Neural Super Sampling (NSS)

Neural Super Sampling (NSS) is a parameter prediction model for real-time temporal super sampling developed by Arm, optimized for execution on Neural Accelerators (NX) in mobile GPUs. It enables high-resolution rendering at a lower compute cost by reconstructing high-quality output frames from low-resolution temporal inputs. NSS is particularly suited for mobile gaming, XR, and other power-constrained graphics use cases.View on Hugging Face

Get Started with this Model

Code example

Enchanted Castle Neural Graphics Demo Assets

Use the Enchanted Castle project assets to demonstrate and evaluate Arm neural-graphics features in Unreal Engine.

Code example

Neural Graphics Model Gym Example Notebooks

Follow Jupyter notebooks for NSS and NFRU training, fine-tuning, quantization-aware training, VGF export, evaluation, and custom-model integration.