Automatic image description by using word-level features

Shingo Horiuchi, Hirotaka Moriguchi, Shinichi Honiden, Shengbo Xu

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

Abstract

Automatic image description is one of the challenging tasks of image recognitions. However, there are image descriptions that contain some too specific phrases that cannot be judged only from appearance of images. In this paper, we propose a novel approach to collect general phrases for generating image descriptions. On the assumption that there are high frequency phrases related to an query image in the image descriptions of similar images, we select nouns and their attribute phrases from the image descriptions of similar images based on their frequency. In order to evaluate the relevance of our image description, we conduct comparative experiments with existing approaches. Our experimental results show that our image descriptions are short, concise and visually relevant to query images.

Original languageEnglish
Title of host publicationICIMCS 2013 - Proceedings of the 5th International Conference on Internet Multimedia Computing and Service
Pages309-314
Number of pages6
DOIs
Publication statusPublished - 2013 Sep 16
Externally publishedYes
Event5th International Conference on Internet Multimedia Computing and Service, ICIMCS 2013 - Huangshan, China
Duration: 2013 Aug 172013 Aug 19

Other

Other5th International Conference on Internet Multimedia Computing and Service, ICIMCS 2013
CountryChina
CityHuangshan
Period13/8/1713/8/19

Fingerprint

Image recognition
Experiments

Keywords

  • automatic image annotation
  • image description
  • image recognition

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Software

Cite this

Horiuchi, S., Moriguchi, H., Honiden, S., & Xu, S. (2013). Automatic image description by using word-level features. In ICIMCS 2013 - Proceedings of the 5th International Conference on Internet Multimedia Computing and Service (pp. 309-314) https://doi.org/10.1145/2499788.2499823

Automatic image description by using word-level features. / Horiuchi, Shingo; Moriguchi, Hirotaka; Honiden, Shinichi; Xu, Shengbo.

ICIMCS 2013 - Proceedings of the 5th International Conference on Internet Multimedia Computing and Service. 2013. p. 309-314.

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

Horiuchi, S, Moriguchi, H, Honiden, S & Xu, S 2013, Automatic image description by using word-level features. in ICIMCS 2013 - Proceedings of the 5th International Conference on Internet Multimedia Computing and Service. pp. 309-314, 5th International Conference on Internet Multimedia Computing and Service, ICIMCS 2013, Huangshan, China, 13/8/17. https://doi.org/10.1145/2499788.2499823
Horiuchi S, Moriguchi H, Honiden S, Xu S. Automatic image description by using word-level features. In ICIMCS 2013 - Proceedings of the 5th International Conference on Internet Multimedia Computing and Service. 2013. p. 309-314 https://doi.org/10.1145/2499788.2499823
Horiuchi, Shingo ; Moriguchi, Hirotaka ; Honiden, Shinichi ; Xu, Shengbo. / Automatic image description by using word-level features. ICIMCS 2013 - Proceedings of the 5th International Conference on Internet Multimedia Computing and Service. 2013. pp. 309-314
@inproceedings{0ab333f1c64741628097d92b47d6a2b8,
title = "Automatic image description by using word-level features",
abstract = "Automatic image description is one of the challenging tasks of image recognitions. However, there are image descriptions that contain some too specific phrases that cannot be judged only from appearance of images. In this paper, we propose a novel approach to collect general phrases for generating image descriptions. On the assumption that there are high frequency phrases related to an query image in the image descriptions of similar images, we select nouns and their attribute phrases from the image descriptions of similar images based on their frequency. In order to evaluate the relevance of our image description, we conduct comparative experiments with existing approaches. Our experimental results show that our image descriptions are short, concise and visually relevant to query images.",
keywords = "automatic image annotation, image description, image recognition",
author = "Shingo Horiuchi and Hirotaka Moriguchi and Shinichi Honiden and Shengbo Xu",
year = "2013",
month = "9",
day = "16",
doi = "10.1145/2499788.2499823",
language = "English",
isbn = "9781450322522",
pages = "309--314",
booktitle = "ICIMCS 2013 - Proceedings of the 5th International Conference on Internet Multimedia Computing and Service",

}

TY - GEN

T1 - Automatic image description by using word-level features

AU - Horiuchi, Shingo

AU - Moriguchi, Hirotaka

AU - Honiden, Shinichi

AU - Xu, Shengbo

PY - 2013/9/16

Y1 - 2013/9/16

N2 - Automatic image description is one of the challenging tasks of image recognitions. However, there are image descriptions that contain some too specific phrases that cannot be judged only from appearance of images. In this paper, we propose a novel approach to collect general phrases for generating image descriptions. On the assumption that there are high frequency phrases related to an query image in the image descriptions of similar images, we select nouns and their attribute phrases from the image descriptions of similar images based on their frequency. In order to evaluate the relevance of our image description, we conduct comparative experiments with existing approaches. Our experimental results show that our image descriptions are short, concise and visually relevant to query images.

AB - Automatic image description is one of the challenging tasks of image recognitions. However, there are image descriptions that contain some too specific phrases that cannot be judged only from appearance of images. In this paper, we propose a novel approach to collect general phrases for generating image descriptions. On the assumption that there are high frequency phrases related to an query image in the image descriptions of similar images, we select nouns and their attribute phrases from the image descriptions of similar images based on their frequency. In order to evaluate the relevance of our image description, we conduct comparative experiments with existing approaches. Our experimental results show that our image descriptions are short, concise and visually relevant to query images.

KW - automatic image annotation

KW - image description

KW - image recognition

UR - http://www.scopus.com/inward/record.url?scp=84883687652&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84883687652&partnerID=8YFLogxK

U2 - 10.1145/2499788.2499823

DO - 10.1145/2499788.2499823

M3 - Conference contribution

AN - SCOPUS:84883687652

SN - 9781450322522

SP - 309

EP - 314

BT - ICIMCS 2013 - Proceedings of the 5th International Conference on Internet Multimedia Computing and Service

ER -