DeepFidelity: Perceptual Forgery Fidelity Assessment for Deepfake Detection

TL;DR

DeepFidelity introduces a novel face forgery detection framework that assesses perceptual fidelity to improve accuracy across varying face image qualities, utilizing a specialized transformer network and outperforming existing methods.

Contribution

The paper proposes DeepFidelity, a new deepfake detection approach that incorporates perceptual fidelity assessment and a symmetric spatial attention transformer to handle diverse face image qualities.

Findings

Outperforms state-of-the-art methods on benchmark datasets.

Effectively distinguishes real and fake faces across different image qualities.

Enhances detection accuracy by modeling perceptual forgery fidelity.

Abstract

Deepfake detection refers to detecting artificially generated or edited faces in images or videos, which plays an essential role in visual information security. Despite promising progress in recent years, Deepfake detection remains a challenging problem due to the complexity and variability of face forgery techniques. Existing Deepfake detection methods are often devoted to extracting features by designing sophisticated networks but ignore the influence of perceptual quality of faces. Considering the complexity of the quality distribution of both real and fake faces, we propose a novel Deepfake detection framework named DeepFidelity to adaptively distinguish real and fake faces with varying image quality by mining the perceptual forgery fidelity of face images. Specifically, we improve the model's ability to identify complex samples by mapping real and fake face data of different qualities to different scores to distinguish them in a more detailed way. In addition, we propose a network structure called Symmetric Spatial Attention Augmentation based vision Transformer (SSAAFormer), which uses the symmetry of face images to promote the network to model the geographic long-distance relationship at the shallow level and augment local features. Extensive experiments on multiple benchmark datasets demonstrate the superiority of the proposed method over state-of-the-art methods.