FSBI: Deepfakes Detection with Frequency Enhanced Self-Blended Images
This addresses the challenge of detecting increasingly realistic deepfakes for security and media integrity, representing an incremental improvement over existing methods.
The paper tackles the problem of detecting deepfakes by proposing a Frequency Enhanced Self-Blended Images (FSBI) approach, which uses Discrete Wavelet Transforms on self-blended images to extract features for a convolutional network, achieving results that outperform state-of-the-art techniques on FF++ and Celeb-DF datasets in cross-dataset evaluations.
Advances in deepfake research have led to the creation of almost perfect manipulations undetectable by human eyes and some deepfakes detection tools. Recently, several techniques have been proposed to differentiate deepfakes from realistic images and videos. This paper introduces a Frequency Enhanced Self-Blended Images (FSBI) approach for deepfakes detection. This proposed approach utilizes Discrete Wavelet Transforms (DWT) to extract discriminative features from the self-blended images (SBI) to be used for training a convolutional network architecture model. The SBIs blend the image with itself by introducing several forgery artifacts in a copy of the image before blending it. This prevents the classifier from overfitting specific artifacts by learning more generic representations. These blended images are then fed into the frequency features extractor to detect artifacts that can not be detected easily in the time domain. The proposed approach has been evaluated on FF++ and Celeb-DF datasets and the obtained results outperformed the state-of-the-art techniques with the cross-dataset evaluation protocol.