Systems and methods for generating images to achieve a style

Patent metadata

Field	Value
Publication number	US-12299774-B2
Application number	US-202218083456-A
Country	US
Kind code	B2
Filing date	Dec 16, 2022
Priority date	Dec 16, 2022
Publication date	May 13, 2025
Grant date	May 13, 2025

Abstract

A method generating an image according to a style is described. The method includes receiving a textual description describing a first image. The method further includes applying an artificial intelligence (AI) model to determine the style of a second image based on the textual description, the first image, a plurality of descriptions, and a plurality of images to generate a suggestion. The style provides a context to the second image. The style is lacking in the first image. The method includes generating the second image with the suggestion according to the style and providing the second image to a client device for display.

First claim

The invention claimed is:

1. A method generating an image according to a style, comprising:
receiving a textual description describing a first image;
applying an artificial intelligence (AI) model to determine the style of a second image based on the textual description, the first image, a plurality of descriptions, and a plurality of images to generate a suggestion, wherein the style provides a context to the second image, wherein the style is lacking in the first image, wherein said applying the AI model includes:
identifying each word of the textual description as being esoteric or exoteric;
assigning a first weight to said each word of the textual description based on whether the word is esoteric or exoteric; and
determining the style based on the first weight;

generating the second image with the suggestion according to the style; and
providing the second image to a client device for display.

2. The method of claim 1 , wherein said applying the AI model includes:
identifying each word of the textual description as belonging to a first lexical category or a second lexical category;
assigning a second weight to said each word of the textual description according to the first lexical category or the second lexical category; and
determining the style based on the second weight.

3. The method of claim 1 , wherein said applying the AI model includes:
identifying two or more words of the textual description as belonging the same one of a plurality of lexical categories;
identifying locations of the two or more words in the textual description;
assigning a plurality of weights to the two or more words based on the locations; and
determining the style based on the plurality of weights.

4. The method of claim 1 , wherein said applying the AI model includes:
identifying a plurality of gaze directions of a user from whom the textual description is received;
assigning a plurality of weights to two or more words of the textual description based on the plurality of gaze directions; and
determining the style based on the plurality of weights.

5. The method of claim 1 , wherein said applying the AI model includes:
identifying amounts of emphasis by one or more users on audio data describing the textual description;
assigning a plurality of weights to two or more words of the textual description based on the amounts of emphasis; and
determining the style based on the plurality of weights.

6. The method of claim 1 , wherein said applying the AI model includes:
identifying amounts of input variables used for inputting the textual description;
assigning a plurality of weights to two or more words of the textual description based on the amounts of input variables; and
determining the style based on the plurality of weights.

7. The method of claim 1 , wherein the suggestion is a textual description.

8. The method of claim 1 , wherein the suggestion includes a reordering of words of the textual description, or a reduction in a number of the words of the textual description, or a combination thereof.

9. The method of claim 1 , further comprising:
providing the suggestion to the client device;
receiving a response to the suggestion, wherein the response indicates an acceptance or a denial of the suggestion.

10. A server system for customizing an image based on user preferences, comprising:
a processor configured to:
receive a textual description describing a first image;
apply an artificial intelligence (AI) model to determine a style of a second image based on the textual description, the first image, a plurality of descriptions, and a plurality of images to generate a suggestion, wherein the style provides a context to the second image, wherein the style is lacking in the first image, wherein to apply the AI model, the processor is configured to:
identify each word of the textual description as being esoteric or exoteric;
assign a first weight to said each word of the textual description based on whether the word is esoteric or exoteric; and
determine the style based on the first weight;

generate the second image with the suggestion according to the style;
provide the second image to a client device for display; and

a memory device coupled to the processor.

11. The server system of claim 10 , wherein to apply the AI model, the processor is configured to:
identify each word of the textual description as belonging to a first lexical category or a second lexical category;
assign a second weight to said each word of the textual description according to the first lexical category or the second lexical category; and
determine the style based on the second weight.

12. The server system of claim 10 , wherein to apply the AI model, the processor is configured to:
identify two or more words of the textual description as belonging the same one of a plurality of lexical categories;
identify locations of the two or more words in the textual description;
assign a plurality of weights to the two or more words based on the locations; and
determine the style based on the plurality of weights.

13. The server system of claim 10 , wherein to apply the AI model, the processor is configured to:
identify a plurality of gaze directions of a user from whom the textual description is received;
assign a plurality of weights to two or more words of the textual description based on the plurality of gaze directions; and
determine the style based on the plurality of weights.

14. The server system of claim 10 , wherein to apply the AI model, the processor is configured to:
identify amounts of emphasis by one or more users on audio data describing the textual description;
assign a plurality of weights to two or more words of the textual description based on the amounts of emphasis; and
determine the style based on the plurality of weights.

15. The server system of claim 10 , wherein to apply the AI model, the processor is configured to:
identify amounts of input variables used for inputting the textual description; and
assign a plurality of weights to two or more words of the textual description based on the amounts of input variables; and
determine the style based on the plurality of weights.

16. A non-transitory computer-readable medium containing program instructions for generating an image according to a style, 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 describing a first image;
applying an artificial intelligence (AI) model to determine the style of a second image based on the textual description, the first image, a plurality of descriptions, and a plurality of images to generate a suggestion, wherein the style provides a context to the second image, wherein the style is lacking in the first image, wherein the operation of applying the AI model includes:
identifying each word of the textual description as being esoteric or exoteric;
assigning a first weight to said each word of the textual description based on whether the word is esoteric or exoteric; and
determining the style based on the first weight;

generating the second image with the suggestion according to the style; and
providing the second image to a client device for display.

17. The non-transitory computer-readable medium of claim 16 , wherein the operation of applying the AI model includes:
identifying each word of the textual description as belonging to a first lexical category or a second lexical category;
assigning

Assignees

• Sony Interactive Entertainment Inc

Inventors

• Green Arran
• Bean Celeste
• Taylor Michael

Classifications

• G10L15/1807

  using prosody or stress · CPC title

• G06F3/013

  Eye tracking input arrangements (G06F3/015 takes precedence) · CPC title

• G06F40/166

  Editing, e.g. inserting or deleting · CPC title

• G06F40/284

  Lexical analysis, e.g. tokenisation or collocates · CPC title

• G06F16/583

  using metadata automatically derived from the content · CPC title