Sparse Neural Graphics Using Cross Vendor GPU Optimization

PI: Michael Doggett, Lund University
co-PI: Jonas Unger, Linköping University

Real-time rendering technology underlies all interactive 3D visual experiences on all displays across all modern computing devices and output modalities. Numerous industrially and societally important applications range from visualization of complex data, medical imaging and interaction with AI systems to digital product design and modern interactive entertainment. In this proposal, we aim to address the key challenge in all these applications, which is to reduce the compute requirements for real-time rendering by creating new algorithms and integrate them into systems capable of creating 3D images using Artificial Neural Networks (ANN) combined with recent research and new insights.

Project number: G4