Noise-Resilient and Lightweight Deep Hybrid Framework for Age-Invariant Face Recognition Using Advanced Preprocessing and Optimization Techniques
TL;DR
This paper introduces a lightweight, noise-resilient deep hybrid framework for age-invariant face recognition, addressing accuracy and efficiency issues in noisy and diverse real-world scenarios using advanced preprocessing and optimization techniques.
Noise-Resilient and Lightweight Deep Hybrid Framework for Age-Invariant Face Recognition Using Advanced Preprocessing and Optimization Techniques
K. Ramya; M. Jasmine Pemeena Priyadarsini
https://doi.org/10.1109/ACCESS.2025.3647577
Volume 14
Real-world face recognition programs lose accuracy and computing efficiency due to image quality degradation and ageing sets. A lightweight, noise-resilient deep hybrid architecture for age-invariant face detection in noisy, low-quality, and demographically diverse contexts is presented in this paper. Denoising Autoencoders (DAE) decrease noise and Cycle-Consistent Generative Adversarial Networks ...