US 12,229,686 B2
Detecting digital human face image manipulations
Lakshmanan Nataraj, Chennai (IN); Tajuddin Manhar Mohammed, Goleta, CA (US); Tejaswi Nanjundaswamy, San Jose, CA (US); Michael Gene Goebel, Santa Barbara, CA (US); Bangalore S. Manjunath, Santa Barbara, CA (US); and Shivkumar Chandrasekaran, Santa Barbara, CA (US)
Assigned to Mayachitra, Inc., Santa Barbara, CA (US)
Filed by Mayachitra, Inc., Santa Barbara, CA (US)
Filed on Oct. 26, 2023, as Appl. No. 18/384,144.
Application 18/384,144 is a continuation of application No. 17/834,455, filed on Jun. 7, 2022, granted, now 11,836,634.
Application 17/834,455 is a continuation of application No. 16/801,866, filed on Feb. 26, 2020, granted, now 11,429,812.
Claims priority of provisional application 62/956,999, filed on Jan. 3, 2020.
Prior Publication US 2024/0070467 A1, Feb. 29, 2024
Int. Cl. G06N 3/088 (2023.01); G06F 18/21 (2023.01); G06F 18/214 (2023.01); G06N 3/045 (2023.01); G06V 10/44 (2022.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01)
CPC G06N 3/088 (2013.01) [G06F 18/2148 (2023.01); G06F 18/217 (2023.01); G06N 3/045 (2023.01); G06V 10/454 (2022.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01)] 18 Claims
OG exemplary drawing
 
1. A method, comprising:
accessing, using one or more processors, an image;
training, using the one or more processors, a convolutional neural network using a set of authentic human face images and a set of digitally manipulated human face images;
generating co-occurrence matrices on pixel values of the image, the generating comprises;
applying a position operator to pixels in the image to determine a defined offsets; and
defining a co-occurrence matrix over the image representing distributions of co-occurring pixel values at the defined offsets;
analyzing the image based on the generated co-occurrence matrices using the trained convolutional neural network, the trained convolutional network trained to detect images generated by a generative adversarial network (GAN); and
classifying the image based on the analysis.