Skip to main content
. 2025 Dec 10;8:1720525. doi: 10.3389/fdata.2025.1720525

Figure 1.

Flowchart illustrating a facial recognition and anti-forensics process. An image enters two parallel paths: one through Arcface and an anti-forensics module, producing identity semantics and reconstruction, and another through spatial and frequency domains using FFT and autoencoder for feature extraction. Features are fused and optimized using an MLP.

Framework of the proposed method.