What-if scenario based and generative adversarial network generated design adaptation

Patent metadata

Field	Value
Publication number	US-2023334720-A1
Application number	US-202217659024-A
Country	US
Kind code	A1
Filing date	Apr 13, 2022
Priority date	Apr 13, 2022
Publication date	Oct 19, 2023
Grant date	—

Abstract

A computer-implemented method, a computer program product, and a computer system for design adaptation. One or more computing devices receive images provided by a user, receive form a designer one or more designer provided what-if scenarios of a design related to the images, receive form the user one or more user selected what-if scenarios from the one or more designer provided what-if scenarios, and retrieve information about influencing factors form an artificial intelligence (AI) knowledge corpus, where the influencing factors affect qualities of the images. One or more computing devices input the retrieved information into a generative adversarial network (GAN) module. One or more computing devices change the images by the GAN module to generate adapted images for the one or more user selected what-if scenarios, based on the retrieved information.

First claim

What is claimed is:

1 . A computer-implemented method for what-if scenario based and generative adversarial network (GAN) generated design adaptation, the method comprising:
receiving images provided by a user; receiving, form a designer, one or more designer provided what-if scenarios of a design related to the images; receiving, form the user, one or more user selected what-if scenarios from the one or more designer provided what-if scenarios; retrieving, form an artificial intelligence (AI) knowledge corpus, information about influencing factors for the one or more user selected what-if scenarios, wherein the influencing factors affect qualities of the images; inputting the information about the influencing factors into a GAN module; and changing the images by the GAN module to generate adapted images for the one or more user selected what-if scenarios, based on the information about the influencing factors.

2 . The computer-implemented method of claim 1 , further comprising:
providing the adapted images to the user; receiving from user feedback on the one or more user selected what-if scenarios with the adapted images; and determining whether the user is satisfied with the one or more user selected what-if scenarios.

3 . The computer-implemented method of claim 2 , further comprising:
in response to determining that the user is satisfied with the one or more user selected what-if scenarios, updating the knowledge corpus with user positive feedback.

4 . The computer-implemented method of claim 2 , further comprising:
in response to determining that the user is not satisfied with the one or more user selected what-if scenarios, updating the knowledge corpus with user negative feedback.

5 . The computer-implemented method of claim 2 , further comprising:
in response to determining that the user is not satisfied with the one or more user selected what-if scenarios, receiving, form the user, one or more user newly selected what-if scenarios from one or more designer newly provided what-if scenarios; and iterating steps of the what-if scenario based and GAN generated design adaptation, until the user is satisfied with the one or more user newly selected what-if scenarios.

6 . The computer-implemented method of claim 2 , further comprising:
in response to determining that the user is not satisfied with the one or more user selected what-if scenarios, receiving, form the user, one or more user re-selected what-if scenarios from the one or more designer provided what-if scenarios; and iterating steps of the what-if scenario based and GAN generated design adaptation, until the user is satisfied with the one or more user re-selected what-if scenarios.

7 . A computer program product for what-if scenario based and generative adversarial network (GAN) generated design adaptation, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by one or more processors, the program instructions executable to:
receive images provided by a user; receive, form a designer, one or more designer provided what-if scenarios of a design related to the images; receive, form the user, one or more user selected what-if scenarios from the one or more designer provided what-if scenarios; retrieve, form an artificial intelligence (AI) knowledge corpus, information about influencing factors for the one or more user selected what-if scenarios, wherein the influencing factors affect qualities of the images; input the information about the influencing factors into a GAN module; and change the images by the GAN module to generate adapted images for the one or more user selected what-if scenarios, based on the information about the influencing factors.

8 . The computer program product of claim 7 , further comprising the program instructions executable to:
provide the adapted images to the user; receive from user feedback on the one or more user selected what-if scenarios with the adapted images; and determine whether the user is satisfied with the one or more user selected what-if scenarios.

9 . The computer program product of claim 8 , further comprising the program instructions executable to:
in response to determining that the user is satisfied with the one or more user selected what-if scenarios, update the knowledge corpus with user positive feedback.

10 . The computer program product of claim 8 , further comprising the program instructions executable to:
in response to determining that the user is not satisfied with the one or more user selected what-if scenarios, update the knowledge corpus with user negative feedback.

11 . The computer program product of claim 8 , further comprising the program instructions executable to:
in response to determining that the user is not satisfied with the one or more user selected what-if scenarios, receive, form the user, one or more user newly selected what-if scenarios from one or more designer newly provided what-if scenarios; and iterate steps of the what-if scenario based and GAN generated design adaptation, until the user is satisfied with the one or more user newly selected what-if scenarios.

12 . The computer program product of claim 8 , further comprising the program instructions executable to:
in response to determining that the user is not satisfied with the one or more user selected what-if scenarios, receive, form the user, one or more user re-selected what-if scenarios from the one or more designer provided what-if scenarios; and iterate steps of the what-if scenario based and GAN generated design adaptation, until the user is satisfied with the one or more user re-selected what-if scenarios.

13 . A computer system for what-if scenario based and generative adversarial network (GAN) generated design adaptation, the computer system comprising one or more processors, one or more computer readable tangible storage devices, and program instructions stored on at least one of the one or more computer readable tangible storage devices for execution by at least one of the one or more processors, the program instructions executable to:
receive images provided by a user; receive, form a designer, one or more designer provided what-if scenarios of a design related to the images; receive, form the user, one or more user selected what-if scenarios from the one or more designer provided what-if scenarios; retrieve, form an artificial intelligence (AI) knowledge corpus, information about influencing factors for the one or more user selected what-if scenarios, wherein the influencing factors affect qualities of the images; input the information about the influencing factors into a GAN module; and change the images by the GAN module to generate adapted images for the one or more user selected what-if scenarios, based on the information about the influencing factors.

14 . The computer system of claim 13 , further comprising the program instructions executable to:
provide the adapted images to the user; receive from user feedback on the one or more user selected what-if scenarios with the adapted images; and determine whether the user is satisfied with the one or more user selected what-if scenarios.

15 . The computer system of claim 14 , further comprising the program instructions executable to:
in response to determining that the user is satisfied with the one or more user selected what-if scenarios, update the knowledge corpus with user positive feedback.

16 . The computer system of claim 14 , further c

Assignees

• IBM

Inventors

• Mene Atul
• Fox Jeremy R
• Agrawal Tushar
• Rakshit Sarbajit K

Classifications

• G06T11/00 Primary

  Two-dimensional [2D] image generation · CPC title

• G06N3/0454

  Physics · mapped topic

• G06N3/0427

  Physics · mapped topic

• G06T2200/24

  involving graphical user interfaces [GUIs] · CPC title

• G06N3/042

  Knowledge-based neural networks; Logical representations of neural networks · CPC title