Many people across the world don't want to or can't shell out hundreds for a high-resolution camera/smartphone. But we think everyone should be able to have crisp, high-resolution images of their memories regardless of the quality of their physical camera. With Enhance, anyone across the world with access to the internet will be able to upscale their images.
What it does
Enhance is a web application that takes low-resolution images as input and upscales them by 4x with deep learning, making features more distinguishable and increasing image resolution.
How I built it
Enhance uses a novel Generative Adversarial Network architecture trained on two diverse high resolution image datasets: the DIV2K dataset and the Microsoft COCO dataset. We did modeling with Tensorflow and Keras, and to prevent our computers from catching fire, trained it through Google’s Cloud ML Engine. Reprocessing was done through basic libraries like NumPy and Matplotlib.
We then strapped up an easy-to-use, intuitive web application in Flask so anyone can use our model.