A hierarchical feature selection strategy for deepfake video detection

Digital face manipulation has become a concern in the last few years due to its harmful impacts on society. It is especially concerning for high-profile celebrities because their identities can be easily manipulated using mobile or web applications such as FaceSwap and FaceApp. These manipulated faces are so close to real ones that it becomes really hard to detect them, even with bare eyes. Though deep learning-based models are predominantly used by researchers, they hardly check for the presence of irrelevant or redundant features produced by those models. To this end, we have proposed a hierarchical feature selection (HFS)-based method to detect deepfake images or videos. First, we have extracted both handcrafted and deep learning features from the inputs. Next, the HFS is applied to select a near optimal set of features. In each stage of it, a hybrid feature selection method is employed that integrates a population-based meta-heuristic model, called Grey Wolf Optimization, and a single solution-based meta-heuristic model, called the Vortex Search algorithm. We have evaluated our model on three publicly available datasets, namely Celeb-DF (V2), FaceForensics++, and Deepfake Detection Challenge (DFDC). The model provides 99.35%, 99.16%, and 85.67% AUC scores on the Celeb-DF (V2), FaceForensics++, and DFDC datasets, respectively, while using only 11.50%, 12.65%, and 10.22% of actual features. Besides, our model outperforms most of the state-of-the-art methods found in our literature review and evaluated on these three datasets.

Automated summary

Digital face manipulation has become a significant concern recently due to its harmful effects on society, particularly for high-profile celebrities who can easily be targeted using apps like FaceSwap and FaceApp. Detecting deepfake images or videos is challenging, and existing models often fail to check for irrelevant or redundant features. In this study, a hierarchical feature selection (HFS) method using a hybrid population-based meta-heuristic model and a single solution-based meta-heuristic model was proposed. The model achieved high AUC scores on three publicly available datasets and outperformed most state-of-the-art methods.