Bahdanau Attention Based Bengali Image Caption Generation

M. M. Alam, M. Rahman, M. Hosen, Khairul Anam Mubin, S. Hossen, M. F. Mridha
{"title":"Bahdanau Attention Based Bengali Image Caption Generation","authors":"M. M. Alam, M. Rahman, M. Hosen, Khairul Anam Mubin, S. Hossen, M. F. Mridha","doi":"10.1109/DASA54658.2022.9765268","DOIUrl":null,"url":null,"abstract":"In the past few years, many works are done in object detection using images and machine translation. Inspired by those works we introduced Bahdanau Attention Based Bengali Image Caption Generation (BABBICG) that generate automatically bangla caption based on images. The Conventional encoder-decoder architectures performance curse will reduce by Bahdanau Attention and achieving momentous improvements over encoder-decoder architectures. In this work, we extract features from images using InceptionV3 neural network and generate caption using RNN decoder. We used Gated Recurrent Unit (GRU) approach as RNN. We evaluate the model using BanglaLekhaImageCaptions dataset from Mendeley Data that can help to generate bangla caption.","PeriodicalId":231066,"journal":{"name":"2022 International Conference on Decision Aid Sciences and Applications (DASA)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 International Conference on Decision Aid Sciences and Applications (DASA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/DASA54658.2022.9765268","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

In the past few years, many works are done in object detection using images and machine translation. Inspired by those works we introduced Bahdanau Attention Based Bengali Image Caption Generation (BABBICG) that generate automatically bangla caption based on images. The Conventional encoder-decoder architectures performance curse will reduce by Bahdanau Attention and achieving momentous improvements over encoder-decoder architectures. In this work, we extract features from images using InceptionV3 neural network and generate caption using RNN decoder. We used Gated Recurrent Unit (GRU) approach as RNN. We evaluate the model using BanglaLekhaImageCaptions dataset from Mendeley Data that can help to generate bangla caption.
基于Bahdanau注意的孟加拉语图像标题生成
在过去的几年里,很多工作都是利用图像和机器翻译来进行目标检测的。受这些作品的启发,我们引入了基于Bahdanau注意力的孟加拉语图像标题生成(BABBICG),它可以根据图像自动生成孟加拉语标题。传统的编码器-解码器体系结构的性能缺陷将随着巴赫达瑙关注的减少而得到显著改善。在这项工作中,我们使用InceptionV3神经网络从图像中提取特征,并使用RNN解码器生成标题。我们使用门控循环单元(GRU)方法作为RNN。我们使用Mendeley Data的BanglaLekhaImageCaptions数据集来评估模型,该数据集可以帮助生成孟加拉语标题。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信