US 11,854,245 B2
Generative adversarial networks having ranking loss
Felix Juefei Xu, Pittsburgh, PA (US); and Marios Savvides, Pittsburgh, PA (US)
Assigned to CARNEGIE MELLON UNIVERSITY, Pittsburgh, PA (US)
Appl. No. 16/976,417
Filed by CARNEGIE MELLON UNIVERSITY, Pittsburgh, PA (US)
PCT Filed Apr. 29, 2019, PCT No. PCT/US2019/029645
§ 371(c)(1), (2) Date Aug. 27, 2020,
PCT Pub. No. WO2019/210303, PCT Pub. Date Oct. 31, 2019.
Claims priority of provisional application 62/762,293, filed on Apr. 27, 2018.
Prior Publication US 2021/0049464 A1, Feb. 18, 2021
Int. Cl. G06V 10/82 (2022.01); G06N 20/20 (2019.01); G06F 17/11 (2006.01); G06N 3/08 (2023.01); G06F 18/2113 (2023.01); G06N 3/045 (2023.01); G06V 10/74 (2022.01); G06V 10/44 (2022.01); G06F 18/22 (2023.01); G06N 3/047 (2023.01); G06N 3/084 (2023.01); G06N 3/088 (2023.01); G06N 3/0455 (2023.01); G06N 3/0475 (2023.01); G06V 40/16 (2022.01)
CPC G06V 10/82 (2022.01) [G06F 17/11 (2013.01); G06F 18/2113 (2023.01); G06F 18/22 (2023.01); G06N 3/045 (2023.01); G06N 3/047 (2023.01); G06N 3/0455 (2023.01); G06N 3/0475 (2023.01); G06N 3/08 (2013.01); G06N 3/084 (2013.01); G06N 3/088 (2013.01); G06N 20/20 (2019.01); G06V 10/454 (2022.01); G06V 10/761 (2022.01); G06V 40/168 (2022.01)] 20 Claims
OG exemplary drawing
 
12. A system for generating images using a generative network comprising:
a processor; and
software that, when executed by the processor, cause the system to implement:
a generative network for generating an image based on an input vector;
a discriminator network for generating a discriminator loss representing the difference between the generated image and a real image represented by the input vector; and
a ranker network for producing a ranking loss between a current iteration and a subsequent iteration of the generative network;
wherein the generator network, the discriminator network and the ranker network are optimized based on the discriminator loss and the ranking loss.