The Role of Attention Mechanism in Generating Image Captions: An Innovative Approach with Neural Network-Based Seq2seq Model

Image-to-text generation contributes significantly across various domains such as entertainment, communication, commerce, security, and education by establishing a connection between visual and textual content through the creation of explanations. This process aims to transform image data into meani...

Full description

Saved in:
Bibliographic Details
Main Authors: Zeynep Karaca, Bihter Daş
Format: Article
Language:English
Published: Sakarya University 2024-04-01
Series:Sakarya University Journal of Computer and Information Sciences
Subjects:
Online Access:https://dergipark.org.tr/en/download/article-file/3317713
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850146205955260416
author Zeynep Karaca
Bihter Daş
author_facet Zeynep Karaca
Bihter Daş
author_sort Zeynep Karaca
collection DOAJ
description Image-to-text generation contributes significantly across various domains such as entertainment, communication, commerce, security, and education by establishing a connection between visual and textual content through the creation of explanations. This process aims to transform image data into meaningful text, enhancing content accessibility, comprehensibility, and processability. Hence, advancements and studies in this field hold paramount importance. This study focuses on how the fusion of the Sequence-to-Sequence (Seq2seq) model and attention mechanism enhances the generation of more meaningful captions from images. Experiments conducted on the Flickr8k dataset highlight the Seq2seq model's capacity to produce captions in alignment with reference sentences. Leveraging the dynamic focus of the attention mechanism, the model effectively captures detailed aspects of images.
format Article
id doaj-art-66ce7166af23450fad36d3d36253cf8a
institution OA Journals
issn 2636-8129
language English
publishDate 2024-04-01
publisher Sakarya University
record_format Article
series Sakarya University Journal of Computer and Information Sciences
spelling doaj-art-66ce7166af23450fad36d3d36253cf8a2025-08-20T02:27:54ZengSakarya UniversitySakarya University Journal of Computer and Information Sciences2636-81292024-04-01719210210.35377/saucis...133993128The Role of Attention Mechanism in Generating Image Captions: An Innovative Approach with Neural Network-Based Seq2seq ModelZeynep Karaca0https://orcid.org/0000-0002-7751-8567Bihter Daş1https://orcid.org/0000-0002-2498-3297FIRAT ÜNİVERSİTESİ, TEKNOLOJİ FAKÜLTESİFIRAT ÜNİVERSİTESİImage-to-text generation contributes significantly across various domains such as entertainment, communication, commerce, security, and education by establishing a connection between visual and textual content through the creation of explanations. This process aims to transform image data into meaningful text, enhancing content accessibility, comprehensibility, and processability. Hence, advancements and studies in this field hold paramount importance. This study focuses on how the fusion of the Sequence-to-Sequence (Seq2seq) model and attention mechanism enhances the generation of more meaningful captions from images. Experiments conducted on the Flickr8k dataset highlight the Seq2seq model's capacity to produce captions in alignment with reference sentences. Leveraging the dynamic focus of the attention mechanism, the model effectively captures detailed aspects of images.https://dergipark.org.tr/en/download/article-file/3317713attention mechanismseq2seg modelimage capturingdeep learningimage-to-text
spellingShingle Zeynep Karaca
Bihter Daş
The Role of Attention Mechanism in Generating Image Captions: An Innovative Approach with Neural Network-Based Seq2seq Model
Sakarya University Journal of Computer and Information Sciences
attention mechanism
seq2seg model
image capturing
deep learning
image-to-text
title The Role of Attention Mechanism in Generating Image Captions: An Innovative Approach with Neural Network-Based Seq2seq Model
title_full The Role of Attention Mechanism in Generating Image Captions: An Innovative Approach with Neural Network-Based Seq2seq Model
title_fullStr The Role of Attention Mechanism in Generating Image Captions: An Innovative Approach with Neural Network-Based Seq2seq Model
title_full_unstemmed The Role of Attention Mechanism in Generating Image Captions: An Innovative Approach with Neural Network-Based Seq2seq Model
title_short The Role of Attention Mechanism in Generating Image Captions: An Innovative Approach with Neural Network-Based Seq2seq Model
title_sort role of attention mechanism in generating image captions an innovative approach with neural network based seq2seq model
topic attention mechanism
seq2seg model
image capturing
deep learning
image-to-text
url https://dergipark.org.tr/en/download/article-file/3317713
work_keys_str_mv AT zeynepkaraca theroleofattentionmechanismingeneratingimagecaptionsaninnovativeapproachwithneuralnetworkbasedseq2seqmodel
AT bihterdas theroleofattentionmechanismingeneratingimagecaptionsaninnovativeapproachwithneuralnetworkbasedseq2seqmodel
AT zeynepkaraca roleofattentionmechanismingeneratingimagecaptionsaninnovativeapproachwithneuralnetworkbasedseq2seqmodel
AT bihterdas roleofattentionmechanismingeneratingimagecaptionsaninnovativeapproachwithneuralnetworkbasedseq2seqmodel