IMAGE CAPTION GENERATOR USING DEEP LEARNING

IMAGE CAPTION GENERATOR USING DEEP LEARNING

1 SAHANA K, 2 YASHODA P G
[1]   Student, Department of MCA, BIET, Davangere

[2]   Assistant Professor Department of MCA, BIET, Davangere

ABSTRACT
Deep learning's ability to combine computer vision with natural language processing has generated a lot of attention in the field of artificial intelligence research. With an emphasis on using English descriptions to convey the context of photos, this study investigates the use of Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) architectures for automatic image captioning. Using object state, property, and relationship analysis, the suggested models seek to produce high-level semantic descriptions that are meaningful. In particular, a CNN uses the COCO Dataset 2017 to precisely detect objectsfrom grayscale-reduced input image data. The final objective is to create a system that can interpret graphical images into meaningful text messages, improving accessibility for people with visual impairments. Through the use of innovative AI technology, this research attempts to enhance users' cognitive capacities and communication abilities in addition to making it easier to analyze visual information.

Keyword: Convolutional Neural Network, Deep learning, Natural Language, Long Short-Term Memory (LSTM).