Grant: Alliance Grants
McGill University
$60,000.00
Implicit modelling and rendering of automobile interiors
FY 2024-25
Summary
The design of modern automobile interiors is a very time-consuming process, involving building detailed 3D graphics models and physical prototyping of many different styles and layouts. Our primary goal in this project is to give car interior designers a streamlined creative process which enables rapid and seamless exploration of various display options, including layouts, colours, and information configurations, and to allow the designers to judge the quality and visibility of displayed content.
To do this, we propose applying Neural Radiance Field (NeRF) technologies. This is a novel-view synthesis method that uses an implicit neural network scene representation to rapidly learn the geometry and lighting of a 3D scene from a set of hand-captured photographs or videos. By applying NeRFs to car interiors, we can create realistic and dynamic views from any angle, enhancing the crafting of driver experience by car designers. NeRFs allow for self-supervised training using easily acquired multi-view images, eliminating the need for 3D supervision, and bypasses the time-consuming construction of scenes needed for standard 3D graphics modeling approaches. Applying NeRFs to automotive interior design, including displays, enables rapid re-rendering of interior views, improving responsiveness and adaptability, but with much less labor from graphic designers.
Our research will enhance the existing NeRF technology to provide implementations that perform efficient dynamic relighting of rendered automotive interiors, using accurate environment lighting maps consistent with different driving situations.
Program
Alliance Grants
Recipient
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
601823-2024
Name nom
Clark, James J
Department département
Electrical and Computer Engineering
Organization id
61
Institution établissement
McGill University
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
60000
Program id
ALLRP
Program name en
Alliance Grants
Program name fr
Subventions Alliance
Group en
RESEARCH PARTNERSHIPS
Group fr
PARTENARIATS DE RECHERCHE
Committee code
96
Committee name en
RPP Internal Decision Cttee
Committee name fr
Cté de décision interne - PPR
Area of application code
802
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Information systems and technology
Area of application fr
Systèmes d'information et technologie de l'information
Research subject code
2708
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Image processing
Research subject fr
Traitement des images
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Implicit modelling and rendering of automobile interiors
Keywords
3d rendering; automotive interior design; computer graphics; modeling of 3d environments; neural networks; scene re-lighting
Application summary
The design of modern automobile interiors is a very time-consuming process, involving building detailed 3D graphics models and physical prototyping of many different styles and layouts. Our primary goal in this project is to give car interior designers a streamlined creative process which enables rapid and seamless exploration of various display options, including layouts, colours, and information configurations, and to allow the designers to judge the quality and visibility of displayed content.
To do this, we propose applying Neural Radiance Field (NeRF) technologies. This is a novel-view synthesis method that uses an implicit neural network scene representation to rapidly learn the geometry and lighting of a 3D scene from a set of hand-captured photographs or videos. By applying NeRFs to car interiors, we can create realistic and dynamic views from any angle, enhancing the crafting of driver experience by car designers. NeRFs allow for self-supervised training using easily acquired multi-view images, eliminating the need for 3D supervision, and bypasses the time-consuming construction of scenes needed for standard 3D graphics modeling approaches. Applying NeRFs to automotive interior design, including displays, enables rapid re-rendering of interior views, improving responsiveness and adaptability, but with much less labor from graphic designers.
Our research will enhance the existing NeRF technology to provide implementations that perform efficient dynamic relighting of rendered automotive interiors, using accurate environment lighting maps consistent with different driving situations.