Context based visual content verification

Martin Lukac, Aigerim Bazarbciyeva, Michitaka Kameyama

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

Abstract

In this paper the intermediary visual content verification method based on multi-level co-occurrences is studied. The co-occurrence statistics are in general used to determine relational properties between objects based on information collected from data. As such these measures are heavily subject to relative number of occurrences and give only limited amount of accuracy when predicting objects in real world. In order to improve the accuracy of this method in the verification task, we include the context information such as location, type of environment etc. In order to train our model we provide new annotated dataset the Advanced Attribute VOC (AAVOC) that contains additional properties of the image. We show that the usage of context greatly improve the accuracy of verification with up to 16% improvement.

Original languageEnglish
Title of host publicationProceedings of the International Conference on Information and Digital Technologies, IDT 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages234-239
Number of pages6
ISBN (Electronic)9781509056880
DOIs
Publication statusPublished - Sep 1 2017
Event2017 International Conference on Information and Digital Technologies, IDT 2017 - Zilina, Slovakia
Duration: Jul 5 2017Jul 7 2017

Conference

Conference2017 International Conference on Information and Digital Technologies, IDT 2017
CountrySlovakia
CityZilina
Period7/5/177/7/17

Fingerprint

occurrences
volatile organic compounds
Volatile organic compounds
Statistics
statistics

ASJC Scopus subject areas

  • Instrumentation
  • Computer Networks and Communications
  • Hardware and Architecture
  • Electrical and Electronic Engineering
  • Safety, Risk, Reliability and Quality

Cite this

Lukac, M., Bazarbciyeva, A., & Kameyama, M. (2017). Context based visual content verification. In Proceedings of the International Conference on Information and Digital Technologies, IDT 2017 (pp. 234-239). [8024302] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/DT.2017.8024302

Context based visual content verification. / Lukac, Martin; Bazarbciyeva, Aigerim; Kameyama, Michitaka.

Proceedings of the International Conference on Information and Digital Technologies, IDT 2017. Institute of Electrical and Electronics Engineers Inc., 2017. p. 234-239 8024302.

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

Lukac, M, Bazarbciyeva, A & Kameyama, M 2017, Context based visual content verification. in Proceedings of the International Conference on Information and Digital Technologies, IDT 2017., 8024302, Institute of Electrical and Electronics Engineers Inc., pp. 234-239, 2017 International Conference on Information and Digital Technologies, IDT 2017, Zilina, Slovakia, 7/5/17. https://doi.org/10.1109/DT.2017.8024302
Lukac M, Bazarbciyeva A, Kameyama M. Context based visual content verification. In Proceedings of the International Conference on Information and Digital Technologies, IDT 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 234-239. 8024302 https://doi.org/10.1109/DT.2017.8024302
Lukac, Martin ; Bazarbciyeva, Aigerim ; Kameyama, Michitaka. / Context based visual content verification. Proceedings of the International Conference on Information and Digital Technologies, IDT 2017. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 234-239
@inproceedings{2e44898d806f4350ae8cfac18d42f002,
title = "Context based visual content verification",
abstract = "In this paper the intermediary visual content verification method based on multi-level co-occurrences is studied. The co-occurrence statistics are in general used to determine relational properties between objects based on information collected from data. As such these measures are heavily subject to relative number of occurrences and give only limited amount of accuracy when predicting objects in real world. In order to improve the accuracy of this method in the verification task, we include the context information such as location, type of environment etc. In order to train our model we provide new annotated dataset the Advanced Attribute VOC (AAVOC) that contains additional properties of the image. We show that the usage of context greatly improve the accuracy of verification with up to 16{\%} improvement.",
author = "Martin Lukac and Aigerim Bazarbciyeva and Michitaka Kameyama",
year = "2017",
month = "9",
day = "1",
doi = "10.1109/DT.2017.8024302",
language = "English",
pages = "234--239",
booktitle = "Proceedings of the International Conference on Information and Digital Technologies, IDT 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - Context based visual content verification

AU - Lukac, Martin

AU - Bazarbciyeva, Aigerim

AU - Kameyama, Michitaka

PY - 2017/9/1

Y1 - 2017/9/1

N2 - In this paper the intermediary visual content verification method based on multi-level co-occurrences is studied. The co-occurrence statistics are in general used to determine relational properties between objects based on information collected from data. As such these measures are heavily subject to relative number of occurrences and give only limited amount of accuracy when predicting objects in real world. In order to improve the accuracy of this method in the verification task, we include the context information such as location, type of environment etc. In order to train our model we provide new annotated dataset the Advanced Attribute VOC (AAVOC) that contains additional properties of the image. We show that the usage of context greatly improve the accuracy of verification with up to 16% improvement.

AB - In this paper the intermediary visual content verification method based on multi-level co-occurrences is studied. The co-occurrence statistics are in general used to determine relational properties between objects based on information collected from data. As such these measures are heavily subject to relative number of occurrences and give only limited amount of accuracy when predicting objects in real world. In order to improve the accuracy of this method in the verification task, we include the context information such as location, type of environment etc. In order to train our model we provide new annotated dataset the Advanced Attribute VOC (AAVOC) that contains additional properties of the image. We show that the usage of context greatly improve the accuracy of verification with up to 16% improvement.

UR - http://www.scopus.com/inward/record.url?scp=85030094947&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85030094947&partnerID=8YFLogxK

U2 - 10.1109/DT.2017.8024302

DO - 10.1109/DT.2017.8024302

M3 - Conference contribution

AN - SCOPUS:85030094947

SP - 234

EP - 239

BT - Proceedings of the International Conference on Information and Digital Technologies, IDT 2017

PB - Institute of Electrical and Electronics Engineers Inc.

ER -