Matthew T. Uyttendaele - Seattle WA, US Jonathan Ryan Howell - Seattle WA, US Jeremy Eric Elson - Kirkland WA, US Drew Edward Steedly - Redmond WA, US Peter Pesti - Atlanta GA, US
Assignee:
MICROSOFT CORPORATION - Redmond WA
International Classification:
G06K 9/36
US Classification:
382284
Abstract:
Systems and methods are provided for the production of seamless, geo-referenced orthographic images that can comprise a composite of two or more underlying images. Illustratively, an exemplary image processing environment comprises an image processing engine and an instruction set comprising at least one instruction to instruct the image processing engine to process data representative of two or more images. Illustratively, the two or more images can comprise data representative of correspondence points between the two or more images and the underlying area (e.g., ground control points). Illustratively, the exemplary image processing engine can identify features that the overlapping photos have in common (e.g., feature match points) and place and re-project (e.g., distort) each of the two or more images to achieve a selected balance of correct position (e.g., based on ground control points) and seamless overlap (e.g., based on feature match points) which can be composited into a single image.
Google - Seattle, WA since Oct 2012
Software Engineer
Georgia Institute of Technology - Atlanta, GA Aug 2006 - Dec 2011
Graduate Research Assitant
Google - Seattle, WA May 2011 - Aug 2011
Software Engineering Intern
Google - Seattle, WA May 2010 - Aug 2010
Software Engineering Intern
Google - Kirkland, WA May 2009 - Aug 2009
Software Engineering Intern
Education:
Georgia Institute of Technology 2006 - 2012
PhD, Computer Science
Georgia Institute of Technology 2004 - 2006
MS, Computer Science
Budapesti Mûszaki és Gazdaságtudományi Egyetem 2000 - 2006
MSc, Technical Informatics