With the recent spurt in technological advancements, especially the advent of autonomous navigation, image processing has become ubiquitous in everyday life. Image reconstruction is an imaging application aimed at rebuilding images from a variety of features obtained from sparse input ground truths or input sources. In this work, we present an image reconstruction method to recreate and represent detailed images solely from binary edges and flat color using Generative Adversarial Networks (GANs) for a universal application, whereas all major recent breakthroughs are related to either medical imaging or noise removal. Our work has yielded promising results and we were able to obtain realistic images of objects and human faces with original as well as altered feature sets. We present here the results with both qualitative and quantitative comparisons with some benchmark frameworks.