This system takes in 480p (640x480) videos from the Davis dataset and uses a Convolutional Autoencoder to reduce them to 60x80 before upscaling back to 480p.
The end-goal is to use this project to take 480p videos, compress them, transmit the compressed videos and use the decoder to upscale and also interpolate frames of the 480p footage to increase FPS.
I aim to use this project to explore the creation of intelligent compression for videos.
(This project is currently in development)