Proceedings of the Advances in Materials, Machinery, Electrical Engineering (AMMEE 2017)

Improve VLAD Using the Entropy Produced by BOW

Authors
Hongwei Zhao, Yeran Wang, Pingping Liu, Chaoran Zhao, Xiang Li
Corresponding Author
Hongwei Zhao
Available Online June 2017.
DOI
10.2991/ammee-17.2017.135How to use a DOI?
Keywords
Image retrieval, VLAD, BOW, entropy.
Abstract

VLAD (vector of locally aggregated descriptors) is a type of global features extracted from the image, which is always used in image retrieval. Although VLAD is effective, it still needs to improve. VLAD is obtained by accumulating residuals, ignoring the number information of descriptors in a cluster. BOW (bag of words) is also a type of features, which describe the amounts of the descriptors within a cluster. In this paper, the concept of the quantity entropy is proposed based on the BOW method. The amount of descriptors is processed by the calculation steps of entropy to obtain the quantity entropy, and we add the quantity entropy to VLAD, which makes the image retrieval ability of VLAD improved.

Copyright
© 2017, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

Volume Title
Proceedings of the Advances in Materials, Machinery, Electrical Engineering (AMMEE 2017)
Series
Advances in Engineering Research
Publication Date
June 2017
ISBN
978-94-6252-350-0
ISSN
2352-5401
DOI
10.2991/ammee-17.2017.135How to use a DOI?
Copyright
© 2017, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Cite this article

TY  - CONF
AU  - Hongwei Zhao
AU  - Yeran Wang
AU  - Pingping Liu
AU  - Chaoran Zhao
AU  - Xiang Li
PY  - 2017/06
DA  - 2017/06
TI  - Improve VLAD Using the Entropy Produced by BOW
BT  - Proceedings of the Advances in Materials, Machinery, Electrical Engineering (AMMEE 2017)
PB  - Atlantis Press
SP  - 702
EP  - 705
SN  - 2352-5401
UR  - https://doi.org/10.2991/ammee-17.2017.135
DO  - 10.2991/ammee-17.2017.135
ID  - Zhao2017/06
ER  -