Related Experiment Video
Updated: Jan 9, 2026
06:37
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
Published on: December 15, 2023
5.2K
AIFS: an efficient face recognition method based on AI and enhanced few-shot learning.
1Smart Systems Engineering Laboratory, Communications and Networks Engineering Department, College of Engineering, Prince Sultan University, Riyadh, 11586, Saudi Arabia. mnasralla@psu.edu.sa.
Scientific Reports
|November 29, 2025
Summary
This study presents AIFS, a hybrid facial recognition system combining traditional and deep learning methods. It achieves 99% accuracy in low-data scenarios, offering a scalable solution for real-time AI applications.
Area of Science:
- Artificial Intelligence
- Computer Vision
- Machine Learning
Background:
- Facial recognition demand is rising in resource-limited settings.
- Current systems struggle with low data and limited computation.
- Scalable AI solutions are needed for real-time adaptive facial recognition.
Purpose of the Study:
- Introduce AIFS, an efficient hybrid facial recognition framework.
- Unify traditional feature-based learning with few-shot deep learning.
- Address limitations of existing facial recognition systems.
Main Methods:
- A hybrid Siamese architecture combining edge and cloud paths.
- Edge path: Viola-Jones algorithm with Particle Swarm Optimization (PSO).
- Cloud path: Siamese network with triplet loss, EfficientNetV2, and InceptionV3.
Main Results:
- AIFS framework validated on CPUs, Raspberry Pi, and GPUs.
- Achieved up to 99% accuracy on the Kaggle Face Recognition Dataset.
- Demonstrated balance between latency, inference speed, and resource efficiency.
Conclusions:
- AIFS is a scalable and robust solution for real-time facial recognition.
- Effective in heterogeneous computing environments and low-data settings.
- Suitable for telemedicine, surveillance, and biometric authentication.
