Skip to content
Conference Open access

HEL-CAW-Net: Face Liveness Detection Using Light Intensity-Aware Convolutional Attentional W-Net Model

Sep 2026 · 2026 7th International Conference on Smart Electronics and Communication (ICOSEC) · pp. 1730-1737 · 0 citations · 17 references

Abstract

Face liveness detection is a major component in securing biometric authentication systems from spoofing attacks with recorded video and printed photos. Therefore, numerous detection methods have been developed to detect face liveness, but they face notable disadvantages, including higher computational complexity, poor accuracy, higher false positive rate, limited generalization ability, and so on. Hence, this research proposes a Hybrid Encoder Light Intensity-aware Convolution Attentional W-Net (HEL-CAW-Net) model to detect face liveness accurately. The Hybrid W-Net encoded De-fake Face Generator enables the model to identify the anomalous cue for accurate face liveness detection. Moreover, HEL-CAW-Net minimizes the false positive rates by incorporating the Mutual Cross Attention (MCA) network and Convolutional Neural Network (CNN) model. Based on the experimental results, HEL-CAW-Net outperforms the existing methods in terms of various evaluation metrics. The proposed method achieved an accuracy of 98.58%, precision of 99.12%, sensitivity of 98.31%, and F1-score of 98.71% using Casia-fasd with 90% of training samples, respectively.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.