Faster Than Lies: Real-time Deepfake Detection using Binary Neural Networks

View on arXiv ← Back to list

Authors: Lanzino Romeo, Fontana Federico, Diko Anxhelo, Marini Marco Raoul, Cinque Luigi

Published: 2024-06-07 13:37:36+00:00

AI Summary

This paper introduces a novel deepfake detection approach using Binary Neural Networks (BNNs) for real-time inference. The method incorporates FFT and LBP features to enhance detection accuracy, achieving state-of-the-art performance with a significant reduction in computational cost.

Abstract

Deepfake detection aims to contrast the spread of deep-generated media that undermines trust in online content. While existing methods focus on large and complex models, the need for real-time detection demands greater efficiency. With this in mind, unlike previous work, we introduce a novel deepfake detection approach on images using Binary Neural Networks (BNNs) for fast inference with minimal accuracy loss. Moreover, our method incorporates Fast Fourier Transform (FFT) and Local Binary Pattern (LBP) as additional channel features to uncover manipulation traces in frequency and texture domains. Evaluations on COCOFake, DFFD, and CIFAKE datasets demonstrate our method's state-of-the-art performance in most scenarios with a significant efficiency gain of up to a $20times$ reduction in FLOPs during inference. Finally, by exploring BNNs in deepfake detection to balance accuracy and efficiency, this work paves the way for future research on efficient deepfake detection.

Key findings

The proposed BNN-based method achieves state-of-the-art or competitive performance on various datasets. It significantly reduces computational complexity (up to 20x reduction in FLOPs) compared to existing methods, enabling real-time deepfake detection. The ablation study demonstrates the effectiveness of combining FFT and LBP features for improved accuracy.

Approach

The authors use a Binary Neural Network (BNN) architecture, specifically BNext, as the backbone for deepfake detection. They augment the input RGB image with FFT magnitude and LBP features to improve detection capabilities, and utilize an adapter layer to process the combined features before feeding them into the BNN.

Datasets

COCOFake, DFFD, CIFAKE

Model(s)

BNext (Binary Neural Network)

Author countries

Italy

← Previous