AI-based semantic-based primitive generation system, information processing method, and program

Abstract

This system provides an information processing system that can dynamically assign meaning to anything in any field, such as games or business processes, using AI, in response to user prompts, without relying on meaning-fixed objects. [Solution] The system comprises a primitive engine that holds attribute-less primitive objects, a semantic assignment module that assigns semantic information to the primitive objects using AI, and a rendering module that generates a virtual space based on the semantic information.

Description

[Technical Field]

[0001] The present invention relates to an information processing technology using generative AI, and more particularly to an information processing system that dynamically generates games, UIs, business systems and the like in a virtual space by adding semantic information to attribute-free primitive objects. [Background Art]

[0002] Conventional dynamic generation systems or no-code UI generation systems are configured on the premise of objects having fixed meanings.

[0003] However, with this configuration, since generatable experiences are limited to a predefined range, it has been difficult to implement dynamic semantic conversion and concept generation by generative AI. [Prior Art Documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent No. 7545191 (P7545191) [Summary of the Invention] [Problem to be Solved by the Invention]

[0005] An object of the present invention is to provide an information processing system that allows AI to dynamically add meanings to the same object without depending on objects with fixed meanings, and can generate various virtual experiences. [Means for Solving the Problem]

[0006] The information processing system according to the present invention comprises: a primitive engine that holds attribute-free primitive objects such as rectangles, circles, lines, and particles; a meaning addition module that adds semantic information to the primitive objects by AI; and a rendering module that generates a virtual space based on the semantic information. [Effect of the Invention]

[0007] This invention eliminates the need for semantically fixed objects and enables AI to assign different meanings to the same primitive. [Brief explanation of the drawing]

[0008] [Figure 1] System Configuration Diagram of the Present Invention [Figure 2] Processing flow diagram of the present invention [Figure 3] Conceptual diagram for assigning semantic information to the present invention [Figure 4] Dynamic change diagram of the present invention [Modes for carrying out the invention]

[0009] The primitive engine manages attribute-less primitive objects such as rectangles, circles, lines, particles, and layers.

[0010] The semantics module parses user prompts and assigns semantic information to each primitive object.

[0011] The rendering module renders the virtual space based on semantic information. [Examples]

[0012] For example, in response to user prompts, semantic information such as "field" or "UI panel" is assigned to rectangular objects, and semantic information such as "ball" or "energy body" is assigned to circular objects, and these are then represented in the virtual space. [Industrial applicability]

[0013] This invention makes it possible to use virtual space for a wide range of fields, including game generation, UI generation, education, simulation, and business applications, depending on the semantic information. [Explanation of Symbols]

[0014] 11…Primitive engine 12…Meaning assignment module 13…Semantic vector management module 14…Rendering module 15…Verification module 16…Application management module S101…User prompt acquisition S102…Attributeless primitive acquisition S103…Semantic information generation (AI) S104…Semantic vectorization and registration S105…Verification processing S106…Application target determination S107…Semantic information application S108…Rendering and output

Claims

1. An information processing system comprising: a primitive engine that holds attributeless primitive objects consisting of rectangles, circles, lines, particles, or layers, which do not retain semantic attributes; a semantic assignment module that assigns semantic information to the attributeless primitive objects based on user prompts; and a rendering module that renders the attributeless primitive objects based on the semantic information to generate a virtual space, wherein the semantic information is applied as configuration information and does not generate or execute executable code.

2. An information processing system according to claim 1, wherein the primitive engine manages attribute-free primitive objects consisting of rectangles, circles, lines, particles, and layers.

3. An information processing system according to claim 2, wherein the semantic assignment module assigns semantic information to the attributeless primitive object based on a user prompt, and generates inter-object relationships, event rules, physical parameters, and visual representations.

4. An information processing system according to claim 3, characterized in that the semantic information comprises a semantic vector management module in which the semantic information is managed as an embedding representation or a vector space representation.

5. An information processing system according to claim 4, wherein the information processing system comprises a verification module that performs validation processing including at least one of schema verification, access control, moderation, or runtime restriction.

6. An information processing system according to claim 5, characterized in that it comprises an application control module for rendering the semantic information.

7. An information processing system according to claim 6, wherein the rendering module dynamically changes the UI configuration, rules, workflow, or virtual space layout based on the semantic information.

8. An information processing system according to claim 7, characterized in that the semantic information is applied as configuration information without generating or executing executable code.

Images