A Visual Question Answering System for Graphical Plots and Figures
Abstract
Visual Question Answering (VQA) systems are one of the fascinating examples of application of Artificial Intelligence. A lot of VQA models are proposed over time which are focused on datasets such as VQA, MS-COCO, Flicker30k etc. which mostly contain randomly taken abstract natural images. But these models dont perform well if the images are graphical plots or say generated from numeric data. Figure-QA dataset is a recently available dataset which contain such statistical images like pie charts, line plots, histograms etc. But the models available are not efficient enough. And there is still lot of work needed to be done to create an efficient enough model which can provide satisfactory results on these statistical images. Taking a step towards that goal, we are proposing a model based on baseline model [3] and embeds attention mechanism to the model and modifying layers accordingly to improve the model. We are committed to develop a Visual Question Answering System for synthetic images representing numerical data that can perform better than the state of art models available.
Keywords : Artificial Intelligence, Machine Learning, Ma-chine Vision, Object detection, Context awareness, Prediction Methods.