Facial Authentication With TOF Depth Mapping for Spoof Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing facial recognition systems struggle to effectively detect identity theft attempts using faces created or reproduced on surfaces, such as photographs or video streams, often leading to high false alarm rates and processing inefficiencies.
Innovation Solution
A method combining visible range imaging with depth mapping using a Time Of Flight (TOF) camera to verify the authenticity of a face by analyzing geometric and depth characteristics, including inter-eye distance and depth differences between the face and background, coupled with facial recognition algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If facial recognition systems use only visible range images, then the system is simple and fast, but it cannot effectively detect spoofing attempts using photographs or video streams
Solution Approach 1:
The patent transitions from 2D visible range images to 3D depth information by introducing a depth map acquisition device. This additional dimensional data allows the system to detect spatial characteristics of real faces versus spoofing artifacts, significantly improving spoofing detection capability while maintaining system simplicity through efficient depth image processing algorithms
2Reliability
If the system uses depth mapping and geometric analysis, then spoofing detection accuracy improves, but processing time increases
Solution Approach 1:
The system performs preliminary face detection and geometric shape definition before detailed depth analysis. By pre-identifying facial regions and key geometric features in the depth map, the system prepares data structures that accelerate subsequent authentication processing, thereby reducing overall processing time while maintaining high spoofing detection rates
Solution Approach 2:
The patent extracts only the essential geometric features (face outline, inter-eye distance, depth characteristics) from the depth map rather than processing the entire depth image. This selective extraction of critical geometric parameters significantly reduces computational load and processing time while preserving authentication accuracy
3Measurement precision
If the system analyzes multiple geometric features and depth characteristics, then authentication accuracy improves, but false alarm rate increases
Solution Approach 1:
The patent combines multiple geometric features (face outline shape, inter-eye distance, depth map characteristics) into a unified authentication decision process. By merging these complementary features and evaluating them collectively against spoofing patterns, the system achieves high measurement precision while reducing false alarms through cross-validation of multiple indicators
Solution Approach 2:
The system uses the depth map information to provide feedback on the validity of detected facial features. The depth characteristics serve as a verification layer that confirms whether geometric features belong to a real face, creating a feedback mechanism that reduces false alarms by validating feature authenticity before authentication decisions
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
Achieves a spoofing detection rate of over 90% with a false alarm rate below 5% and processing times comparable to standard facial recognition, without requiring complex movements or additional biometric features.
Implementation Method 1
coupling the information from a camera acquiring images in the visible domain with that of a camera configured to obtain a depth map of the scene
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
Facial recognition method comprising, in combination, at least the following steps: - having an image of given quality containing at least one face recognised with respect to identification data and defining a geometric area around said face in the image acquired in the visible range and projecting the geometric area into a depth map acquired at the same time t, - determining a difference value Δ( P(Zfond), P(Zgeo)) between the average depth of the clipped face area Zgeo and the average depth of a background area Zfond of the depth map, and comparing this difference value with a threshold value (seuil1), sending an alert signal if this threshold value is not complied with, and otherwise, - detecting, in the image, at least two characteristic points associated with a biometric parameter of the individual, and projecting said detected characteristic points into the depth map, - checking that the size of the face corresponds to a size of a living individual, - checking the relative depths of the characteristic points in the depth map, and confirming that the face is alive in order to definitively identify the individual or to transmit an attempted fraud signal.