Authors :
Akash Chaudhary; AnkitaSingh; Km.Yachana
Volume/Issue :
Volume 8 - 2023, Issue 5 - May
Google Scholar :
https://bit.ly/3TmGbDi
Scribd :
https://bit.ly/3q1sgb6
DOI :
https://doi.org/10.5281/zenodo.7953326
Abstract :
The ability to detect spoofed faces has become
a critical concern in various applications, such as face
recognition systems, banking, and security measures.
Thisresearchpresentsa simple system that can detect
whether a facein video stream is spoofed or real using
pre-trained models for face detection and anti-spoofing.
The system uses a continuous loop to read each frame of
the video stream, to assess whether a face image is real
or spoof, first detect faces using the pre-trained face
detection model, then crop and resize the face image. If
the model predicts that the face is fake, the system draws
a red rectangle around the face and displays the label
"spoof." If the model predicts that the face is real, the
system draws a green rectangle around the face and
displays the label "real." The proposed system achieved
a high accuracy rate in detecting spoofed faces, making it
suitable for real-world applications.
Keywords :
Facebiometric, liveness detection, Anti-spoofing, Fraud prevention, Face Spoofing detection, Convolutional neural networks.
The ability to detect spoofed faces has become
a critical concern in various applications, such as face
recognition systems, banking, and security measures.
Thisresearchpresentsa simple system that can detect
whether a facein video stream is spoofed or real using
pre-trained models for face detection and anti-spoofing.
The system uses a continuous loop to read each frame of
the video stream, to assess whether a face image is real
or spoof, first detect faces using the pre-trained face
detection model, then crop and resize the face image. If
the model predicts that the face is fake, the system draws
a red rectangle around the face and displays the label
"spoof." If the model predicts that the face is real, the
system draws a green rectangle around the face and
displays the label "real." The proposed system achieved
a high accuracy rate in detecting spoofed faces, making it
suitable for real-world applications.
Keywords :
Facebiometric, liveness detection, Anti-spoofing, Fraud prevention, Face Spoofing detection, Convolutional neural networks.