Face Spoofing Detection Using Camera Motion and Biometric Points

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing biometric security systems are vulnerable to face spoofing attacks, such as using counterfeit images or videos, which are difficult to detect without high computing resources and operator intervention.

Innovation Solution

A method using a communicating object with a camera and accelerometer to record a video sequence while moving along a curved path, extracting biometric points and accelerometer data, and applying machine learning algorithms to classify the authenticity of the face, without requiring 3D reconstruction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If high computing resources are used for face spoofing detection, then detection accuracy is improved, but system complexity and resource consumption increase

Engineering Contradiction:
Improveface spoofing detection accuracyVSAvoidcomputational resource requirements
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the essential features needed for spoofing detection: 2D face image data from the camera and acceleration data from the accelerometer. By extracting only these critical data elements rather than processing complete 3D reconstructions or comprehensive biometric datasets, the system achieves effective spoofing detection with reduced computational complexity.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent replaces complex computational mechanics (3D reconstruction algorithms, heavy machine learning models) with a simpler approach combining basic 2D image processing and accelerometer data analysis. This substitution reduces the mechanical complexity of the computational system while maintaining detection effectiveness.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If 3D reconstruction is implemented for spoofing detection, then detection capability is improved, but processing time and computational load increase

Engineering Contradiction:
Improvespoofing detection capabilityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent uses inexpensive 2D camera images and accelerometer readings instead of expensive 3D sensing hardware or complex reconstruction processes. These simple, easily acquired data sources are processed quickly without requiring time-consuming 3D reconstruction, achieving fast spoofing detection with readily available sensors.

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

Solution Approach 2:

The patent applies partial action by using only 2D face images rather than complete 3D facial data. This partial approach (using only necessary image data combined with acceleration patterns) is sufficient for detecting spoofing attempts and eliminates the need for time-consuming full 3D reconstruction while maintaining detection reliability.

Inventive Principle:
Principle #16Partial or excessive action

3Measurement precision

If pre-registration of reference faces is required, then authentication accuracy is improved, but system complexity and setup time increase

Engineering Contradiction:
Improveauthentication accuracyVSAvoidpre-registration requirements
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs self-service by using the accelerometer data from the device itself to detect spoofing attempts. The accelerometer embedded in the communication object automatically provides motion pattern data that enables spoofing detection without requiring external reference databases or pre-registration of multiple facial views, simplifying the system setup.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent makes the system universal by enabling spoofing detection with a single 2D face image and accelerometer data, without requiring pre-registered reference faces or multiple enrollment photos. This multi-functional approach (using standard camera and accelerometer already present in communication objects) eliminates the need for complex pre-registration procedures while maintaining authentication capability.

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

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach allows for fast and resource-efficient detection of face spoofing, confirming the presence of a physical person rather than a mask or projected image, with reduced computational demands and no need for pre-registration.

Implementation Method 1

record (300) the acceleration undergone by the camera

Methodology Applied
Scientific EffectAcceleration: Accelerometer

Data Source

PatentEP3881222B1Method for automatically detecting facial impersonation
Publication Date: 2025.12.24 SURYS
  • EP3881222B1 patent drawingFigure 1~2
  • EP3881222B1 patent drawingFigure 3~4
  • EP3881222B1 patent drawingFigure 5~6

AI summary

The invention relates to a method for automatically detecting the impersonation of a face (30) comprising: - acquiring (200) a sequence of video frames of the face (30) using a camera. It is basically characterised by: - recording (300) the acceleration of the camera during the acquisition; - selecting (210) at least one frame, and applying a facial recognition algorithm (220) and extracting (230) a set of predetermined biometric points (10); determining (310) the value of the acceleration experienced by the camera for each frame; - generating (400) a data file (40) synchronously comprising: mathematical processing of the coordinates of the biometric points and mathematical processing of the values of the acceleration; - applying an automatic classification algorithm by automatic learning (500) to said data file (40); and - transmitting (600) a signal (OK, NOK) representing the category assigned by the automatic classification algorithm by automatic learning to the data file.