Phase-Difference Liveness Detection for 3D Face Verification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing biometric authentication systems, particularly face verification technology, are vulnerable to attacks and lack effective methods to distinguish between live and non-live subjects, which undermines their security and reliability.

Innovation Solution

A liveness detection method and apparatus utilizing a phase difference-based approach, employing a multiphase detection sensor to generate phase images and a minimum map, combined with a neural network model, to analyze disparities and detect liveness by distinguishing between two-dimensional and three-dimensional objects.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional face verification technology is used, then authentication convenience and efficiency are improved, but security against attacks deteriorates

Engineering Contradiction:
Improveauthentication efficiencyVSAvoidsecurity against attacks
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent transitions from traditional 2D face verification to 3D depth-aware verification by introducing phase difference information. The depth map generator creates a depth map representing distance information, and the liveness detector uses this third dimension (depth) to distinguish real faces from spoofing attacks, thereby maintaining authentication efficiency while improving security.

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

Solution Approach 2:

The patent introduces a depth map as an intermediary element between the captured face image and the liveness detection process. The depth map, generated from phase difference calculations, serves as a mediator that provides additional spatial information to the liveness detector, enabling it to differentiate between live and spoofed faces without compromising authentication speed.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of operation

If conventional liveness detection methods are used, then implementation simplicity is maintained, but detection accuracy deteriorates

Engineering Contradiction:
Improveimplementation simplicityVSAvoidliveness detection accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent makes the image sensor multi-functional by enabling it to perform both standard 2D image capture and 3D depth measurement using phase difference detection. The same sensor hardware generates both a captured image and a depth map, eliminating the need for separate detection devices and maintaining implementation simplicity while significantly improving liveness detection accuracy through the added depth dimension.

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

Enhances the security of biometric authentication by effectively differentiating between live and non-live subjects, improving defense against attacks and maintaining system reliability.

Implementation Method 1

generates phase images based on visual information of a plurality of phases sensed by an image sensor

Methodology Applied
Scientific EffectPhase difference detection: Parallax

Data Source

PatentEP3885979B1Method and apparatus for detecting liveness based on phase difference
Publication Date: 2025.11.05 SAMSUNG ELECTRONICS CO LTD
  • EP3885979B1 patent drawingFigure 1
  • EP3885979B1 patent drawingFigure 2~3
  • EP3885979B1 patent drawingFigure 4~5

AI summary

A method and apparatus for detecting a liveness based on a phase difference. The method includes generating a first phase image based on first visual information of a first phase, generating a second phase image based on second visual information of a second phase, generating a minimum map based on a disparity between the first phase image and the second phase image, and detecting a liveness based on the minimum map.