US 9,813,621 B2
Omnistereo capture for mobile devices
Robert Anderson, Seattle, WA (US); Steven Maxwell Seitz, Seattle, WA (US); and Carlos Hernandez Esteban, Kirkland, WA (US)
Assigned to Google LLC, Mountain View, CA (US)
Filed by GOOGLE INC., Mountain View, CA (US)
Filed on May 26, 2015, as Appl. No. 14/721,612.
Prior Publication US 2016/0353018 A1, Dec. 1, 2016
Int. Cl. G02B 27/01 (2006.01); G09G 5/14 (2006.01); H04N 5/232 (2006.01); G06T 3/40 (2006.01); H04N 1/00 (2006.01); H04N 13/02 (2006.01); G06T 1/00 (2006.01)
CPC H04N 5/23238 (2013.01) [G02B 27/017 (2013.01); G06T 1/0007 (2013.01); G06T 3/4038 (2013.01); G09G 5/14 (2013.01); H04N 1/00183 (2013.01); H04N 5/23293 (2013.01); H04N 13/0221 (2013.01); G06T 2200/32 (2013.01); G06T 2207/30244 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
obtaining, at a computing device, images captured with a mobile device, the images including content captured from multiple points around a circular path;
adjusting, with the computing device, one or more of the images for rendering a scene in 3D space, the adjusting including:
in response to determining at least one overlap between at least two of the images, the overlap including a match between pixels in the at least two images, selecting, for each overlap, a portion of the pixels included in the respective overlap and combining the portions so that each portion of pixels is vertically aligned to generate a 3D segment of the scene; and
in response to determining that there is not an overlap between the at least two images, generating additional image content to be placed between the at least two images by sampling a plurality of pixels from the at least two images and combining the additional image content to generate a 3D segment of the scene; and
rendering, in a head-mounted display, at least one of the 3D segments as part of the scene.