Deepfake Detection via Exploring Degradation Inconsistency

Summary

Researchers propose the Degradation Consistency Learning Framework to improve deepfake face detection on forgery methods unseen during training. The method exploits degradation inconsistencies between background and manipulated face regions, using a data generation network and a detection network mining both spatial and frequency domain clues, coupled through adversarial training. The authors report effectiveness on benchmark datasets under in-dataset and cross-dataset protocols.

Topics

• Deepfakes and impersonation