Detection of Anti-spoofing Face using Yolo

Abstract

This paper presents a novel approach for face anti-spoofing detection using the YOLO object detection algorithm. By training a YOLO model on a diverse dataset of real and spoofed facial images, the system effectively identifies and classifies faces as genuine or fake in real-time. The proposed method leverages YOLO's high processing speed to analyze facial features and subtle visual cues, enabling robust detection of presentation attacks commonly used to deceive facial recognition systems, thus enhancing security in applications requiring facial authentication.