TGEN - Text Generation from Scenic Images
Our project aims to generate a system which analyzes scenic images and generate a caption based on the analysis of the image. For analyzing the image we are using CNN(Convolutional Neural Network) model VGG16 and for sequential modelling we are using LSTM(Long Short Time Model). We have created our own dataset project in kaggle to use for model building and training. METEOR is used for the evaluation of the caption generator. Due to the limited amount of data present in the data we haven't gotten an output that satisfying our expectation. Currently we are attempting to correct and make our generated caption more acurate and precise.