US 11,842,274 B2
Electronic apparatus and controlling method thereof
Gilwoo Song, Suwon-si (KR); Jaehyun Kwon, Suwon-si (KR); Jinwoo Nam, Suwon-si (KR); Heeseung Shin, Suwon-si (KR); and Minseok Choi, Suwon-si (KR)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Jan. 4, 2021, as Appl. No. 17/140,706.
Claims priority of application No. 10-2020-0002257 (KR), filed on Jan. 7, 2020.
Prior Publication US 2021/0209374 A1, Jul. 8, 2021
Int. Cl. G06N 3/00 (2023.01); G06N 3/08 (2023.01); G06V 20/40 (2022.01); G06F 18/22 (2023.01); G06N 3/045 (2023.01); G06V 10/22 (2022.01); G06V 10/25 (2022.01); G06V 10/75 (2022.01)
CPC G06N 3/08 (2013.01) [G06F 18/22 (2023.01); G06N 3/045 (2023.01); G06V 10/22 (2022.01); G06V 20/41 (2022.01); G06V 20/46 (2022.01); G06V 10/25 (2022.01); G06V 10/751 (2022.01)] 12 Claims
OG exemplary drawing
 
1. A controlling method of an electronic apparatus, the controlling method comprising:
obtaining image data and metadata regarding the image data, the image data comprising a first image frame and a second image frame that is subsequent to the first image frame;
identifying a manipulated area of the first image frame by inputting the first image frame to a first neural network model trained to detect whether an input image or video has been manipulated;
obtaining a similarity between the first image frame and the second image frame based on motion vector information included in the metadata; and
detecting whether there is a manipulated area in the second image frame based on the similarity between the first image frame and the second image frame, the detecting comprising:
based on the similarity being greater than a predetermined threshold value, setting a region of interest of the second image frame based on the manipulated area of the first image frame and the motion vector information,
obtaining first information on a manipulation probability of the region of interest of the second image frame by inputting at least the region of interest of the second image frame to the trained first neural network model,
based on a size of a motion vector in the motion vector information being greater than a predetermined value, obtaining second information on a manipulation probability of the second image frame by inputting at least the region of interest of the second image frame to a second neural network model, and determining whether there is a manipulated area in the second image frame by determining whether the region of interest of the second image frame has been manipulated based on the first information and the second information, and
based on the size of the motion vector in the motion vector information being less than the predetermined value, obtaining third information on a manipulation probability of the region of interest of the second image frame by inputting at least the region of interest of the second image frame to a trained third neural network model, and determining whether there is a manipulated area in the second image frame by determining whether the region of interest of the second image frame has been manipulated based on the first information and the third information.