Python library for handling most vision tasks .\

In this project, we implement SRGAN. This paper attempts to upscale images up to a factor of 4x without losing the finer textural details. We extended the scope of this idea to videos. We demonstrate that the model generalizes well on out-of-domain inputs through various biased and unbiased inputs.