Open Access

ARTICLE

SMNDNet for Multiple Types of Deepfake Image Detection

Qin Wang¹, Xiaofeng Wang^2,*, Jianghua Li², Ruidong Han², Zinian Liu¹, Mingtao Guo³

1 Department of Computer Science and Engineering, Xi’an University of Technology, Xi’an, 710048, China
2 Department of Mathematics, Xi’an University of Technology, Xi’an, 710054, China
3 The National Key Laboratory of Fundamental Science on Synthetic Vision, Sichuan University, Chengdu, 610065, China

* Corresponding Author: Xiaofeng Wang. Email:

Computers, Materials & Continua 2025, 83(3), 4607-4621. https://doi.org/10.32604/cmc.2025.063141

Received 06 January 2025; Accepted 27 February 2025; Issue published 19 May 2025

Abstract

The majority of current deepfake detection methods are constrained to identifying one or two specific types of counterfeit images, which limits their ability to keep pace with the rapid advancements in deepfake technology. Therefore, in this study, we propose a novel algorithm, Stereo Mixture Density Network (SMNDNet), which can detect multiple types of deepfake face manipulations using a single network framework. SMNDNet is an end-to-end CNN-based network specially designed for detecting various manipulation types of deepfake face images. First, we design a Subtle Distinguishable Feature Enhancement Module to emphasize the differentiation between authentic and forged features. Second, we introduce a Multi-Scale Forged Region Adaptive Module that dynamically adapts to extract forged features from images of varying synthesis scales. Third, we integrate a Nonlinear Expression Capability Enhancement Module to augment the model’s capacity for capturing intricate nonlinear patterns across various types of deepfakes. Collectively, these modules empower our model to efficiently extract forgery features from diverse manipulation types, ensuring a more satisfactory performance in multiple-types deepfake detection. Experiments show that the proposed method outperforms alternative approaches in detection accuracy and AUC across all four types of deepfake images. It also demonstrates strong generalization on cross-dataset and cross-type detection, along with robust performance against post-processing manipulations.

---

Keywords

---