TermiNeRF: ray termination prediction for efficient neural rendering
File(s)
Accepted version
Author(s)
Piala, Martin
Clark, Ronald
Type
Conference Paper
Abstract
Volume rendering using neural fields has shown great promise in capturing and synthesizing novel views of 3D scenes. However, this type of approach requires querying the volume network at multiple points along each viewing ray in order to render an image, resulting in very slow rendering times. In this paper, we present a method that overcomes this limitation by learning a direct mapping from camera rays to locations along the ray that are most likely to influence the pixel’s final appearance. Using this approach we are able to render, train and fine-tune a volumetrically-rendered neural field model an order of magnitude faster than standard approaches. Unlike existing methods, our approach works with general volumes and can be trained end-to-end.
Date Issued
2022-01-06
Date Acceptance
2022-01-01
Citation
2021 International Conference on 3D Vision (3DV), 2022, pp.1106-1114
Publisher
IEEE
Start Page
1106
End Page
1114
Journal / Book Title
2021 International Conference on 3D Vision (3DV)
Copyright Statement
© 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Identifier
https://ieeexplore.ieee.org/document/9665832
Source
2021 International Conference on 3D Vision (3DV)
Publication Status
Published
Start Date
2021-12-01
Finish Date
2021-12-03
Coverage Spatial
London, United Kingdom
Date Publish Online
2022-01-06