Systems and methods for customizing images based on user preferences

Patent metadata

Field	Value
Publication number	US-2024201833-A1
Application number	US-202218081638-A
Country	US
Kind code	A1
Filing date	Dec 14, 2022
Priority date	Dec 14, 2022
Publication date	Jun 20, 2024
Grant date	—

Abstract

Systems and methods for customizing an image based on user preferences are described. One of the methods includes receiving a textual description with a request to generate an image, accessing a user account to identify a characteristic of a user and a profile of the user, and generating the image by applying an image generation artificial intelligence (IGAI) model to the textual description based on the characteristic of the user and the profile of the user. The IGAI model is trained based on a plurality of images and a plurality of textual descriptions received from a plurality of users. The method further includes conditioning the image to confirm that the image satisfies a plurality of constraints to output a conditioned image and providing the conditioned image for display on a client device via the user account.

First claim

1 . A method for customizing an image based on user preferences, comprising:
receiving a textual description with a request to generate an image; accessing a user account to identify a characteristic of a user and a profile of the user; generating the image by applying an image generation artificial intelligence (IGAI) model to the textual description based on the characteristic of the user and the profile of the user, wherein the IGAI model is trained based on a plurality of images and a plurality of textual descriptions received from a plurality of users; conditioning the image to confirm that the image satisfies a plurality of constraints to output a conditioned image; and providing the conditioned image for display on a client device via the user account.

2 . The method of claim 1 , further comprising training the IGAI model based on the plurality of images and the plurality of textual descriptions received from the plurality of users, wherein said training the IGAI model includes determining a similarity between each of the plurality of textual descriptions and a respective one of the plurality of images.

3 . The method of claim 2 , wherein the textual description is received via the user account, wherein said applying the IGAI model includes:
determining a similarity between the textual description received with the request to generate the image and each of the plurality of textual descriptions to generate image data of the image; applying the profile of the user and a plurality of profiles of the plurality of users to the image; and applying the characteristic of the user to the image.

4 . The method of claim 1 , wherein the profile of the user includes an age of the user and the characteristic includes a geographic location of the user, a plurality of game titles played by the user via the user account, a comment made by the user, and a preference of the user, wherein the profile is stored within the user account.

5 . The method of claim 4 , further comprising receiving the geographic location within a predetermined time period from receiving the textual description.

6 . The method of claim 1 , wherein the plurality of constraints include image data of the plurality of images.

7 . The method of claim 1 , wherein said conditioning the image includes upscaling the image and including missing portions within the image.

8 . A server system for customizing an image based on user preferences, comprising:
a processor configured to:
receive a textual description with a request to generate an image;
access a user account to identify a characteristic of a user and a profile of the user;
generate the image by applying an image generation artificial intelligence (IGAI) model to the textual description based on the characteristic of the user and the profile of the user, wherein the IGAI model is trained based on a plurality of images and a plurality of textual descriptions received from a plurality of users;
condition the image to confirm that the image satisfies a plurality of constraints to output a conditioned image; and
provide the conditioned image for display on a client device via the user account; and
a memory device coupled to the processor.

9 . The server system of claim 8 , wherein to train the IGAI model based on the plurality of images and the plurality of textual descriptions received from the plurality of users, the processor is configured to determine a similarity between each of the plurality of textual descriptions and a respective one of the plurality of images.

10 . The server system of claim 9 , wherein the textual description is received via the user account, wherein to apply the IGAI model, the processor is configured to:
determine a similarity between the textual description received with the request to generate the image and each of the plurality of textual descriptions to generate image data of the image; apply the profile of the user and a plurality of profiles of the plurality of users to the image; and apply the characteristic of the user to the image.

11 . The server system of claim 8 , wherein the profile of the user includes an age of the user and the characteristic includes a geographic location of the user, a plurality of game titles played by the user via the user account, a comment made by the user, and a preference of the user, wherein the profile is stored within the user account.

12 . The server system of claim 11 , wherein the processor is configured to receive the geographic location within a predetermined time period from receiving the textual description.

13 . The server system of claim 8 , wherein the plurality of constraints include image data of the plurality of images.

14 . The server system of claim 8 , wherein to condition the image, the processor is configured to upscale the image and include missing portions within the image.

15 . A non-transitory computer-readable medium containing program instructions for customizing an image based on user preferences, wherein execution of the program instructions by one or more processors of a computer system causes the one or more processors to carry out operations of:
receiving a textual description with a request to generate an image; accessing a user account to identify a characteristic of a user and a profile of the user; generating the image by applying an image generation artificial intelligence (IGAI) model to the textual description based on the characteristic of the user and the profile of the user, wherein the IGAI model is trained based on a plurality of images and a plurality of textual descriptions received from a plurality of users; conditioning the image to confirm that the image satisfies a plurality of constraints to output a conditioned image; and providing the conditioned image for display on a client device via the user account.

16 . The non-transitory computer-readable medium of claim 15 , wherein the operations further comprise training the IGAI model based on the plurality of images and the plurality of textual descriptions received from the plurality of users, wherein said training the IGAI model includes determining a similarity between each of the plurality of textual descriptions and a respective one of the plurality of images.

17 . The non-transitory computer-readable medium of claim 16 , wherein the textual description is received via the user account, wherein the operation of applying the IGAI model includes:
determining a similarity between the textual description received with the request to generate the image and each of the plurality of textual descriptions to generate image data of the image; applying the profile of the user and a plurality of profiles of the plurality of users to the image; and applying the characteristic of the user to the image.

18 . The non-transitory computer-readable medium of claim 15 , wherein the profile of the user includes an age of the user and the characteristic includes a geographic location of the user, a plurality of game titles played by the user via the user account, a comment made by the user, and a preference of the user, wherein the profile is stored within the user account.

19 . The non-transitory computer-readable medium of claim 18 , further comprising receiving the geographic location within a predetermined time period from receiving the textual description.

20 . The non-transitory computer-readable medium of claim 15 , wherein the plurality of constraints include image data of the plurality of images, whe

Assignees

• Sony Interactive Entertainment Inc

Inventors

• Abel Rebecca
• Taylor Michael
• Fortuna Ellana
• Benedetto Warren M

Classifications

• G06T3/40

  Scaling of whole images or parts thereof, e.g. expanding or contracting · CPC title

• G06F18/22

  Matching criteria, e.g. proximity measures · CPC title

• G06T11/00

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

• G06F3/04845 Primary

  for image manipulation, e.g. dragging, rotation, expansion or change of colour · CPC title

• G06F16/5866 Primary

  using information manually generated, e.g. tags, keywords, comments, manually generated location and time information · CPC title