Modeling Spoof Noise by De-spoofing Diffusion and its Application in Face Anti-spoofing
This addresses security issues in face recognition systems by improving generalization over GAN-based methods, though it appears incremental as it adapts diffusion models to an existing approach.
The paper tackled the problem of face anti-spoofing by using diffusion models to denoise spoof images and extract spoof noise as a discriminative cue, achieving competitive performance in accuracy and generalization on intra- and inter-testing protocols.
Face anti-spoofing is crucial for ensuring the security and reliability of face recognition systems. Several existing face anti-spoofing methods utilize GAN-like networks to detect presentation attacks by estimating the noise pattern of a spoof image and recovering the corresponding genuine image. But GAN's limited face appearance space results in the denoised faces cannot cover the full data distribution of genuine faces, thereby undermining the generalization performance of such methods. In this work, we present a pioneering attempt to employ diffusion models to denoise a spoof image and restore the genuine image. The difference between these two images is considered as the spoof noise, which can serve as a discriminative cue for face anti-spoofing. We evaluate our proposed method on several intra-testing and inter-testing protocols, where the experimental results showcase the effectiveness of our method in achieving competitive performance in terms of both accuracy and generalization.