Real-Time Realistic Pixel Synthesis using Deep Learning for Augmented and Virtual Reality

Michael Doggett, LU & Patric Ljung, LiU

Augmented and Virtual Reality will have a major impact on future Human Computing Interfaces. This project aims to solve the challenges of generating the realistic images necessary to ensure the level of immersion to make these new platforms essential. By using high quality physical accurate resources, combined with the latest techniques in Deep Learning combined with Real-Time Rendering, we will address the challenges of immersive realistic imagery for Augmented and Virtual Reality.

Project number: C03