US 12,073,595 B2
AI encoding apparatus and operation method of the same, and AI decoding apparatus and operation method of the same
Daesung Cho, Suwon-si (KR); Daeeun Kim, Suwon-si (KR); Bongjoe Kim, Suwon-si (KR); Taejun Park, Suwon-si (KR); Sangjo Lee, Suwon-si (KR); and Kyuha Choi, Suwon-si (KR)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Nov. 9, 2021, as Appl. No. 17/522,579.
Application 17/522,579 is a continuation of application No. PCT/KR2021/013814, filed on Oct. 7, 2021.
Claims priority of application No. 10-2020-0171373 (KR), filed on Dec. 9, 2020.
Prior Publication US 2022/0180568 A1, Jun. 9, 2022
Int. Cl. G06N 3/045 (2023.01); G06T 3/4046 (2024.01); G06T 7/00 (2017.01); G06T 9/00 (2006.01)
CPC G06T 9/002 (2013.01) [G06N 3/045 (2023.01); G06T 3/4046 (2013.01); G06T 7/0002 (2013.01)] 14 Claims
OG exemplary drawing
 
1. An artificial intelligence (AI) encoding apparatus comprising:
at least one processor configured to:
determine quality information of an original image by using a quality measurement network,
select a target detail intensity according to the quality information of the original image,
determine a downscaling target, based on a target resolution of a first image,
obtain the first image by AI-downscaling the original image using an AI-downscaling neural network corresponding to the downscaling target,
generate image data by encoding the first image,
select AI-upscaling neural network set identification information, based on the target resolution of the first image, characteristic information of the original image, and the target detail intensity,
generate AI data including the target resolution of the first image, bit depth information of the first image, the AI-upscaling neural network set identification information, and encoding control information, and
generate AI encoding data including the image data and the AI data; and
a communication interface configured to transmit the AI encoding data to an AI decoding apparatus,
wherein the AI data includes information about an AI-upscaling neural network corresponding to the AI-downscaling neural network.