Improved Xception with Dual Attention Mechanism and Feature Fusion for Face Forgery Detection

View on arXiv ← Back to list

Authors: Hao Lin, Weiqi Luo, Kangkang Wei, Minglin Liu

Published: 2021-09-29 01:54:13+00:00

AI Summary

This paper proposes an improved Xception model for face forgery detection, incorporating a dual attention mechanism (CBAM and self-attention) and feature fusion to enhance feature representation and improve classification accuracy. Experimental results on three deepfake datasets demonstrate superior performance compared to existing methods.

Abstract

With the rapid development of deep learning technology, more and more face forgeries by deepfake are widely spread on social media, causing serious social concern. Face forgery detection has become a research hotspot in recent years, and many related methods have been proposed until now. For those images with low quality and/or diverse sources, however, the detection performances of existing methods are still far from satisfactory. In this paper, we propose an improved Xception with dual attention mechanism and feature fusion for face forgery detection. Different from the middle flow in original Xception model, we try to catch different high-semantic features of the face images using different levels of convolution, and introduce the convolutional block attention module and feature fusion to refine and reorganize those high-semantic features. In the exit flow, we employ the self-attention mechanism and depthwise separable convolution to learn the global information and local information of the fused features separately to improve the classification the ability of the proposed model. Experimental results evaluated on three Deepfake datasets demonstrate that the proposed method outperforms Xception as well as other related methods both in effectiveness and generalization ability.

Key findings

The proposed method outperforms other state-of-the-art deepfake detection methods on three datasets (TIMIT, FF++, WildDeepfake), showing improvements in both effectiveness and generalization ability. Ablation studies confirm the contribution of the dual attention mechanism and feature fusion.

Approach

The authors improve the Xception architecture by adding a dual attention mechanism (CBAM in the middle flow, self-attention in the exit flow) and feature fusion in the middle flow. This enhances feature extraction and classification by refining features and integrating global and local information.

Datasets

Deepfake-TIMIT (TIMIT), FaceForensics++ (FF++), WildDeepfake, Celeb-DF

Model(s)

Improved Xception with CBAM and self-attention mechanisms, feature fusion

Author countries

China

← Previous