Partial plane sweep volume for deep learning based view synthesis

Kouta Takeuchi, Kazuki Okami, Daisuke Ochi, Hideaki Kimata

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

Abstract

We propose a partial plane sweep volume that can be a more suitable input format for deep-learning-based view synthesis approaches. Our approach makes it possible to synthesize higher quality images with a smaller number of learning iterations, while keeping the number of depth planes.

Original languageEnglish
Title of host publicationACM SIGGRAPH 2017 Posters, SIGGRAPH 2017
PublisherAssociation for Computing Machinery, Inc
ISBN (Electronic)9781450350150
DOIs
Publication statusPublished - 2017 Jul 30
Event44th International Conference on Computer Graphics and Interactive Techniques, ACM SIGGRAPH 2017 - Los Angeles, United States
Duration: 2017 Jul 302017 Aug 3

Publication series

NameACM SIGGRAPH 2017 Posters, SIGGRAPH 2017

Other

Other44th International Conference on Computer Graphics and Interactive Techniques, ACM SIGGRAPH 2017
CountryUnited States
CityLos Angeles
Period17/7/3017/8/3

Keywords

  • Novel view synthesis
  • Plane sweep volume

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Software
  • Computer Graphics and Computer-Aided Design

Fingerprint Dive into the research topics of 'Partial plane sweep volume for deep learning based view synthesis'. Together they form a unique fingerprint.

Cite this