Go to Sessions

Running Neural Networks in a Real-time Game Engine


Oct 21
15:30 - 16:20

SPEAKERS

Neural networks have brought significant improvements to many areas of real-time 3D, such as scene construction, gameplay, rendering and so on. However, deploying neural networks across different consumer devices at an interactive framerate is not yet a solved problem.

We will lay out practical considerations for building a real-time neural network inference library in a 3D engine. We will discuss and outline potential solutions relating to model management, optimizations and deployment.
SEE MORE

RELATED CONTENT


ARM ECOSYSTEM SESSIONS

Reducing Carbon Emissions: From SoC Development through Product Deployment and End-of-Life

Joshua Randall

Staff Research Engineer / Arm

Hannah Peeler

Graduate Researcher / Arm


The IC industry is set to consume 20% of the world's energy supply in 2050, driven largely by data centers. While run-time power requirements continue to decrease, other factors increase carbon emissions: IP verification, SoC manufacturing, and shipp...

Go to Session

EVENT PARTNER SESSIONS

Enabling Compute in All its Forms: Arm and Synopsys and the Rise of Democratized HPC

Brian Millar

Principal Engineer / Synopsys


With the "-as-a-service" model already bringing advances and efficiencies to our daily lives, the democratization of AI-enabled high-performance computing (HPC) promises to go beyond addressing the pressing, big-data-centric scientific problems of to...

Go to Session

EVENT PARTNER CAMPFIRES

Arm & Synopsys: Empowering the Next Generation of High-Performance, Power-Efficient Servers

Brian Millar

Principal Engineer / Synopsys


The long-standing partnership between Synopsys and Arm continues to enable mutual customers to realize innovative solutions, accelerating to market secure, AI-enabled, and power-efficient HPC systems that lay at the center of an increasingly insight-...

Go to Session