Real-Time 3D Model Generation with Anomaly Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing 3D model generation technologies do not enable real-time or near real-time generation of 3D models from captured image sequences, leading to potential inaccuracies or incompleteness, with users often unaware of insufficiencies until after processing is complete.

Innovation Solution

A system and method that allows for real-time or near real-time generation of 3D point clouds and models by capturing image sequences, with features to detect anomalies, notify users, and provide options for correction, such as capturing additional images or using algorithms to fill gaps, while also monitoring and adjusting the speed of image capture to ensure data sufficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If image sequences are captured and processed concurrently to generate 3D models in real-time or near real-time, then the generation speed and user feedback capability are improved, but the computational complexity and processing requirements increase

Engineering Contradiction:
Improve3D model generation speedVSAvoidprocessing system complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The patent segments the 3D model generation process into distinct functional modules: image capture module, real-time processing module, anomaly detection module, and model generation module. This segmentation allows concurrent processing of multiple tasks while managing computational complexity through modular architecture, directly resolving the contradiction between generation speed and system complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary actions by detecting anomalies and determining data sufficiency during the image capture phase itself, before complete processing is finished. This early detection and feedback mechanism enables real-time intervention and correction, improving generation speed while maintaining manageable processing requirements through proactive rather than reactive processing.

Inventive Principle:
Principle #10Preliminary action

2Loss of time

If the image capture device moves quickly to capture more frames, then the capture time is reduced, but the accuracy and completeness of the 3D model deteriorate

Engineering Contradiction:
Improveimage capture timeVSAvoid3D model accuracy
Core Design Contradiction:
Loss of timeVSManufacturing precision

Solution Approach 1:

The patent implements continuous feedback mechanisms where the processing system monitors image quality metrics and anomaly detection results in real-time, then provides feedback to control the movement speed of the image capture device. This closed-loop feedback system dynamically adjusts capture speed to maintain optimal accuracy while minimizing capture time, resolving the contradiction between speed and precision.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system dynamically adapts the capture process by adjusting movement speed based on real-time analysis of captured frames and anomaly detection results. This dynamic control allows the system to optimize the trade-off between capture time and model accuracy for each specific scanning scenario, rather than using a fixed speed approach.

Inventive Principle:
Principle #15Dynamics

3Manufacturing precision

If complete image sequences are captured before processing to ensure model accuracy, then the model quality is improved, but the processing time and user feedback delay increase

Engineering Contradiction:
Improve3D model completenessVSAvoidprocessing time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The system performs preliminary processing and anomaly detection on captured image sequences before complete processing is finished. By identifying potential quality issues early in the pipeline and providing user feedback during rather than after capture, the system enables corrective actions while maintaining high model completeness standards, thus resolving the time-quality trade-off.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent maintains continuous useful action by processing images concurrently with capture rather than waiting for complete sequences. This continuous processing approach ensures model completeness through real-time anomaly detection and feedback while minimizing processing delays, allowing the system to maintain high quality standards without sacrificing time efficiency.

Inventive Principle:
Principle #20Continuity of useful action

4Manufacturing precision

If anomaly detection and correction options are provided in real-time, then the model accuracy is improved, but the system complexity and user interaction requirements increase

Engineering Contradiction:
Improve3D model accuracyVSAvoidsystem complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent implements feedback mechanisms that detect anomalies during processing and present correction options to users in real-time. This feedback loop enables accuracy improvement through user-informed corrections while managing system complexity through automated detection and structured presentation of correction options, rather than requiring complex manual intervention systems.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3164848B1System and method of three-dimensional model generation
Publication Date: 2022.07.06 QUALCOMM INC
  • EP3164848B1 patent drawingFigure 1
  • EP3164848B1 patent drawingFigure 2
  • EP3164848B1 patent drawingFigure 3

AI summary

A method includes generating, at an electronic device, a three-dimensional model of an object based on a sequence of images captured by an image capture device associated with the electronic device. The method further includes displaying the three-dimensional model via a display device associated with the electronic device. The method also includes, based on detecting that the three-dimensional model includes an anomaly, presenting, via the display device, one or more selectable options to enable correction of the anomaly.