	US 12,307,490 B2
	Image/text-based design creating device and method
Woo Sang Song, Seoul (KR); Jianri Li, Seoul (KR); and Ki-Young Shin, Seoul (KR)
Filed by Woo Sang Song, Seoul (KR); Jianri Li, Seoul (KR); and Ki-Young Shin, Seoul (KR)
Filed on Mar. 1, 2024, as Appl. No. 18/593,491.
Application 18/593,491 is a continuation of application No. 16/991,311, filed on Aug. 12, 2020, granted, now 11,948,177.
Application 16/991,311 is a continuation of application No. PCT/KR2019/001645, filed on Feb. 11, 2019.
Claims priority of application No. 10-2018-0017229 (KR), filed on Feb. 12, 2018.
Prior Publication US 2024/0202786 A1, Jun. 20, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 30/27 (2020.01); G06F 18/21 (2023.01); G06F 40/205 (2020.01); G06N 3/045 (2023.01); G06Q 30/06 (2023.01)

CPC G06Q 30/06 (2013.01) [G06F 18/21 (2023.01); G06F 30/27 (2020.01); G06F 40/205 (2020.01); G06N 3/045 (2023.01)]

5 Claims

1. A design generation device comprising:

a parsing unit configured to parse image data and text data in real time online;

a learning unit configured to train a design generation model using artificial intelligence based on the parsed image data and text data;

a design generation unit configured to generate a new design or a design element of a specific item by using the design generation model trained in the learning unit based on the real time parsed image data and text data, without separate input from a user,

wherein the learning unit and the design generation unit are configured to perform the training of the design generation model, and the generation of the design or the design element using at least one of generative adversarial network (GAN), variational autoencoder (VAE)+GAN, genetic algorithm (GA)+GAN, conditional GAN, and style transfer, and

wherein the design generation unit further comprises a design changing unit configured to generate the new design or the design element of the specific item by reflecting a received vector value as an input;

a processor; and

a memory.