Image description generation without image processing using fuzzy inference

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

1 Citation (Scopus)

Abstract

We propose a sentence generation method that describes images. We do not use image processing technique in our proposed method. Human annotated image tags are used as image information to generate sentence. By using human annotated tags, we think this enables to describe image more relevant and user specific. Our method uses Kyoto University's case frame data and Google N-gram to generate candidate sentences. We extend these candidates to describe images more relevant. To be more precise, we added segments with missing semantic role, and added modification segments. To select one output sentence, we used fuzzy rules to grade naturalness of candidate sentences. To grading image relevance of the sentence, we scored word similarity for each word. The performance of the proposed system has been evaluated by subjective experiments and obtained satisfactory results.

Original languageEnglish
Title of host publicationIEEE International Conference on Fuzzy Systems
DOIs
Publication statusPublished - 2012
Event2012 IEEE International Conference on Fuzzy Systems, FUZZ 2012 - Brisbane, QLD, Australia
Duration: 2012 Jun 102012 Jun 15

Other

Other2012 IEEE International Conference on Fuzzy Systems, FUZZ 2012
CountryAustralia
CityBrisbane, QLD
Period12/6/1012/6/15

Fingerprint

Fuzzy Inference
Fuzzy inference
Fuzzy rules
Image Processing
Image processing
Semantics
Experiments
N-gram
Grading
Fuzzy Rules
Output
Experiment

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence
  • Applied Mathematics
  • Theoretical Computer Science

Cite this

Image description generation without image processing using fuzzy inference. / Ito, Naho; Hagiwara, Masafumi.

IEEE International Conference on Fuzzy Systems. 2012. 6250835.

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

Ito, N & Hagiwara, M 2012, Image description generation without image processing using fuzzy inference. in IEEE International Conference on Fuzzy Systems., 6250835, 2012 IEEE International Conference on Fuzzy Systems, FUZZ 2012, Brisbane, QLD, Australia, 12/6/10. https://doi.org/10.1109/FUZZ-IEEE.2012.6250835
Ito, Naho ; Hagiwara, Masafumi. / Image description generation without image processing using fuzzy inference. IEEE International Conference on Fuzzy Systems. 2012.
@inproceedings{7cf6c73067e44e0090c45025d3ee4aba,
title = "Image description generation without image processing using fuzzy inference",
abstract = "We propose a sentence generation method that describes images. We do not use image processing technique in our proposed method. Human annotated image tags are used as image information to generate sentence. By using human annotated tags, we think this enables to describe image more relevant and user specific. Our method uses Kyoto University's case frame data and Google N-gram to generate candidate sentences. We extend these candidates to describe images more relevant. To be more precise, we added segments with missing semantic role, and added modification segments. To select one output sentence, we used fuzzy rules to grade naturalness of candidate sentences. To grading image relevance of the sentence, we scored word similarity for each word. The performance of the proposed system has been evaluated by subjective experiments and obtained satisfactory results.",
author = "Naho Ito and Masafumi Hagiwara",
year = "2012",
doi = "10.1109/FUZZ-IEEE.2012.6250835",
language = "English",
isbn = "9781467315067",
booktitle = "IEEE International Conference on Fuzzy Systems",

}

TY - GEN

T1 - Image description generation without image processing using fuzzy inference

AU - Ito, Naho

AU - Hagiwara, Masafumi

PY - 2012

Y1 - 2012

N2 - We propose a sentence generation method that describes images. We do not use image processing technique in our proposed method. Human annotated image tags are used as image information to generate sentence. By using human annotated tags, we think this enables to describe image more relevant and user specific. Our method uses Kyoto University's case frame data and Google N-gram to generate candidate sentences. We extend these candidates to describe images more relevant. To be more precise, we added segments with missing semantic role, and added modification segments. To select one output sentence, we used fuzzy rules to grade naturalness of candidate sentences. To grading image relevance of the sentence, we scored word similarity for each word. The performance of the proposed system has been evaluated by subjective experiments and obtained satisfactory results.

AB - We propose a sentence generation method that describes images. We do not use image processing technique in our proposed method. Human annotated image tags are used as image information to generate sentence. By using human annotated tags, we think this enables to describe image more relevant and user specific. Our method uses Kyoto University's case frame data and Google N-gram to generate candidate sentences. We extend these candidates to describe images more relevant. To be more precise, we added segments with missing semantic role, and added modification segments. To select one output sentence, we used fuzzy rules to grade naturalness of candidate sentences. To grading image relevance of the sentence, we scored word similarity for each word. The performance of the proposed system has been evaluated by subjective experiments and obtained satisfactory results.

UR - http://www.scopus.com/inward/record.url?scp=84867618953&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867618953&partnerID=8YFLogxK

U2 - 10.1109/FUZZ-IEEE.2012.6250835

DO - 10.1109/FUZZ-IEEE.2012.6250835

M3 - Conference contribution

AN - SCOPUS:84867618953

SN - 9781467315067

BT - IEEE International Conference on Fuzzy Systems

ER -