Grant: Discovery Grants Program - Individual
Ryerson University
$15,000.00
Development of computer vision based localization and mapping for robotics and augmented reality
FY 2010-11
Summary
The ability to self-localize with vision is an important capability that humans and animals possess, there is a need to implement vision in automatic systems such as robotics and Augmented Reality (AR). Mobile AR holds the promise of providing new ways of viewing data and electronic content annotated over the real world,
and has been recognized by many including MIT's Technology Review as one of the top ten emerging technologies. For a mobile robot to navigate or an AR system to operate in an unknown and unprepared environment, the pose (position and orientation) must be constantly measured. Computer vision can be used to build a model of the environment and localize within it, this model can be shared with other robots or AR devices. Computer vision has seen major recent developments that suggest a breakthrough in these areas to produce real world robotics and AR systems may soon be possible.
The goals of this research are to develop such a vision based localization system, specifically in the context of mobile devices. Aspects of feature detection in images, matching between viewpoints, and fusion into an automatically generated map will be addressed. The research will focus on two aspects for mobile devices: processing power and devices, and compact scene map representations. The former will look for theory and algorithms that can run on lower processing power devices such as mobile phones, or can be implemented in custom programmable devices or Graphic Processing Units (GPU). Secondly, research will attempt to find compact representations of the map of vision features for efficient download to other robots or use by mobile users. This compact data will allow location-based AR content viewable on consumer camera cell phones when a user enters an area previously mapped and aligned with virtual content by others.
Program
Discovery Grants Program - Individual
Recipient
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
356072-2009
Name nom
Fiala, Mark
Department département
Computer Science
Organization id
43
Institution établissement
Ryerson University
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2010
Competition year année de concours
2009
Award amount
15000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
331
Committee name en
Computing and Information Sciences - B
Committee name fr
Informatique - B
Area of application code
1207
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Information, computer and communication technologies
Area of application fr
Technologie de l'information, des ordinateurs et des réseaux de communications
Research subject code
2603
Research subject group en
Robotics
Research subject group fr
Robotique
Research subject en
Computer vision
Research subject fr
Vision artificielle
Partie
2
Nb partie
5
Application title
Development of computer vision based localization and mapping for robotics and augmented reality
Keyword
NSERC
Application summary
The ability to self-localize with vision is an important capability that humans and animals possess, there is a need to implement vision in automatic systems such as robotics and Augmented Reality (AR). Mobile AR holds the promise of providing new ways of viewing data and electronic content annotated over the real world,
and has been recognized by many including MIT's Technology Review as one of the top ten emerging technologies. For a mobile robot to navigate or an AR system to operate in an unknown and unprepared environment, the pose (position and orientation) must be constantly measured. Computer vision can be used to build a model of the environment and localize within it, this model can be shared with other robots or AR devices. Computer vision has seen major recent developments that suggest a breakthrough in these areas to produce real world robotics and AR systems may soon be possible.
The goals of this research are to develop such a vision based localization system, specifically in the context of mobile devices. Aspects of feature detection in images, matching between viewpoints, and fusion into an automatically generated map will be addressed. The research will focus on two aspects for mobile devices: processing power and devices, and compact scene map representations. The former will look for theory and algorithms that can run on lower processing power devices such as mobile phones, or can be implemented in custom programmable devices or Graphic Processing Units (GPU). Secondly, research will attempt to find compact representations of the map of vision features for efficient download to other robots or use by mobile users. This compact data will allow location-based AR content viewable on consumer camera cell phones when a user enters an area previously mapped and aligned with virtual content by others.