Paper Title
Image Reconstruction using Lighter Version of Conditional Generative Adversarial Networks

Abstract
A new model to fill the tampered portion of an image using lighter version of Conditional Generative Adversarial Network (GANs) has been proposed in this paper. Lighter mean a system which uses lesser number of parameters and simple network structure. Also, a User Interface (UI) for the object removal and reconstruction part has been made in this work, where a mouse controls is used to mark the object for removal. It gives better results when compared to other similar works in the literature and it also facilitates lesser number of model arguments. Keywords - GANs, Image Reconstruction, Deep Learning, Open CV