Deferred Neural Rendering for View Extrapolation

Tobias Bertel, Yusuke Tomoto, Srinivas Rao, Rodrigo Ortiz-Cayon, Stefan Holzer, Christian Richardt

Research output: Chapter in Book/Report/Conference proceedingConference contribution

17 Downloads (Pure)

Abstract

Image-based rendering methods that support visually pleasing specular surface reflections require accurate surface geometry and a large number of input images. Recent advances in neural scene representations show excellent visual quality while requiring only imperfect mesh proxies or no surface-based proxies at all. While providing state-of-the-art visual quality, the inference time of learned models is usually too slow for interactive applications. While using a casually captured circular video sweep as input, we extend Deferred Neural Rendering to extrapolate smooth viewpoints around specular objects like a car.
Original languageEnglish
Title of host publicationSA '20 Posters: SIGGRAPH Asia 2020 Posters
PublisherAssociation for Computing Machinery
Pages1-2
Number of pages2
DOIs
Publication statusPublished - 31 Dec 2020
EventACM SIGGRAPH Conference and Exhibition on Computer Graphics and Interactive Techniques in Asia - Online
Duration: 4 Dec 202013 Dec 2020
Conference number: 13
https://sa2020.siggraph.org/en/

Conference

ConferenceACM SIGGRAPH Conference and Exhibition on Computer Graphics and Interactive Techniques in Asia
Abbreviated titleSIGGRAPH Asia
Period4/12/2013/12/20
Internet address

Keywords

  • novel-view synthesis
  • surface light field
  • extrapolation

Cite this