• Prathima Gamini
• Pururva Manikanta Bavana
• Bhargavi Sai Dharmavarapu
• Madhulatha Ayireddy
• Rajesh Gadham

This paper presents an approach to image captioning using a combination of Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks. The ResNET-50 model serves as an encoder to extract meaningful features from images, while the LSTM-based decoder generates coherent and contextually relevant captions. The dataset utilized is the Flickr8k dataset, comprising 8,000 images, each associated with five human-generated captions. This dataset facilitated training the model to understand diverse contexts and generate descriptive captions capturing various aspects of the images. The pre-trained ResNET-50 encoder extracts high-level features from input images, which are then fed into the LSTM-based decoder responsible for generating sequential descriptions. The LSTM network will be trained to grasp temporal dependencies and relationship between the extracted features, thereby ensuring the production of accurate and contextually rich captions.

Deep Learning, CNN, ResNet50, VGG16, InceptionV3, Xception, MobileNet, BLEU Score, Long-Short Term Memory(LSTM).

"Caption Generator of Image to Text using Deep Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 5, page no.e427-e431, May-2024, Available :http://www.jetir.org/papers/JETIR2405448.pdf

"Caption Generator of Image to Text using Deep Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 5, page no. ppe427-e431, May-2024, Available at : http://www.jetir.org/papers/JETIR2405448.pdf