Geometric Scene Feature Detection for Real-Time Gameplay Assistance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Players in computer games face challenges in navigating complex virtual environments due to obstacles and hazards, leading to frustration and dissatisfaction, as existing systems lack real-time assistance for efficient decision-making.

Innovation Solution

A system that uses segment casts to detect geometric features in virtual scenes, maps these features to recommended actions, and controls player avatars in real-time, utilizing parallel processing to enhance gameplay efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If real-time detection and analysis of geometric features is implemented to provide assisted gameplay, then player decision-making efficiency is improved, but computational complexity and processing requirements increase

Engineering Contradiction:
Improvedecision-making efficiencyVSAvoidcomputational complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system segments the virtual environment into discrete geometric features (planes, lines, points, shapes) that can be independently detected and processed. By breaking down complex scenes into fundamental geometric elements, the system achieves real-time analysis without overwhelming computational demands, as each feature type can be processed through specialized detection algorithms.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary detection of geometric features in advance of when they are needed for gameplay decisions. By continuously pre-processing the virtual environment to identify and catalog geometric features beforehand, the system reduces real-time computational burden during critical gameplay moments, enabling faster decision-making support.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If comprehensive geometric feature detection is performed to enhance gameplay assistance, then detection accuracy is improved, but processing time increases

Engineering Contradiction:
Improvedetection accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system dynamically adjusts detection parameters such as feature sensitivity thresholds, detection depth, and geometric complexity levels based on gameplay context. By changing these parameters adaptively, the system maintains high detection accuracy for critical features while reducing processing time for less important elements, achieving a balance between precision and speed.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system applies different detection accuracy levels to different regions of the virtual environment based on their importance to gameplay. High-precision detection is concentrated on critical geometric features directly affecting player decisions, while less critical areas receive reduced processing, thereby maintaining overall detection accuracy while minimizing total processing time.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS12533588B2Techniques for assisted gameplay using geometric features
Publication Date: 2026.01.27 ELECTRONIC ARTS INC
  • US12533588B2 patent drawing
  • US12533588B2 patent drawing
  • US12533588B2 patent drawing

AI summary

The techniques described herein include using a system for enabling assisted gameplay in a computer game using real-time detection of predefined scene features and mapping of the detected features to recommended actions. For example, the system may generate a scanning query (e.g., a segment cast) toward a target area within a virtual scene, determine a geometric feature based on the scanning query, determine a scene feature based on the geometric feature, determine an action associated with the scene feature, and control an avatar based on the action. Examples of scene features that may have mappings to recommended actions include obstacles within a predicted trajectory of the avatar and transitions in the ground level of the virtual scene.