US 12,169,912 B2
Surface spectral reflection estimation in computer vision
Hamidreza Vaezi Joze, Redmond, WA (US)
Assigned to Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed on Jun. 21, 2021, as Appl. No. 17/352,469.
Prior Publication US 2022/0405521 A1, Dec. 22, 2022
Int. Cl. G06T 5/60 (2024.01); G06T 5/50 (2006.01); G06T 5/90 (2024.01); G06T 7/174 (2017.01); G06T 7/90 (2017.01)
CPC G06T 5/60 (2024.01) [G06T 5/50 (2013.01); G06T 5/90 (2024.01); G06T 7/174 (2017.01); G06T 7/90 (2017.01); G06T 2207/20081 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A method for processing images, the method comprising:
receiving first image data representing an image, the first image data comprising a plurality of color values corresponding to a plurality of pixels in the image;
determining, using a machine learning model trained for a type of a camera sensor having a camera spectral sensitivity function, second image data based on the first image data, the second image data including a spectral power distribution of light in the image and surface spectral reflection values corresponding to the plurality of pixels in the image, wherein the spectral power distribution of light and the surface spectral reflection values are estimated, using the machine learning model, as a function of wavelength of visible light in the image captured by the camera sensor and the camera spectral sensitivity function; and
performing, using the second image data, at least one image processing operation with respect to the image.