Development of Hybrid Image Caption Generation Method using Deep Learning

The image captioning is a process of generating descriptive sentence for a given image in human understandable language and such sentence is known as caption of the image. The automatic image caption generated is a result of deep analysis performed on the image which involves detecting objects in th...

Full description

Saved in:
Bibliographic Details
Main Authors: Namdev, Anuja, Reddy, S.R.N.
Format: Conference Proceeding
Language:English
Subjects:
Artificial intelligence
Computer vision
Deep learning
Hybrid power systems
image captioning
Natural language processing
Signal processing
Visualization
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The image captioning is a process of generating descriptive sentence for a given image in human understandable language and such sentence is known as caption of the image. The automatic image caption generated is a result of deep analysis performed on the image which involves detecting objects in the image as well as the relationship between them. The generated caption should be meaningful and related to the context of the image. Image captioning techniques are the most researched area. It involves expertise of computer vision (CV), natural language processing (NLP) and artificial intelligence (AI). This paper proposes a novel hybrid approach for higher accuracy of image captioning. A detailed review of traditional methods and methods based on deep learning developed for image captioning are analyzed. The different method has significantly different Bilingual Evaluation Understudy (BLEU) scores on similar images. Thus, the hybrid approach is developed to get high BLEU score for each input images. The data set generation, implementation of hybrid approach, and the challenges along with the future work are discussed.
ISSN:2688-769X
DOI:10.1109/SPIN57001.2023.10116140