INTERNATIONAL JOURNAL OF NOVEL RESEARCH AND DEVELOPMENT International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2456-4184 | Impact factor: 8.76 | ESTD Year: 2016
Scholarly open access journals, Peer-reviewed, and Refereed Journals, Impact factor 8.76 (Calculate by google scholar and Semantic Scholar | AI-Powered Research Tool) , Multidisciplinary, Monthly, Indexing in all major database & Metadata, Citation Generator, Digital Object Identifier(DOI)
Automatically generating descriptive captions for images is indeed a crucial task for machines to understand and describe natural scenes accurately. Current state-of-the-art models utilize deep convolutional neural networks (CNNs) to extract visual features from the images, followed by recurrent neural networks for the task of image caption generation. However, these models often overlook the textual information present in images, limiting the effectiveness of the generated captions. This study focuses on utilizing the text present in the images along with the visual features extracted from CNN. In this paper, we propose integration of deep convolutional neural networks with long-short term memory to generate context related captions. We achieve this by fusing the textual features obtained from text along with the visual features obtained from the CNN model. We use the benchmark Flickr8k dataset for validating the efficiency of proposed approach. The experiments reveal that fusing of visual-textual features performs better than existing state-of-the-art models.
Keywords:
Image captioning, Convolutional neural networks, Long short term memory, Visual cues, Textual cues
Cite Article:
"A Multi Model CNN-LSTM For Image Caption Generation", International Journal of Novel Research and Development (www.ijnrd.org), ISSN:2456-4184, Vol.9, Issue 3, page no.f411-f418, March-2024, Available :http://www.ijnrd.org/papers/IJNRD2403547.pdf
Downloads:
00047
ISSN:
2456-4184 | IMPACT FACTOR: 8.76 Calculated By Google Scholar| ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 8.76 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Facebook Twitter Instagram LinkedIn