US 12,314,862 B2
Optimizing generative networks via latent space regularizations
Sheng Zhong, Santa Clara, CA (US)
Assigned to Agora Lab, Inc., Santa Clara, CA (US)
Filed by Agora Lab, Inc., Santa Clara, CA (US)
Filed on Apr. 26, 2024, as Appl. No. 18/647,545.
Application 18/647,545 is a continuation of application No. 18/319,109, filed on May 17, 2023, granted, now 12,001,956.
Application 18/319,109 is a continuation of application No. 17/324,831, filed on May 19, 2021, granted, now 11,694,085, issued on Jul. 4, 2023.
Application 17/324,831 is a continuation of application No. 16/530,692, filed on Aug. 2, 2019, granted, now 11,048,980, issued on Jun. 29, 2021.
Claims priority of provisional application 62/840,635, filed on Apr. 30, 2019.
Prior Publication US 2024/0296332 A1, Sep. 5, 2024
Int. Cl. G06N 3/084 (2023.01); G06F 18/21 (2023.01); G06N 20/00 (2019.01); G06T 3/4053 (2024.01); G06T 5/00 (2006.01); G06V 10/776 (2022.01); G06V 10/82 (2022.01)
CPC G06N 3/084 (2013.01) [G06F 18/217 (2023.01); G06N 20/00 (2019.01); G06T 3/4053 (2013.01); G06T 5/00 (2013.01); G06V 10/776 (2022.01); G06V 10/82 (2022.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for image generation based on a Generative AI Network, the Generative AI Network comprising a generator and an encoder, the method comprising:
determining, by the encoder, a first encoding E(Y) of a target image Y;
generating, by the generator, a generated image G(Z) corresponding to the target image Y, wherein the generated image G(Z) is located in a close vicinity of a target neighborhood of the target image Y, and outputs of the generator are mapped, by the encoder, to a latent space adaptable to manipulate at least one characteristics of images generated by the Generative AI Network; and
generating, by the encoder, a second encoding E(G(Z)) of the generated image G(Z) corresponding to the target image Y, wherein the first and second encodings E(Y) and E(G(Z)) map the target image Y and the generated image G(Z) to the latent space.