Deep Pixel-wise Binary Supervision for Face Presentation Attack Detection
This addresses security vulnerabilities in face recognition systems for biometric authentication, particularly in unattended scenarios, with incremental improvements in detection accuracy.
The paper tackles face presentation attack detection by introducing a CNN framework with deep pixel-wise supervision, achieving 0% HTER on Replay Mobile and 0.42% ACER on OULU Protocol-1, outperforming state-of-the-art methods.
Face recognition has evolved as a prominent biometric authentication modality. However, vulnerability to presentation attacks curtails its reliable deployment. Automatic detection of presentation attacks is essential for secure use of face recognition technology in unattended scenarios. In this work, we introduce a Convolutional Neural Network (CNN) based framework for presentation attack detection, with deep pixel-wise supervision. The framework uses only frame level information making it suitable for deployment in smart devices with minimal computational and time overhead. We demonstrate the effectiveness of the proposed approach in public datasets for both intra as well as cross-dataset experiments. The proposed approach achieves an HTER of 0% in Replay Mobile dataset and an ACER of 0.42% in Protocol-1 of OULU dataset outperforming state of the art methods.