processing, Security, and Applications 2016. SPIE, 2016. (SPIE Conference Proceedings).

Bibtex - Download

@inproceedings{feeabfd6333d4af78446b96ad5f748c4,
title = "Simultenious binary hash and features learning for image retrieval",
abstract = "Content-based image retrieval systems have plenty of applications in modern world. The most important one is the image search by query image or by semantic description. Approaches to this problem are employed in personal photo-collection management systems, web-scale image search engines, medical systems, etc. Automatic analysis of large unlabeled image datasets is virtually impossible without satisfactory image-retrieval technique. It's the main reason why this kind of automatic image processing has attracted so much attention during recent years. Despite rather huge progress in the field, semantically meaningful image retrieval still remains a challenging task. The main issue here is the demand to provide reliable results in short amount of time. This paper addresses the problem by novel technique for simultaneous learning of global image features and binary hash codes. Our approach provide mapping of pixel-based image representation to hash-value space simultaneously trying to save as much of semantic image content as possible. We use deep learning methodology to generate image description with properties of similarity preservation and statistical independence. The main advantage of our approach in contrast to existing is ability to fine-tune retrieval procedure for very specific application which allow us to provide better results in comparison to general techniques. Presented in the paper framework for data- dependent image hashing is based on use two different kinds of neural networks: convolutional neural networks for image description and autoencoder for feature to hash space mapping. Experimental results confirmed that our approach has shown promising results in compare to other state-of-the-art methods.",
keywords = "autoencoder, content{based image retrieval, deep convolutional neural network, semantic hashing",
author = "Frantc, {V. A.} and Makov, {S. V.} and Voronin, {V. V.} and Marchuk, {V. I.} and Semenishchev, {E. A.} and Egiazarian, {K. O.} and S. Agaian",
year = "2016",
doi = "10.1117/12.2223605",
language = "English",
series = "SPIE Conference Proceedings",
publisher = "SPIE",
booktitle = "Mobile Multimedia/Image Processing, Security, and Applications 2016",
address = "United States",

}

RIS (suitable for import to EndNote) - Download

TY - GEN

T1 - Simultenious binary hash and features learning for image retrieval

AU - Frantc, V. A.

AU - Makov, S. V.

AU - Voronin, V. V.

AU - Marchuk, V. I.

AU - Semenishchev, E. A.

AU - Egiazarian, K. O.

AU - Agaian, S.

PY - 2016

Y1 - 2016

N2 - Content-based image retrieval systems have plenty of applications in modern world. The most important one is the image search by query image or by semantic description. Approaches to this problem are employed in personal photo-collection management systems, web-scale image search engines, medical systems, etc. Automatic analysis of large unlabeled image datasets is virtually impossible without satisfactory image-retrieval technique. It's the main reason why this kind of automatic image processing has attracted so much attention during recent years. Despite rather huge progress in the field, semantically meaningful image retrieval still remains a challenging task. The main issue here is the demand to provide reliable results in short amount of time. This paper addresses the problem by novel technique for simultaneous learning of global image features and binary hash codes. Our approach provide mapping of pixel-based image representation to hash-value space simultaneously trying to save as much of semantic image content as possible. We use deep learning methodology to generate image description with properties of similarity preservation and statistical independence. The main advantage of our approach in contrast to existing is ability to fine-tune retrieval procedure for very specific application which allow us to provide better results in comparison to general techniques. Presented in the paper framework for data- dependent image hashing is based on use two different kinds of neural networks: convolutional neural networks for image description and autoencoder for feature to hash space mapping. Experimental results confirmed that our approach has shown promising results in compare to other state-of-the-art methods.

AB - Content-based image retrieval systems have plenty of applications in modern world. The most important one is the image search by query image or by semantic description. Approaches to this problem are employed in personal photo-collection management systems, web-scale image search engines, medical systems, etc. Automatic analysis of large unlabeled image datasets is virtually impossible without satisfactory image-retrieval technique. It's the main reason why this kind of automatic image processing has attracted so much attention during recent years. Despite rather huge progress in the field, semantically meaningful image retrieval still remains a challenging task. The main issue here is the demand to provide reliable results in short amount of time. This paper addresses the problem by novel technique for simultaneous learning of global image features and binary hash codes. Our approach provide mapping of pixel-based image representation to hash-value space simultaneously trying to save as much of semantic image content as possible. We use deep learning methodology to generate image description with properties of similarity preservation and statistical independence. The main advantage of our approach in contrast to existing is ability to fine-tune retrieval procedure for very specific application which allow us to provide better results in comparison to general techniques. Presented in the paper framework for data- dependent image hashing is based on use two different kinds of neural networks: convolutional neural networks for image description and autoencoder for feature to hash space mapping. Experimental results confirmed that our approach has shown promising results in compare to other state-of-the-art methods.

KW - autoencoder

KW - content{based image retrieval

KW - deep convolutional neural network

KW - semantic hashing

U2 - 10.1117/12.2223605

DO - 10.1117/12.2223605

M3 - Conference contribution

T3 - SPIE Conference Proceedings

BT - Mobile Multimedia/Image Processing, Security, and Applications 2016

PB - SPIE

ER -