Build real-time neural rendering pipelines with Metal

Key Points

• Shows how to integrate machine learning into the real-time rendering pipeline using Metal 4.
• Covers practical adoption patterns and best practices for achieving production-quality results with MetalFX neural denoising, featuring real-world insights from Maxon Redshift Live.
• Explains how to train and deploy a neural tone mapper using the ML command encoder inline with the graphics work.
• Finally, dive into the new tensor API to build and evaluate small, specialized neural networks directly within the shaders.
• The session moves through MetalFX Denoising, Deploy custom ML networks with Metal 4, Inline neural networks with tensorOps.
• Key concepts include Metal, MetalFX, TensorOps, GPUs, MTLPackage, PyTorch, HDRNet.