# DS-Xception: Dual-Stream Noise Residual Architecture with Few-Shot Domain Anchor for Generalizable Diffusion-Based Deepfake Video Detection > Slat C.C. URL kanonis: https://discover.unhas.ac.id/publications/pub_scopus_105055441101 Jurnal / Konferensi: 2026 IEEE International Conference on Artificial Intelligence and Mechatronics Systems Aims 2026 Tahun terbit: 2026 DOI: https://doi.org/10.1109/AIMS71270.2026.11712590 Citations: 0 ## Authors - Slat C.C. ## Abstract The rapid shift from Generative Adversarial Networks (GANs) to diffusion models has rendered conventional deepfake detectors ineffective, as GAN-optimized architectures suffer catastrophic cross-domain performance drops. We reveal a critical motion-field divergence where GAN content exhibits erratic temporal fields while diffusion pipelines produce excessively smooth trajectories, creating a motion smoothness trap that yields an average cross-domain predicted fake probability of only 7.43%. To address this, we propose DS-Xception, a dual-stream architecture fusing a spatial stream with a Photo Response Non-Uniformity (PRNU) noise residual stream, augmented by a Few-Shot Domain Anchor strategy injecting only 50 diffusion samples (0.5% of training data) to calibrate cross-paradigm decision boundaries. DS-Xception achieves a cross-domain AUC of 0.9257 and recall of 86.06%, up from a baseline of 0.6234, while preserving an intra-domain AUC of 0.9989. ## Keywords - Computer science - Residual - Domain (mathematical analysis) - Noise (video) - Artificial intelligence - Computer vision - Pattern recognition (psychology) - Architecture - Algorithm - Background noise - Key (lock) - Object detection - Machine learning - Signal processing - Measure (data warehouse) - Noise reduction --- Sumber: Discover Unhas — RIMS Universitas Hasanuddin. Saat mengutip, gunakan DOI bila tersedia atau URL kanonis di atas.