Presentation
Neural BRDF Importance Sampling by Reparameterization
SessionNeural Materials & LOD
DescriptionWe introduce a reparameterization-based formulation of neural BRDF importance sampling. Comparing to previous methods that construct a probability transform to the BRDF through multi-step invertible neural networks, our BRDF sampling is in single step without needing network invertibility, achieving higher inference speed with the best variance reduction.
Event Type
Technical Paper
TimeThursday, 14 August 20252:20pm - 2:30pm PDT
LocationWest Building, Rooms 301-305
ACM Digital Library
Journal Papers' PDFs
Conference Papers' PDFs
Conference Papers' PDFs
Session Time & Location
Sunday, 10 August 20256:00pm - 8:45pm PDTWest Building, Ballroom AB
Thursday, 14 August 20252:00pm - 3:30pm PDTWest Building, Rooms 301-305
Livestreamed
Recorded
