Temporal Facial Analysis for Image Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current face recognition systems can be fooled by pictures or masks, leading to unauthorized access, as they lack the ability to differentiate between real and fake images of a user.
Innovation Solution
A computer-implemented method using a deep neural network trained on both verified and fake face images to perform temporal facial analysis, generating a verification signal to determine if the image data is fake, and granting access only if the signal indicates a real user.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional face recognition algorithms are used, then access convenience is improved, but security reliability deteriorates because the system can be fooled by pictures or videos
Solution Approach 1:
A neural network classifier is introduced as an intermediary component between the face recognition system and the access grant decision. This classifier analyzes temporal facial data and determines whether the input is a real face or a fake image/video, thereby mediating the security verification process and preventing unauthorized access while maintaining convenience for legitimate users
Solution Approach 2:
The traditional mechanical face recognition system is enhanced by substituting it with a neural network-based temporal facial analysis system. The neural network processes sequences of facial images over time, replacing simple static face matching with a more sophisticated temporal analysis that can distinguish real faces from fake ones, thus improving security while maintaining ease of use
2Speed
If face recognition systems process only static images, then processing speed is improved, but security vulnerability increases due to inability to detect fake images
Solution Approach 1:
The system transitions from processing static facial images to processing dynamic temporal sequences of facial images. By analyzing how facial features change over time, the neural network can detect inconsistencies in fake images or videos that do not move naturally, thereby improving security without significantly increasing processing time
Solution Approach 2:
The system performs preliminary analysis of temporal facial patterns during the authentication process itself. By continuously monitoring facial movements and comparing them against learned patterns of real faces, the system can preliminarily identify fake images before final access is granted, enhancing security while maintaining efficient processing speeds
Data Source
AI summary
A computer-implemented method verifies an image based authentication via one or more processors performing operations including receiving image data corresponding to a face identified by a facial recognition system, processing the received raw image data via a deep neural network trained on training data that includes images of both verified and fake faces to perform a temporal facial analysis, and generating a verification signal in response to the temporal facial analysis to indicate whether the raw image data is fake.