Grant: Alliance Grants
University of Waterloo
$40,000.00
High-Resolution Projector-Camera and LED-Wall Display Calibration
FY 2024-25
Summary
High-end projection and visual imagery go far beyond the regular display of a single projector onto a flat white screen, as in a movie theatre or university classroom. In this project, our industry partner is interested in displaying onto curved or irregular surfaces, based on multiple (possibly up to hundreds) projectors, or based on tiled LED walls. The goal of this project is to infer the overall 3D environment, based on some number of cameras observing the scene.
That is, we wish to accurately determine the 3D location of the pixels of a display, and then to apply 3D inference to enable correction to some idealized image, using as input only one or more uncalibrated cameras. 3D inference and reconstruction are well-investigated research areas in the literature, however the contextual assumptions render previous efforts impractical for our application, because the established methods require manual camera calibrations, which are expensive and time consuming, or assume the projection onto a flat planar surface, which is here frequently not the case, or assume the use of calibration objects, which may be infeasible in confined spaces or in projecting onto vast buildings. Thus, this project focuses on fully-automated 3D inference and reconstruction with the fewest prior assumptions. To this end, we propose to design an approach to automatically and accurately learn pixel-level 3D geometry of arbitrarily large and arbitrarily shaped displays of two types, either the projection onto unknown surfaces, or discrete-LED displays of nominally planar or non-planar geometry. The only input to the system is intended to be one or more uncalibrated camera, possibly moving (hand-held). Such calibration will allow for 3D inference, in the case of projection, or to identify and correct for pixel-level non-uniformity in LED displays.
Program
Alliance Grants
Recipient
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
581432-2023
Name nom
Fieguth, Paul PW
Department département
Systems Design Engineering
Organization id
33
Institution établissement
University of Waterloo
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2023
Award amount
40000
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
High-Resolution Projector-Camera and LED-Wall Display Calibration
Keywords
3d geometry; camera calibration; data projection; imaging; projector-camera systems
Application summary
High-end projection and visual imagery go far beyond the regular display of a single projector onto a flat white screen, as in a movie theatre or university classroom. In this project, our industry partner is interested in displaying onto curved or irregular surfaces, based on multiple (possibly up to hundreds) projectors, or based on tiled LED walls. The goal of this project is to infer the overall 3D environment, based on some number of cameras observing the scene.
That is, we wish to accurately determine the 3D location of the pixels of a display, and then to apply 3D inference to enable correction to some idealized image, using as input only one or more uncalibrated cameras. 3D inference and reconstruction are well-investigated research areas in the literature, however the contextual assumptions render previous efforts impractical for our application, because the established methods require manual camera calibrations, which are expensive and time consuming, or assume the projection onto a flat planar surface, which is here frequently not the case, or assume the use of calibration objects, which may be infeasible in confined spaces or in projecting onto vast buildings. Thus, this project focuses on fully-automated 3D inference and reconstruction with the fewest prior assumptions. To this end, we propose to design an approach to automatically and accurately learn pixel-level 3D geometry of arbitrarily large and arbitrarily shaped displays of two types, either the projection onto unknown surfaces, or discrete-LED displays of nominally planar or non-planar geometry. The only input to the system is intended to be one or more uncalibrated camera, possibly moving (hand-held). Such calibration will allow for 3D inference, in the case of projection, or to identify and correct for pixel-level non-uniformity in LED displays.