GAN Based Photo-Realistic Image Generation from Sketch using Nested U-Net

In computer vision generative Image modelling is a vast area of research that many studies have been carried out to address such problems as an image to image translation. In this study, we mainly discuss how we can bridge the gap between the industrial designer and their production workflow to reduce the cost of the time they spend on prototyping. We demonstrate an image synthesizing technique to generate a photo-realistic image of a real-world object from a sketch. We integrate a new network architecture to the existing network to improve the system in generating photo realistic images. Compared to the existing systems our system can generate images with more accuracy and more photo-realism.