Enhancing Image Captioning Accuracy through Hybrid Deep Learning Models

This paper introduces an Artificial Neural Network
model that integrates advanced deep learning techniques from
computer vision and natural language processing domains. The
model focuses on automating the captioning process for images, a crucial task in artificial intelligence. By employing Convolutional Neural Networks (CNNs) and Long Short-Term Memory Net-works (LSTMs), the model is trained to optimize the likelihood estimation of descriptive labels corresponding to each image in the dataset. Evaluation of the model’s performance includes both quantitative metrics and qualitative assessment using the state-of-the-art BLEU-1 scoring method.