Image Caption Generation has always been a study of great interest to the researchers in the Artificial Intelligence (AI) department. Being able to program a machine to accurately describe image or an environment like an average human has major application. The neural model which is regenerative is created. It depends on computer vision (CV) and machine translation. This model is used to generate natural sentences which describes the image. This model consists of Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN). This model is enhanced with LSTM in Bidirectional. The CNN is used for extracting features from image and RNN is used for sentence generation. The model is trained in such a way that if any input image is given to model it generates captions which almost describes the image. The accuracy of model and smoothness of language model learns from image descriptions are tested on different datasets. These experiments show that model is often giving accurate descriptions for an input image. In this paper, we present image caption generating model based on deep neural networks, focusing on the various RNN techniques and analysing their influence on the sentence generation. Captions for sample images and compared the different feature extraction and encoder models to analyse which model gives better accuracy and generates the desired results.



Software And Hardware