Augmented Reality Virtual Object Data Acquisition via Component Visibility Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for generating three-dimensional appearance data for augmented reality are inefficient, requiring manual creation and resulting in low data generation efficiency.

Innovation Solution

A computer-implemented method and apparatus for acquiring virtual object data in augmented reality by identifying components and directions of a target device from three-dimensional image data, allocating more data resources to visually important portions, and reducing data while maintaining visual quality, using component and direction weights.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If three-dimensional image data is reduced in size for augmented reality applications, then data processing efficiency improves, but visual quality may deteriorate

Engineering Contradiction:
Improvedata generation efficiencyVSAvoidvisual quality
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent applies local quality by differentiating between important and unimportant components of the target object. It identifies visible components that contribute to visual quality and preserves them in high detail, while removing or simplifying hidden components that do not affect visual appearance. This selective preservation strategy maintains visual quality in critical areas while reducing overall data size, thereby resolving the contradiction between data reduction efficiency and visual quality preservation.

Inventive Principle:
Principle #3Local quality

2Manufacturing precision

If manual methods are used to create three-dimensional appearance data, then data accuracy can be maintained, but data generation time increases significantly

Engineering Contradiction:
Improvedata accuracyVSAvoiddata generation time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent employs copying by extracting three-dimensional image data from existing target objects (such as products) and transforming this copied data into augmented reality-compatible formats. Instead of manually creating three-dimensional models from scratch, the system copies geometric and appearance information from real objects, automatically processes this copied data to identify and preserve visible components, and generates the final AR data. This copying approach maintains data accuracy while dramatically reducing generation time compared to manual methods.

Inventive Principle:
Principle #26Copying

3Loss of information

If all components of a target device are preserved in three-dimensional image data, then complete visual information is maintained, but data size increases unnecessarily

Engineering Contradiction:
Improvevisual information completenessVSAvoiddata size
Core Design Contradiction:
Loss of informationVSQuantity of substance

Solution Approach 1:

The patent applies the extraction principle by selectively removing hidden components from the three-dimensional image data that do not contribute to the visual appearance of the target object. The system analyzes the spatial relationships and visibility of various components, then extracts and removes those that are occluded or not visible from external viewpoints. This extraction process reduces data size by eliminating redundant information while preserving all visually relevant components, thereby maintaining visual information completeness without the bloat of unnecessary data.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP3864630B1Method and apparatus for acquiring virtual object data in augmented reality
Publication Date: 2024.03.27 SAMSUNG ELECTRONICS CO LTD
  • EP3864630B1 patent drawingFigure 1~3
  • EP3864630B1 patent drawingFigure 4~5
  • EP3864630B1 patent drawingFigure 6

AI summary

Provided are a method and apparatus for acquiring virtual object data in augmented reality, the method including identifying at least one exposed component from three-dimensional image data corresponding to a target device, the exposed component being exposed to a view from outside of the target device, identifying a type of the target device from the three-dimensional image data corresponding to the target device, determining at least one main direction of the target device based on the type of the target device, and acquiring virtual object data for representing an appearance of the target device based on the at least one exposed component and at least one main direction.