Detecting Localized Deepfakes: How Well Do Synthetic Image Detectors Handle Inpainting?
This addresses the problem of detecting localized image manipulations for cybersecurity applications, but it is incremental as it focuses on evaluating existing methods rather than introducing new ones.
The paper systematically evaluated how well existing deepfake detectors, trained on fully synthetic images, perform on detecting localized inpainting manipulations, finding that models trained on diverse generators partially transfer and reliably detect medium-to-large or regeneration-style inpainting, outperforming many ad hoc methods.
The rapid progress of generative AI has enabled highly realistic image manipulations, including inpainting and region-level editing. These approaches preserve most of the original visual context and are increasingly exploited in cybersecurity-relevant threat scenarios. While numerous detectors have been proposed for identifying fully synthetic images, their ability to generalize to localized manipulations remains insufficiently characterized. This work presents a systematic evaluation of state-of-the-art detectors, originally trained for the deepfake detection on fully synthetic images, when applied to a distinct challenge: localized inpainting detection. The study leverages multiple datasets spanning diverse generators, mask sizes, and inpainting techniques. Our experiments show that models trained on a large set of generators exhibit partial transferability to inpainting-based edits and can reliably detect medium- and large-area manipulations or regeneration-style inpainting, outperforming many existing ad hoc detection approaches.