US 12,333,675 B2
Image processing apparatus, image processing method, and storage medium
Atsushi Takahama, Kanagawa (JP); and Toru Kokura, Kanagawa (JP)
Assigned to Canon Kabushiki Kaisha, Tokyo (JP)
Filed by CANON KABUSHIKI KAISHA, Tokyo (JP)
Filed on Mar. 14, 2022, as Appl. No. 17/694,483.
Claims priority of application No. 2021-053212 (JP), filed on Mar. 26, 2021.
Prior Publication US 2022/0309612 A1, Sep. 29, 2022
Int. Cl. G06T 3/4046 (2024.01); G06N 3/08 (2023.01); G06T 3/4015 (2024.01); G06T 3/4038 (2024.01); G06T 7/00 (2017.01)
CPC G06T 3/4046 (2013.01) [G06N 3/08 (2013.01); G06T 3/4015 (2013.01); G06T 3/4038 (2013.01); G06T 7/0002 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30168 (2013.01)] 15 Claims
OG exemplary drawing
 
1. An image processing apparatus comprising:
an acquisition unit configured to acquire training image data;
a first extraction unit configured to learn the training image data;
a determination unit configured to determine, based on the learning of the training image data, the difficult training image data and extract a first difficult image data that is difficult to learn;
a generation unit configured to generate student image data by mosaicking the first difficult image data;
a training unit configured to generate a network parameter obtained by re-learning a pair of the first difficult image data and the student image data; and
a second extracting unit configured to extract second difficult image data different from the first difficult image data based on an error evaluation value between a ground-truth image in the training image data and a demosaicked image generated by using the network parameter,
wherein the training unit repeats the re-learning by generating a plurality of times the second difficult image data.