	US 12,468,935 B2
	Method and apparatus with image correspondence
Seungju Han, Seoul (KR); Minsu Cho, Pohang-si (KR); Juhong Min, Pohang-si (KR); Jongmin Lee, Pohang-si (KR); and Changbeom Park, Seoul (KR)
Assigned to Samsung Electronics Co., Ltd., Suwon-si (KR); and POSTECH Research and Business Development Foundation, Pohang-si (KR)
Filed by Samsung Electronics Co., Ltd., Suwon-si (KR); and POSTECH Research and Business Development Foundation, Pohang-si (KR)
Filed on Jun. 9, 2021, as Appl. No. 17/342,858.
Claims priority of application No. 10-2020-0090560 (KR), filed on Jul. 21, 2020; and application No. 10-2020-0161385 (KR), filed on Nov. 26, 2020.
Prior Publication US 2022/0027728 A1, Jan. 27, 2022
Int. Cl. G06N 3/08 (2023.01); G06F 18/211 (2023.01); G06F 18/25 (2023.01); G06N 3/045 (2023.01); G06N 3/0464 (2023.01)

CPC G06N 3/08 (2013.01) [G06F 18/211 (2023.01); G06F 18/25 (2023.01); G06N 3/045 (2023.01); G06N 3/0464 (2023.01)]

26 Claims

1. A method with image correspondence, comprising:

acquiring a plurality of feature map pairs corresponding to outputs of a plurality of layers of a convolutional neural network (CNN) in response to an input of an input image pair;

selecting a portion of feature map pairs from among the plurality of feature map pairs by determining, for each of the plurality of feature map pairs, whether to select the feature map pair based on a feature of the feature map pair;

generating a hyper feature map pair based on the selected portion of feature map pairs; and

generating a correspondence result of the input image pair based on a correlation of the hyper feature map pair.