US 12,340,529 B2
Machine learning-based object-centric approach to image manipulation
Vickie Youmin Wu, Santa Clara, CA (US); and Xiaojia Song, Santa Clara, CA (US)
Assigned to Advanced Micro Devices, Inc., Santa Clara, CA (US)
Filed by Advanced Micro Devices, Inc., Santa Clara, CA (US)
Filed on Dec. 27, 2021, as Appl. No. 17/562,913.
Claims priority of provisional application 63/151,517, filed on Feb. 19, 2021.
Prior Publication US 2022/0270273 A1, Aug. 25, 2022
Int. Cl. G06K 9/00 (2022.01); G06T 5/70 (2024.01); G06T 5/77 (2024.01); G06T 7/50 (2017.01)
CPC G06T 7/50 (2017.01) [G06T 5/70 (2024.01); G06T 5/77 (2024.01); G06T 2207/10016 (2013.01); G06T 2207/20084 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for image manipulation, the method comprising:
determining depth information for an image, the depth information including depth information for an object of interest;
identifying a first depth threshold and a second depth threshold based on the depth information for the object of interest, the first depth threshold being closer to a camera than the object of interest and the second depth threshold being farther from the camera than the object of interest; and
modifying content of the image based on the first depth threshold and the second depth threshold, the modifying including obscuring a portion of the image closer than the first depth threshold and farther than the second depth threshold.