US 12,080,012 B2
Systems and methods for low compute high-resolution depth map generation using low-resolution cameras
Raymond Kirk Price, Carnation, WA (US); Michael Bleyer, Seattle, WA (US); and Christopher Douglas Edmonds, Carnation, WA (US)
Assigned to Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed on Apr. 4, 2023, as Appl. No. 18/295,768.
Application 18/295,768 is a division of application No. 17/235,159, filed on Apr. 20, 2021, granted, now 11,651,506.
Prior Publication US 2023/0274455 A1, Aug. 31, 2023
Int. Cl. G06T 5/00 (2024.01); G02B 27/01 (2006.01); G06T 3/40 (2006.01); G06T 5/20 (2006.01); G06T 5/50 (2006.01); G06T 7/593 (2017.01)
CPC G06T 7/593 (2017.01) [G02B 27/017 (2013.01); G06T 3/40 (2013.01); G06T 5/20 (2013.01); G06T 5/50 (2013.01); G06T 2207/10012 (2013.01); G06T 2207/10048 (2013.01); G06T 2207/20028 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A system for facilitating low compute depth map generation to provide parallax-corrected images, the system comprising:
one or more processors; and
one or more hardware storage devices storing instructions that are executable by the one or more processors to configure the system to facilitate low compute depth map generation to provide parallax-corrected images by configuring the system to:
obtain a first image of an environment;
obtain a second image of the environment, the second image capturing the environment in temporal synchronization with the first image, the second image comprising a higher image resolution than the first image;
obtain a third image of the environment, the third image capturing the environment in temporal synchronization with the second image, the third image having a same image resolution as the second image, wherein parallax exists between capture perspectives associated with the second image and the third image;
generate an upsampled first image, wherein the upsampled first image comprises a same image resolution as the second image;
generate a depth map of the environment by performing stereo matching on the upsampled first image and the second image, wherein the third image is not used to generate the depth map; and
generate a set of parallax-corrected images by reprojecting the second image and the third image to correspond to one or more user perspectives based on depth information from the depth map.