Real-Time Upscaling With Neural Super Sampling in Unreal Engine

Follow this hands-on code-along to integrate Neural Super Sampling (NSS) into an Unreal Engine project and get ready for next-generation neural upscaling techniques, offering stunning image quality at a fraction of the GPU workload cost. Using ML extensions for Vulkan, you configure the environment, enable the upscaling plugin, and run a live demo on Arm emulation layers. This setup demonstrates the effects of neural accelerators coming soon to Arm GPUs, removing the need for dedicated neural accelerator hardware.

What you’ll learn:

• How to set up Vulkan SDK and ML emulation layers for NSS
• How to enable and integrate the NSS plugin into Unreal Engine 5.5
• How to configure Vulkan as Unreal’s rendering backend
• How to visualize ML-powered upscaling in real time using NSS technology

Requirements:

• Be on Windows 11
• Have Visual Studio 2022 with Desktop Development with C++ and the .NET Desktop Development installed
• Have Unreal Engine 5.5 installed
• How to visualize ML-powered upscaling in real time using NSS technology
• In Unreal, have already created a new 5.5 project, using the Third Person template, with C++ as the project default

Watch the on-demand session below, or start building with the Neural Super Sampling in Unreal Engine learning path and follow the same workflow at your own pace.