Eikonal Fields for Refractive Novel-View Synthesis

Mojtaba Bemana; Karol Myszkowski; Jeppe Revall Frisvad; Hans-Peter Seidel; Tobias Ritschel

doi:10.1145/3528233.3530706

Eikonal Fields for Refractive Novel-View Synthesis

Mojtaba Bemana
, Karol Myszkowski
, Jeppe Revall Frisvad
, Hans-Peter Seidel
, Tobias Ritschel

Max Planck Institute for Informatics
University College London

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Abstract

We tackle the problem of generating novel-view images from collections of 2D images showing refractive and reflective objects. Current solutions assume opaque or transparent light transport along straight paths following the emission-absorption model. Instead, we optimize for a field of 3D-varying index of refraction (IoR) and trace light through it that bends toward the spatial gradients of said IoR according to the laws of eikonal light transport.

Original language	English
Title of host publication	Proceedings of SIGGRAPH ’22
Number of pages	9
Publisher	Association for Computing Machinery
Publication date	2022
Pages	1-9
ISBN (Print)	978-1-4503-9337-9
DOIs	https://doi.org/10.1145/3528233.3530706
Publication status	Published - 2022
Event	2022 Special Interest Group on Computer Graphics and Interactive Techniques Conference - Vancouver Convention Centre, Vancouver, Canada Duration: 7 Aug 2022 → 11 Aug 2022

Conference

Conference	2022 Special Interest Group on Computer Graphics and Interactive Techniques Conference
Location	Vancouver Convention Centre
Country/Territory	Canada
City	Vancouver
Period	07/08/2022 → 11/08/2022

Keywords

Refraction
Deep learning
Eikonal rendering

Access to Document

10.1145/3528233.3530706

OpenUrl availability

Check availability in DTU Findit

Cite this

@inproceedings{808ede8d058f41e0b64d0c3ff312e918,

title = "Eikonal Fields for Refractive Novel-View Synthesis",

abstract = "We tackle the problem of generating novel-view images from collections of 2D images showing refractive and reflective objects. Current solutions assume opaque or transparent light transport along straight paths following the emission-absorption model. Instead, we optimize for a field of 3D-varying index of refraction (IoR) and trace light through it that bends toward the spatial gradients of said IoR according to the laws of eikonal light transport.",

keywords = "Refraction, Deep learning, Eikonal rendering",

author = "Mojtaba Bemana and Karol Myszkowski and \{Revall Frisvad\}, Jeppe and Hans-Peter Seidel and Tobias Ritschel",

year = "2022",

doi = "10.1145/3528233.3530706",

language = "English",

isbn = "978-1-4503-9337-9",

pages = "1--9",

booktitle = "Proceedings of SIGGRAPH {\textquoteright}22",

publisher = "Association for Computing Machinery",

address = "United States",

note = "2022 Special Interest Group on Computer Graphics and Interactive Techniques Conference, SIGGRAPH {\textquoteright}22 ; Conference date: 07-08-2022 Through 11-08-2022",

}

Bemana, M, Myszkowski, K, Revall Frisvad, J, Seidel, H-P & Ritschel, T 2022, Eikonal Fields for Refractive Novel-View Synthesis. in Proceedings of SIGGRAPH ’22. Association for Computing Machinery, pp. 1-9, 2022 Special Interest Group on Computer Graphics and Interactive Techniques Conference, Vancouver, British Columbia, Canada, 07/08/2022. https://doi.org/10.1145/3528233.3530706

TY - GEN

T1 - Eikonal Fields for Refractive Novel-View Synthesis

AU - Bemana, Mojtaba

AU - Myszkowski, Karol

AU - Revall Frisvad, Jeppe

AU - Seidel, Hans-Peter

AU - Ritschel, Tobias

PY - 2022

Y1 - 2022

N2 - We tackle the problem of generating novel-view images from collections of 2D images showing refractive and reflective objects. Current solutions assume opaque or transparent light transport along straight paths following the emission-absorption model. Instead, we optimize for a field of 3D-varying index of refraction (IoR) and trace light through it that bends toward the spatial gradients of said IoR according to the laws of eikonal light transport.

AB - We tackle the problem of generating novel-view images from collections of 2D images showing refractive and reflective objects. Current solutions assume opaque or transparent light transport along straight paths following the emission-absorption model. Instead, we optimize for a field of 3D-varying index of refraction (IoR) and trace light through it that bends toward the spatial gradients of said IoR according to the laws of eikonal light transport.

KW - Refraction

KW - Deep learning

KW - Eikonal rendering

U2 - 10.1145/3528233.3530706

DO - 10.1145/3528233.3530706

M3 - Article in proceedings

SN - 978-1-4503-9337-9

SP - 1

EP - 9

BT - Proceedings of SIGGRAPH ’22

PB - Association for Computing Machinery

T2 - 2022 Special Interest Group on Computer Graphics and Interactive Techniques Conference

Y2 - 7 August 2022 through 11 August 2022

ER -

Eikonal Fields for Refractive Novel-View Synthesis

Abstract

Conference

Keywords

Access to Document

OpenUrl availability

Fingerprint

Cite this