1 / 31

Multi-Resolution Screen-Space Ambient Occlusion

Multi-Resolution Screen-Space Ambient Occlusion. Thai-Duong Hoang Kok -Lim Low. National University of Singapore. Ambient Occlusion (AO). AO. Cosine-weighted fraction of a tangent hemisphere that is occluded. :. Screen-Space AO (SSAO). Volumetric AO [SKUT10]. SSAO (Dis)Advantages.

tyrone
Download Presentation

Multi-Resolution Screen-Space Ambient Occlusion

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Multi-Resolution Screen-Space Ambient Occlusion Thai-Duong Hoang Kok-Lim Low National University of Singapore

  2. Ambient Occlusion (AO)

  3. AO Cosine-weightedfraction of a tangent hemisphere that is occluded :

  4. Screen-Space AO (SSAO) Volumetric AO [SKUT10]

  5. SSAO(Dis)Advantages • Inaccurate • Local AO • Over/underestimated AO • Low quality • Noise • Blur • Simple • Fast • General • Easy to integrate

  6. Multi-Resolution AO (MSSAO)Intuition

  7. Multi-Resolution AO (MSSAO)Intuition 256x256 512x512 1024x1024

  8. MSSAOOverview Geometry g-buffer (1024x1024) Downsample AO buffer (256x256) Compute AO (256x256) g-buffer (256x256) Blur & Upsample AO buffer (512x512) Compute AO (512x512) g-buffer (512x512) Blur & Upsample AO buffer (1024x1024) Compute AO (1024x1024) g-buffer (1024x1024)

  9. MSSAOAO from Multiple Resolutions

  10. MSSAOOverview Geometry g-buffer (1024x1024) Downsample AO buffer (256x256) Compute AO (256x256) g-buffer (256x256) Blur & Upsample AO buffer (512x512) Compute AO (512x512) g-buffer (512x512) Blur & Upsample AO buffer (1024x1024) Compute AO (1024x1024) g-buffer (1024x1024)

  11. MSSAODownsampling if else

  12. MSSAOOverview Geometry g-buffer (1024x1024) Downsample AO buffer (256x256) Compute AO (256x256) g-buffer (256x256) Blur & Upsample AO buffer (512x512) Compute AO (512x512) g-buffer (512x512) Blur & Upsample AO buffer (1024x1024) Compute AO (1024x1024) g-buffer (1024x1024)

  13. MSSAONeighborhood Sampling • Project the AO radius of influence to screen space at each pixel at resolution to get (in terms of pixels) • Cap to some value (typical value is 5) 1024x1024 16-point Poisson disk 512x512 256x256 Works well with a 3x3 Gaussian filter

  14. MSSAOComputing AO model after the Monte-Carlo approximation of • Low-variance • Cheap • Biased

  15. MSSAOOverview Geometry g-buffer (1024x1024) Downsample AO buffer (256x256) Compute AO (256x256) g-buffer (256x256) Blur & Upsample AO buffer (512x512) Compute AO (512x512) g-buffer (512x512) Blur & Upsample AO buffer (1024x1024) Compute AO (1024x1024) g-buffer (1024x1024)

  16. MSSAOBilateral Upsampling • Bilinear weights • Depth weights • Normal weights

  17. MSSAOCombining AO Values • and are computed by “propagating” appropriate values across resolutions • Avoid underestimating AO by ensuring • And a plausible heuristic

  18. MSSAOTemporal Filtering

  19. ResultsQuality Blizzard [FM08] MSSAO

  20. ResultsQuality HBAO [BSD08] MSSAO

  21. ResultsQuality MSSAO VAO [SKUT10]

  22. ResultsGround-truth Comparison MSSAO Blender

  23. ResultsNoise/Blur MSSAO HBAO [BSD08] Blizzard [FM08]

  24. ResultsMultiple AO Scales Small AO radius Large AO radius MSSAO

  25. ResultsPerformance • Scenes rendered at 1024x1024 on GeForce GTX 460M • Exclusive of geometry pass • The same parameters used to produce the shown images

  26. MSSAOConclusions • Inaccurate • Local AO • Over/underestimated AO • Low quality • Noise • Blur • Use more memory • Poor temporal coherence on very thin geometry • Not too noticeable • Errors due to the use of coarse resolutions • Not too noticeable unless compared with ground-truths • Simple • Fast • General • Easy to integrate • Capture multiple shadow frequencies

  27. ResultsAdditional Results

  28. ResultsAdditional Results

  29. ResultsAdditional Results

  30. ResultsAdditional Results

  31. Thank You

More Related