Wearable Fit Prediction Using 3D Depth Mapping from Image Data

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing systems for fitting wearable devices, such as smart glasses, lack accuracy in sizing without specialized equipment or technician assistance, leading to potential improper fits that compromise functionality.

Innovation Solution

A method using a computing device to capture image data, detect facial landmarks, and combine this with position/orientation data from sensors to generate a three-dimensional mesh model, predicting the fit of a wearable device without the need for specialized equipment or a technician.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If existing virtual try-on systems are used, then accessibility is improved, but measurement precision deteriorates

Engineering Contradiction:
ImproveaccessibilityVSAvoidsizing accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent transitions from two-dimensional image analysis to three-dimensional depth mapping by incorporating device position and orientation data. Multiple 2D images captured from different angles are transformed into a 3D depth map, enabling accurate measurement of facial features while maintaining the accessibility of virtual try-on systems.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent introduces position/orientation sensors as an intermediary between the camera and the image processing system. These sensors provide metadata about device orientation that mediates the transformation of 2D images into accurate 3D depth information, resolving the contradiction between simplicity and precision.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If specialized equipment is used, then measurement precision is improved, but device complexity worsens

Engineering Contradiction:
Improvesizing accuracyVSAvoidequipment requirements
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent makes the computing device universal by using its existing camera and position sensors for multiple purposes. The same device that captures images also provides orientation data, eliminating the need for specialized equipment while maintaining measurement precision through multi-functional use of standard components.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The computing device serves itself by using its own position and orientation sensors to provide the metadata needed for depth mapping. No external specialized equipment is required as the device leverages its inherent capabilities to achieve accurate 3D measurements.

Inventive Principle:
Principle #25Self-service

3Measurement precision

If multiple images from different positions are captured, then measurement precision is improved, but loss of time worsens

Engineering Contradiction:
Improvedepth accuracyVSAvoiddata collection time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary action by capturing multiple images in rapid succession before the user moves significantly. The position and orientation data is collected concurrently with image capture, preparing all necessary data in advance for efficient 3D reconstruction without requiring lengthy measurement sessions.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS20240242441A1Fit prediction based on detection of metric features in image data
Publication Date: 2024.07.18 GOOGLE LLC
  • US20240242441A1 patent drawing
  • US20240242441A1 patent drawing
  • US20240242441A1 patent drawing

AI summary

A system and method of predicting fit of a wearable device from image data obtained by a computing device together with position and orientation of the computing device is provided. The system and method may include capturing a series of frames of image data, and detecting one or more fixed features in the series of frames of image data. Position and orientation data associated with the capture of the image data is combined with the position data related to the one or more fixed features, to extract depth data from the series of frames of image data. A three-dimensional model is generated based on the extracted depth data. The three-dimensional model and/or key points extracted therefrom, can be processed by a simulator and/or a machine learning model to predict fit of the wearable device for the user.