Authentication Transformation Model for Feature Extractor Compatibility
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing authentication methods face challenges in seamlessly transitioning from one feature extractor to another, especially when updates are made to the feature extraction model, leading to inconvenience for users and potential authentication failures.
Innovation Solution
The proposed solution involves using a transformation model that transforms enrolled features generated by an old feature extractor into enrolled features compatible with a new feature extractor, allowing for authentication to proceed based on the transformed features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the feature extractor is updated to improve authentication accuracy, then authentication accuracy is improved, but existing enrolled features become incompatible causing authentication failures
Solution Approach 1:
A transformation model is introduced as an intermediary component that converts enrolled features from the old feature extractor format into the new feature extractor format. This mediator enables compatibility between the updated authentication system and existing user enrollment data without requiring users to re-enroll, thus maintaining both authentication accuracy improvement and backward compatibility.
2Measurement precision
If the feature extractor is updated to improve authentication accuracy, then authentication accuracy is improved, but user convenience deteriorates due to re-enrollment requirements
Solution Approach 1:
The transformation model serves as a bridge that automatically translates existing enrolled features into the new feature format, eliminating the need for users to perform re-enrollment operations. This maintains user convenience while enabling the system to benefit from the updated feature extractor's improved accuracy.
Solution Approach 2:
The transformation model is pre-trained to handle the conversion of features from the old to the new format, so that when the feature extractor is updated, the system can immediately transform existing enrolled features without requiring any user action or intervention, thus preserving user convenience.
3Adaptability or versatility
If a transformation model is introduced to maintain compatibility, then feature compatibility is improved, but system complexity increases
Solution Approach 1:
The transformation model is positioned as a separate, modular intermediary component that can be independently trained and deployed. This modular approach isolates the complexity of format conversion from the main authentication pipeline, allowing the transformation model to be pre-computed and stored, thereby reducing the operational complexity during authentication while maintaining adaptability.
Data Source
AI summary
An authentication method and apparatus using a transformation model are disclosed. The authentication method includes generating, at a first apparatus, a first enrolled feature based on a first feature extractor, obtaining a second enrolled feature to which the first enrolled feature is transformed, determining an input feature by extracting a feature from input data with a second feature extractor different from the first feature extractor, and performing an authentication based on the second enrolled feature and the input feature.