Presentation

Online Neural Path Guiding with Normalized Anisotropic Spherical Gaussians
DescriptionWe propose a online framework to learn the spatial-varying distribution of the full product of the rendering equation, with a single small neural network using stochastic ray samples, and a novel, expressive, closed-form density model called the Normalized Anisotropic Spherical Gaussian mixture.
Event Type
Technical Paper
TimeWednesday, 31 July 20244:15pm - 4:25pm MDT
LocationMile High 1
Session Time & Location
Sunday, 28 July 20246:00pm - 8:45pm MDTBluebird Ballroom
Wednesday, 31 July 20243:45pm - 5:35pm MDTMile High 1
Interest Areas
Research & Education
Recordings
Livestreamed
Recorded
Keywords
Rendering
Registration Categories
Full Conference
Full Conference Supporter
Virtual Access
Exhibitor Full Conference
Wednesday