Grant: Alliance Grants
Université Laval
$30,000.00
Physics-driven image compositing
FY 2023-24
Summary
Image compositing is the process of piecing together parts of different images in order to create a new image. For example, one might want to select an object from one image, and copy it to another. In this context, two conflicting goals must be met. First, it is important to allow the user a control over the result (e.g., they must be able to specific the position and scale of the object, but perhaps also control the desired lighting). Second, the resulting image must be realistic: the inserted object must blend well with its new surroundings. The main goal of this research proposal is to devise new approaches for compositing images together while allowing for physics-driven control (e.g., change the lighting direction, slightly rotate the object) on the desired result. We propose to use develop the next generation of deep generative models, which will be able to not only generate realistic images, but will do so in a user-controllable, physically-based fashion.
Program
Alliance Grants
Recipient
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
586543-2023
Name nom
Lalonde, Jean-François J-F
Department département
Génie électrique et génie informatique
Organization id
51
Institution établissement
Université Laval
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2023
Competition year année de concours
2023
Award amount
30000
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
801
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Computer software
Area of application fr
Logiciel
Research subject code
9999
Research subject group en
Not available
Research subject group fr
Not available
Research subject en
Non disponible
Research subject fr
Non disponible
Application title
Physics-driven image compositing
Keyword
computer graphics; computer vision; diffusion models; generative models; illumination estimation; image compositing
Application summary
Image compositing is the process of piecing together parts of different images in order to create a new image. For example, one might want to select an object from one image, and copy it to another. In this context, two conflicting goals must be met. First, it is important to allow the user a control over the result (e.g., they must be able to specific the position and scale of the object, but perhaps also control the desired lighting). Second, the resulting image must be realistic: the inserted object must blend well with its new surroundings. The main goal of this research proposal is to devise new approaches for compositing images together while allowing for physics-driven control (e.g., change the lighting direction, slightly rotate the object) on the desired result. We propose to use develop the next generation of deep generative models, which will be able to not only generate realistic images, but will do so in a user-controllable, physically-based fashion.