US 12,106,524 B2
System and method for learning tone curves for local image enhancement
Abdelrahman Abdelhamed, Toronto (CA); Luxi Zhao, Toronto (CA); and Michael Scott Brown, Toronto (CA)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Dec. 30, 2021, as Appl. No. 17/566,448.
Claims priority of provisional application 63/270,817, filed on Oct. 22, 2021.
Prior Publication US 2023/0127327 A1, Apr. 27, 2023
Int. Cl. G06T 7/90 (2017.01); G06T 5/20 (2006.01); G06T 5/94 (2024.01); G06T 7/11 (2017.01); G06T 11/00 (2006.01); G06V 10/56 (2022.01)
CPC G06T 7/90 (2017.01) [G06T 5/20 (2013.01); G06T 5/94 (2024.01); G06T 7/11 (2017.01); G06T 11/001 (2013.01); G06V 10/56 (2022.01); G06T 2207/10024 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A method, comprising:
receiving an input image;
extracting at least one feature from the input image;
partitioning at least a portion of the input image into a plurality of neighboring tiles,
partitioning the plurality of neighboring tiles into a center region comprising portions of a first number of the plurality of neighboring tiles, and a non-center region comprising at least a portion of a second number of the plurality of neighboring tiles, the second number being less than the first number;
determining a plurality of local tone curves respectively for each of the plurality of neighboring tiles based on the extracted at least one feature; and
generating a toned image based on at least one local tone curve of the plurality of local tone curves,
wherein generating the toned image based on the at least one local tone curve comprises:
based on a pixel being in the non-center region, applying each local tone curve for each tile of the second number of the plurality of tiles to the pixel.