Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore a 10-minute IEEE conference talk presenting a lightweight supersampling framework designed to accelerate stereo rendering in virtual reality. The presentation, part of the "Computer graphics techniques" session, details how researchers from the State Key Lab of CAD&CG integrate image projection with spatio-temporal supersampling to improve VR rendering efficiency. Learn how the framework leverages temporal and spatial redundancies in stereo videos to quickly generate high-resolution frames by blending rendered low-resolution frames with temporal samples, then reprojecting these to synthesize images for opposite viewpoints. The talk explains techniques for addressing disocclusions through history data and low-pass filtering, resulting in high-quality, resolution-enhanced, and anti-aliased images with minimal delay. Discover how extensive evaluations have confirmed the framework's efficiency on both PC and standalone VR devices, making it applicable across various VR platforms.
