1 / 6

Conclusions

This paper discusses properties of effective filters for image processing, including the use of auxiliary information, per-pixel features, and complex filter shapes. It also explores techniques such as joint bilateral and NL-means filtering, weighted local regression, and the use of per-pixel filter parameters. The paper discusses industry adoption and open challenges in real-time applications, animation sequences, and beyond image space filtering. It provides resources including a survey paper and source code.

huffaker
Download Presentation

Conclusions

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. Conclusions Matthias Zwicker University of Bern

  2. Properties of effective filters • Exploit auxiliary information from renderer • Per-pixel features (normal, position, albedo, etc.) • Support complex filter shapes • Joint bilateral, NL-means filter • Weighted local regression • Use per-pixel filter parameters • Use input variance • Predict using mutual information, learning • Estimate error of filter output (SURE, bias and variance)

  3. Industry adoption • Pixar RenderMan • Disney Hyperion • innoBright www.innobright.com

  4. Open challenges • Real-time applications • Animation sequences • Beyond image space filtering • Exploit additional path space properties • Leverage theoretical foundations • Sampling theory • Learning-based techniques • Sparse methods • Theoretical analysis, proof of lower bounds on sampling density

  5. Resources • Survey paper ”Recent advances in adaptive sampling and reconstruction for Monte Carlo rendering”, Computer Graphics Forum, 2015 • Source code, partial list • Rousselleet al. [RKZ11]:http://www.cgg.unibe.ch/downloads/asr-auxiliary.zip/at_download/file • Rousselleet al. [RKZ12]: http://www.cgg.unibe.ch/downloads/nlm-code-data.zip/at_download/file • Sen and Darabi [SD12]: http://dx.doi.org/10. 7919/F4MW2F28 • Kalantari and Sen [KS13]: http://www.ece.ucsb.edu/~psen/PaperPages/RemovingMCNoiseStuff/RemovingMCNoise_ v1.0.zip • Moon et al. [MJL∗ 13]: http://sglab.kaist.ac.kr/VFL/ • Rousselleet al. [RMZ13]: http://www.cgg.unibe.ch/downloads/pg2013_code_data.zip/at_download/file • Moon et al. [MCY14]: http://sglab.kaist.ac.kr/WLR/

  6. Thank you! • Funding agencies and partners • NSF, Intel, Nvidia, SNSF, NRF • Co-authors Course organizers and presenters: Nima Kalantari UC Santa Barbara Fabrice Rousselle Disney Research Matthias Zwicker University of Bern Pradeep Sen UC Santa Barbara Sung-Eui Yoon KAIST

More Related