Table 1 Highlights of traditional model applied for literature review.
From: Secure facial biometric authentication in smart cities using multimodal methodology
Reference No/Year | Model | Accuracy | Error/Loss/time | Remarks |
|---|---|---|---|---|
2/2024 | Facial ID with Hormonic Encryption | 96.80 | 0.48 | Hardware Dependent and continuous restart issue arises |
3/2018 | Efficient Homomorphic Face Matching C + + Library | 97 | Minimal loss | High computational cost |
5/2022 | FaceNet Neural Network | 90 | 0.9 | Very Slow compared to many traditional model |
6/2021 | LOP encoding CKKS | 96 | 0.62ms | Limited available dataset |
7/2023 | LR based Facial Mapping | 67.5 | High Loss | Low Accuracy, High computational loss |
8/2020 | Dimensional feature vector with real and imaginary part | 91 | 2.83ms | High result regeneration time |
11/2020 | Adaptive Fuzzy genetic model | 96 | 1.16 | High computational time |
13/2025 | Reinforcement Learning | 99 | Minimal loss | Suitable for medical internet of things |
16/2016 | Multimodal biometric analyser | 97 | 4.89ms | Slow facial recognition process |
