On-the-fly Facial Recognition Enrollment via Virtual Authentication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing facial recognition systems require users to reenroll their data with each software update, leading to a cumbersome experience, especially when new neural network models are introduced, as they need to transition seamlessly from the current model to the updated one without compromising security or user experience.
Innovation Solution
Implementing a 'virtual' facial recognition authentication process that operates alongside the current system, using a template generated from successful unlock attempts to compare performance with the current neural network, allowing for a seamless transition to the new model by assessing and switching to the new neural network once it meets performance criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a new neural network model is introduced for facial recognition, then recognition accuracy and security are improved, but users must reenroll their data causing inconvenience and loss of time
Solution Approach 1:
The system performs preliminary actions by collecting and processing facial images continuously in the background during normal device usage. These images are stored and prepared in advance, so when a new neural network model is introduced, the enrollment process can be completed quickly using pre-collected data rather than requiring users to spend time posing for new images.
Solution Approach 2:
The system enables self-service by automatically collecting facial images during normal device operation without requiring user intervention. The device continuously captures and processes facial data in the background, allowing the enrollment system to service itself and prepare enrollment data autonomously, eliminating the need for users to manually reenroll when models are updated.
2Ease of operation
If facial images are continuously collected for enrollment, then transition to new models becomes seamless, but device complexity and data storage requirements increase
Solution Approach 1:
The system extracts only the necessary facial images that meet quality criteria from the continuous stream of captured images. Rather than storing all captured images, the system identifies and extracts suitable enrollment images based on specific conditions (clear face visibility, proper lighting, etc.), reducing storage requirements and simplifying the overall system architecture.
Solution Approach 2:
The system introduces an intermediary component that acts as a buffer between continuous image capture and the enrollment process. This intermediary layer pre-processes and stores facial images in a ready-to-use format, mediating between the camera system and the neural network enrollment, thereby simplifying the transition process when new models are introduced.
3Reliability
If processed data is stored instead of raw user data, then security is enhanced, but flexibility for model updates is reduced
Solution Approach 1:
The system implements dynamics by maintaining processed enrollment data that can be dynamically regenerated when new neural network models are introduced. Rather than storing static raw images that would require reprocessing, the system stores processed facial representations that can be quickly adapted to new models, providing both security through processed data storage and flexibility for model updates.
Data Source
AI summary
When a software update is provided to a device that implements a facial recognition authentication process, a new authentication algorithm to operate the facial recognition authentication process may be included as part of software update. For a period of time, the new authentication algorithm may operate a “virtual” facial recognition authentication process alongside operation of the existing facial recognition authentication process using the existing (e.g., earlier version) authentication algorithm. The performance of the new authentication algorithm in providing facial recognition authentication (as assessed by the “virtual” process) may be compared to the performance of the existing authentication algorithm in providing facial recognition authentication during the period of time. When the performance of the new authentication algorithm is determined to have a satisfactory performance, operation of the actual facial recognition authentication process on the device may be switched to the new authentication algorithm.