Text-Driven NeRF Game Items for Rapid Character Creation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Creating characters and their accoutrements for computer simulations such as computer games is time-consuming.

Innovation Solution

A system that uses a neural radiance field (NeRF) and a Contrastive Language-Image Pre-training (CLIP) model to generate hyper-personalized game items from text input, allowing for rapid creation of virtual characters and accoutrements in less than two minutes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional methods are used to create characters and accoutrements for computer games, then the quality and detail of the characters can be maintained, but the time required for creation is excessive

Engineering Contradiction:

Improvecharacter creation speedVSAvoidtime required for character and accoutrement creation

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent replaces traditional manual mechanical creation processes with automated neural radiance field (NeRF) technology and CLIP models. The system uses AI to generate 3D characters and accoutrements from text prompts, eliminating the need for manual modeling and animation while maintaining high quality output.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the fundamental parameters of character creation by using neural radiance fields instead of traditional polygonal meshes. This allows for continuous, high-resolution 3D representations that can be generated rapidly from text descriptions, fundamentally altering the creation workflow from manual to automated.

Inventive Principle:

Principle #35Parameter changes

2Ease of operation

If manual creation methods are used, then detailed customization is possible, but the process is time-consuming and labor-intensive

Engineering Contradiction:

Improveintuitiveness of character creationVSAvoidtime required for character creation

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The system enables self-service character creation where users simply provide text prompts describing their desired characters and accoutrements. The AI system automatically processes these prompts, generates the 3D models, and outputs ready-to-use assets without requiring users to have technical skills in 3D modeling or animation.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent introduces AI models (CLIP, NeRF) as intermediaries between the user's text ideas and the final 3D character assets. These intermediaries translate natural language descriptions into accurate 3D representations, bridging the gap between simple user input and complex graphical output.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Productivity

If AI-generated content is used, then creation speed increases, but the uniqueness and personalization of items may be compromised

Engineering Contradiction:

Improvegeneration speed of game itemsVSAvoidhyper-personalization capability

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent applies local quality by generating hyper-personalized items tailored to specific players' preferences, play styles, and in-game achievements. Each player receives customized accoutrements and characters that are uniquely adapted to their individual characteristics, ensuring high adaptability while maintaining rapid generation through AI.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS12427424B2Hyper-personalized game items

Publication Date: 2025.09.30 SONY INTERACTIVE ENTERTAINMENT LLC

AI summary

Two dimensional images are converted to a 3D neural radiance field (NeRF), which is modified based on text personalized to a player and input to resemble the accoutrement for a character demanded by the text. A model scores how well an image matches a line of text to produce a final 3D NeRF, which may be converted to a polygonal mesh and imported into a computer simulation such as a computer game.