Deepfake Detection Scheme Based on Vision Transformer and Distillation

TL;DR

This paper introduces a Vision Transformer-based deepfake detection method utilizing distillation, which outperforms existing CNN-based models in accuracy and reduces false negatives.

Contribution

The study presents a novel Vision Transformer model with a distillation approach that effectively detects deepfakes, surpassing CNN-based methods on benchmark datasets.

Findings

Achieved 0.978 AUC and 91.9 F1 score without ensemble techniques.

Outperformed state-of-the-art CNN-based models on the DFDC dataset.

Demonstrated effectiveness of patch embedding in deepfake detection.

Abstract

Deepfake is the manipulated video made with a generative deep learning technique such as Generative Adversarial Networks (GANs) or Auto Encoder that anyone can utilize. Recently, with the increase of Deepfake videos, some classifiers consisting of the convolutional neural network that can distinguish fake videos as well as deepfake datasets have been actively created. However, the previous studies based on the CNN structure have the problem of not only overfitting, but also considerable misjudging fake video as real ones. In this paper, we propose a Vision Transformer model with distillation methodology for detecting fake videos. We design that a CNN features and patch-based positioning model learns to interact with all positions to find the artifact region for solving false negative problem. Through comparative analysis on Deepfake Detection (DFDC) Dataset, we verify that the proposed scheme with patch embedding as input outperforms the state-of-the-art using the combined CNN features. Without ensemble technique, our model obtains 0.978 of AUC and 91.9 of f1 score, while previous SOTA model yields 0.972 of AUC and 90.6 of f1 score on the same condition.