Facial Liveness Detection Using Device Motion Consistency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing liveness detection methods for facial image data are inadequate in distinguishing between live and fraudulent biometric data, particularly in remote authentication transactions, leading to spoofing attacks that are difficult to detect.

Innovation Solution

An electronic device captures movement data during facial image capture, using a pre-trained machine learning algorithm to analyze this data and generate a confidence score, comparing it against a threshold to determine the authenticity of the facial image data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional liveness detection methods are used, then the authentication process is simple, but the detection accuracy is insufficient and spoofing attacks are difficult to detect

Engineering Contradiction:
Improveliveness detection accuracyVSAvoidauthentication system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies dynamics by capturing movement data of the authentication device during the authentication process. Instead of using static images alone, the system analyzes dynamic movement patterns including device orientation changes, translation movements, and acceleration data to detect spoofing attempts, thereby improving detection accuracy while maintaining reasonable system complexity

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent introduces movement data as an intermediary element between the facial image and the liveness detection decision. This intermediary provides additional information about the physical handling of the device during authentication, enabling more accurate detection of fraudulent attempts without requiring complex additional hardware

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If movement data analysis is added to enhance liveness detection, then spoofing detection accuracy improves, but the processing complexity increases

Engineering Contradiction:
Improveauthentication reliabilityVSAvoiddata processing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by pre-processing and analyzing movement data during the authentication process itself. The system captures and evaluates device movement patterns in real-time before making the final authentication decision, allowing for reliable spoofing detection without requiring complex post-processing operations

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes the parameters being analyzed by incorporating movement data parameters (device orientation, position, acceleration) alongside traditional facial recognition parameters. This multi-parameter approach enhances authentication reliability by providing multiple indicators to detect spoofing attempts

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If multiple data types are captured and analyzed, then authentication accuracy improves, but the time required for authentication increases

Engineering Contradiction:
Improveauthentication precisionVSAvoidauthentication time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies continuity of useful action by capturing movement data continuously during the authentication process rather than as separate discrete steps. The movement data collection occurs naturally throughout the authentication sequence, allowing for accurate analysis without adding significant time overhead to the overall process

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentEP3989184B1Liveness detection of facial image data
Publication Date: 2026.02.18 DAON TECH
  • EP3989184B1 patent drawingFigure 1
  • EP3989184B1 patent drawingFigure 2~4
  • EP3989184B1 patent drawingFigure 5

AI summary

A method for enhanced liveness detection of facial image data is provided that includes capturing movement data of an electronic device while capturing, using the electronic device, facial image data of a user. In response to determining the captured movement data is consistent with movement data expected to be captured during capture of facial image data, the method includes deciding the captured facial image data is genuine. In response to determining the captured movement data is different than movement data expected to be generated during capture of facial image data, the method includes deciding the captured facial image data is fraudulent.