Two-Stream Xception Structure Based on Feature Fusion for DeepFake Detection
Abstract DeepFake may have a crucial impact on people’s lives and reduce the trust in digital media, so DeepFake detection methods have developed rapidly. Most existing detection methods rely on single-space features (mostly RGB features), and there is still relatively little research on multi-space...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2023-08-01
|
| Series: | International Journal of Computational Intelligence Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s44196-023-00312-8 |
| _version_ | 1827633182732713984 |
|---|---|
| author | Bin Wang Liqing Huang Tianqiang Huang Feng Ye |
| author_facet | Bin Wang Liqing Huang Tianqiang Huang Feng Ye |
| author_sort | Bin Wang |
| collection | DOAJ |
| description | Abstract DeepFake may have a crucial impact on people’s lives and reduce the trust in digital media, so DeepFake detection methods have developed rapidly. Most existing detection methods rely on single-space features (mostly RGB features), and there is still relatively little research on multi-space feature fusion. At the same time, a lot of existing methods used a single receptive field, which leads to models that cannot extract information of different scales. In order to solve the above problems, we propose a two-stream Xception network structure (Tception) that fused RGB spatial feature and noise-space feature. This network structure consists of two main parts. The first part is a feature fusion module, which can adaptively fuse RGB feature and noise-space feature generated by RGB images through SRM filters. The second part is the two-stream network structure, which utilizes a parallel structure of convolutional kernels of different sizes allowing the network to learn features of different scales. The experiments show that the proposed method improves performance compared to the Xception network. Compared to SSTNet, the detection accuracy of the Neural Textures is improved by nearly 8%. |
| first_indexed | 2024-03-09T14:55:58Z |
| format | Article |
| id | doaj.art-3f921dacaece4170b13ddba8aa02e879 |
| institution | Directory Open Access Journal |
| issn | 1875-6883 |
| language | English |
| last_indexed | 2024-03-09T14:55:58Z |
| publishDate | 2023-08-01 |
| publisher | Springer |
| record_format | Article |
| series | International Journal of Computational Intelligence Systems |
| spelling | doaj.art-3f921dacaece4170b13ddba8aa02e8792023-11-26T14:11:53ZengSpringerInternational Journal of Computational Intelligence Systems1875-68832023-08-0116111110.1007/s44196-023-00312-8Two-Stream Xception Structure Based on Feature Fusion for DeepFake DetectionBin Wang0Liqing Huang1Tianqiang Huang2Feng Ye3College of Computer and Cyberspace Security, Fujian Normal UniversityCollege of Computer and Cyberspace Security, Fujian Normal UniversityCollege of Computer and Cyberspace Security, Fujian Normal UniversityCollege of Computer and Cyberspace Security, Fujian Normal UniversityAbstract DeepFake may have a crucial impact on people’s lives and reduce the trust in digital media, so DeepFake detection methods have developed rapidly. Most existing detection methods rely on single-space features (mostly RGB features), and there is still relatively little research on multi-space feature fusion. At the same time, a lot of existing methods used a single receptive field, which leads to models that cannot extract information of different scales. In order to solve the above problems, we propose a two-stream Xception network structure (Tception) that fused RGB spatial feature and noise-space feature. This network structure consists of two main parts. The first part is a feature fusion module, which can adaptively fuse RGB feature and noise-space feature generated by RGB images through SRM filters. The second part is the two-stream network structure, which utilizes a parallel structure of convolutional kernels of different sizes allowing the network to learn features of different scales. The experiments show that the proposed method improves performance compared to the Xception network. Compared to SSTNet, the detection accuracy of the Neural Textures is improved by nearly 8%.https://doi.org/10.1007/s44196-023-00312-8Deep learningFeature fusionTwo-stream structure |
| spellingShingle | Bin Wang Liqing Huang Tianqiang Huang Feng Ye Two-Stream Xception Structure Based on Feature Fusion for DeepFake Detection International Journal of Computational Intelligence Systems Deep learning Feature fusion Two-stream structure |
| title | Two-Stream Xception Structure Based on Feature Fusion for DeepFake Detection |
| title_full | Two-Stream Xception Structure Based on Feature Fusion for DeepFake Detection |
| title_fullStr | Two-Stream Xception Structure Based on Feature Fusion for DeepFake Detection |
| title_full_unstemmed | Two-Stream Xception Structure Based on Feature Fusion for DeepFake Detection |
| title_short | Two-Stream Xception Structure Based on Feature Fusion for DeepFake Detection |
| title_sort | two stream xception structure based on feature fusion for deepfake detection |
| topic | Deep learning Feature fusion Two-stream structure |
| url | https://doi.org/10.1007/s44196-023-00312-8 |
| work_keys_str_mv | AT binwang twostreamxceptionstructurebasedonfeaturefusionfordeepfakedetection AT liqinghuang twostreamxceptionstructurebasedonfeaturefusionfordeepfakedetection AT tianqianghuang twostreamxceptionstructurebasedonfeaturefusionfordeepfakedetection AT fengye twostreamxceptionstructurebasedonfeaturefusionfordeepfakedetection |