US 11,704,896 B2
Method, apparatus, device and storage medium for image processing
Meng Yao, Beijing (CN)
Assigned to Beijing Baidu Netcom Science and Technology Co., Ltd., Beijing (CN)
Filed by Beijing Baidu Netcom Science and Technology Co., Ltd., Beijing (CN)
Filed on Mar. 26, 2021, as Appl. No. 17/213,764.
Claims priority of application No. 202010623701.6 (CN), filed on Jun. 30, 2020.
Prior Publication US 2021/0216811 A1, Jul. 15, 2021
Int. Cl. G06V 10/74 (2022.01); G06V 20/40 (2022.01); G06V 10/762 (2022.01); G06F 18/22 (2023.01)
CPC G06V 10/761 (2022.01) [G06F 18/22 (2023.01); G06V 10/762 (2022.01); G06V 20/48 (2022.01); G06V 20/49 (2022.01)] 13 Claims
OG exemplary drawing
 
1. A method for image processing, comprising:
obtaining a set of image sequences, the set of image sequences comprising a plurality of image sequences;
determining a first similarity between each pair of image sequences in the set of image sequences;
dividing the set of image sequences into one or more subset of image sequences based on first similarities, wherein, dividing the set of image sequences into the one or more subset of image sequences based on the first similarities comprises:
determining a base image sequence from the set of image sequences, wherein a second similarity corresponding to the base image sequence is higher than or equal to a second similarity corresponding to each of other image sequences in the set of image sequences, wherein the second similarity corresponding to the base image sequence is determined based on first similarities between the base image sequence and each of the other image sequences, and the second similarity corresponding to each of the other image sequences is determined based on first similarities between the each image sequence and image sequences other than the each image sequence in the set of image sequences; and
dividing the set of image sequences into a plurality of subsets of image sequence, wherein each of the subsets of the image sequences comprises the base image sequence and at least one other image sequence, wherein the first similarity between the base image sequence and the at least one other image sequence is higher than a threshold similarity; and
determining, in each of the subset of image sequences, degrees of correlation of images in one image sequence of the subset of image sequences with images in other image sequence of the subset of image sequences.