Reality Capture Data Classification Using Metadata for Asset Generation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Manually classifying source files of reality capture data into appropriate asset classes is often impossible, impractical, or error-prone, leading to inefficient use of computational and network resources and degraded processing of digital reality capture assets.

Innovation Solution

A method and system for automatically classifying source files of reality capture data into asset classes using metadata, involving a user device, classification server, and classification model to generate payloads and identify the appropriate asset class for digital reality capture assets.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual classification of source files is performed, then classification accuracy may be maintained, but productivity is severely reduced and error-prone

Engineering Contradiction:

Improveclassification accuracyVSAvoidclassification speed

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces the mechanical manual classification process with an automated machine learning classification system. The classification server uses trained models to automatically classify source files into asset classes based on metadata, eliminating the need for manual human intervention while maintaining high accuracy through algorithmic decision-making.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The classification system is self-service in that it automatically processes source files without requiring human operators. The classification server autonomously receives source files, extracts metadata, applies classification models, and generates classification results, making the system self-sufficient for the classification task.

Inventive Principle:

Principle #25Self-service

2Reliability

If manual classification of source files is performed, then detailed review is possible, but loss of time increases significantly

Engineering Contradiction:

Improveclassification reliabilityVSAvoidclassification time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary classification automatically before any potential manual review. By pre-classifying source files using the classification model, the system prepares the data in advance, reducing the time required for subsequent processing while maintaining reliability through the automated classification framework.

Inventive Principle:

Principle #10Preliminary action

3Quantity of substance

If all source files are processed without classification, then processing completeness is maintained, but loss of energy increases due to unnecessary processing

Engineering Contradiction:

Improveprocessing completenessVSAvoidcomputational resource consumption

Core Design Contradiction:

Quantity of substanceVSUse of energy by moving object

Solution Approach 1:

The classification system segments source files into different asset classes based on their metadata and characteristics. This segmentation allows the system to process only the relevant files for each specific asset class, reducing unnecessary computational processing while maintaining complete coverage of all source files through organized categorization.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system applies different processing qualities and methods to different asset classes. By classifying files locally into specific categories, the system can optimize processing parameters for each class, reducing overall energy consumption while maintaining processing completeness across all file types.

Inventive Principle:

Principle #3Local quality

4Reliability

If incorrect classification occurs, then processing errors increase, but device complexity remains low

Engineering Contradiction:

Improveprocessing accuracyVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent replaces error-prone manual classification with an automated machine learning system that consistently applies classification rules. This substitution reduces processing errors by eliminating human variability while managing system complexity through automated algorithmic processes.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The classification system incorporates feedback mechanisms to improve classification accuracy. By analyzing classification results and adjusting the model accordingly, the system reduces processing errors over time while maintaining manageable complexity through iterative improvement.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS12380156B2Methods and systems for automatically classifying reality capture data

Publication Date: 2025.08.05 DRONEDEPLOY INC

AI summary

A method for classifying a set of source files of reality capture data into an asset class of digital reality capture assets may include receiving, from a user device, a set of payloads including metadata of the set of source files of the reality capture data corresponding to a region of interest captured by a camera. The method may include classifying the set of source files of the reality capture data into the asset class of the digital reality capture assets, based on the set of payloads including the metadata of the set of source files of the reality capture data. The method may include providing, to the user device, information identifying the asset class of the digital reality capture assets to which the set of source files are classified to permit a digital reality capture asset, corresponding to the asset class, of the region of interest to be generated based on the set of source files.