Enhancing photo/video quality using Deep Learning

Project description

The goal of the project was to increase photo/video quality and scale it up to 4x without losing its quality. Solution: Deep CNN and GAN based image super-resolution trained on provided domain data. Performance: near-real-time and 60fps, so it can be applied to a live stream.

Technologies:

• PyTorch based CNN model for image super-resolution.
• LibTorch and PyTorch JIT for near-real-time performance.