Depth-Based Facemask Compliance Measurement Without Identity Exposure

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing facemask compliance measurement systems often compromise privacy and identity, suffer from computational inefficiencies, and struggle with systemic inaccuracies, particularly for diverse populations, due to reliance on visible-light images and deep neural networks.

Innovation Solution

Utilizing depth information capture devices to generate depth images, applying irreversible functions to transform depth values into spectral-density values, and employing classification models to assess facemask compliance without revealing individual identities, ensuring robust and efficient mask-wearing analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If visible-light images and deep neural networks are used for facemask compliance measurement, then classification accuracy can be achieved, but privacy and identity are compromised

Engineering Contradiction:
Improvefacemask compliance measurement accuracyVSAvoidprivacy and identity compromise
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent extracts only the necessary geometric features (facial landmarks, mask boundaries, compliance metrics) from depth images while deliberately excluding identifiable visual information. By using depth data instead of visible-light images, the system separates the compliance measurement function from personal identification, achieving accurate measurement without privacy compromise.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces depth images as an intermediary medium between the person and the measurement system. Depth images provide geometric information needed for compliance assessment while inherently protecting identity by not capturing visual appearance. This intermediary layer enables measurement precision while eliminating privacy harm.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If deep neural networks are used for facemask compliance analysis, then classification capability is improved, but computational efficiency and power consumption increase

Engineering Contradiction:
Improvecompliance classification accuracyVSAvoidcomputational power consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent extracts specific geometric features (facial landmarks, contour points, mask boundaries) from depth images and uses these extracted features directly for compliance assessment. By taking out only the necessary geometric information and using simple geometric algorithms rather than deep neural networks, the system achieves accurate classification with minimal computational overhead and low power consumption.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent replaces expensive, computationally intensive deep neural networks with inexpensive, simple geometric algorithms. These lightweight algorithms require minimal processing power and can be executed efficiently on battery-powered devices, achieving the desired classification accuracy without the high energy costs of deep learning models.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

3Adaptability or versatility

If diverse facial features and mask types are accommodated, then adaptability is improved, but measurement accuracy may decrease due to systemic inaccuracies

Engineering Contradiction:
Improveaccommodation of diverse populations and mask typesVSAvoidcompliance measurement accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent develops a universal geometric measurement framework that works across diverse facial features, ethnicities, and mask types. By using fundamental geometric relationships (distances, angles, proportions) that are invariant to appearance variations, the system achieves both high adaptability and maintained measurement accuracy. The geometric approach naturally generalizes to different populations without introducing systemic biases.

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

Solution Approach 2:

The patent measures compliance based on geometric parameters (relative positions of facial landmarks, mask coverage areas, distance ratios) rather than appearance-based features. By changing the measurement parameters from visual appearance to geometric relationships, the system becomes adaptable to diverse faces and mask types while maintaining precision, as geometric relationships remain consistent across different appearances.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4020403B1Systems and methods for privacy-preserving facemask-compliance-level measurement
Publication Date: 2026.03.11 INTEL CORP
  • EP4020403B1 patent drawingFigure 1
  • EP4020403B1 patent drawingFigure 2
  • EP4020403B1 patent drawingFigure 3

AI summary

Disclosed herein are systems and methods for privacy-preserving facemask-compliance-level measurement. In an embodiment, a mask-compliance measurement system includes a processor that is configured to generate, from an image of a person, a facial depth image of a region of a face of the person, and to generate facial wavelet descriptors from the facial depth image. The processor is also configured to determine spectral-density values of the wavelet descriptors, and to analyze the spectral-density values to generate a mask-wearing-compliance result for the person. In an embodiment, the analyzing includes using a classification model that is trained to classify sets of spectral-density values with respect to facemask wearing in images from which the spectral-density values were derived.