The Detection Optimization of Low-Quality Fake Face Images: Feature Enhancement and Noise Suppression Strategies
Source abstract
Abstract
With the rapid advancement of deepfake technology, the detection of low-quality synthetic facial images has become increasingly challenging, particularly in cases involving low resolution, blurriness, or noise. Traditional detection methods often exhibit limited performance under such conditions. To address these limitations, this paper proposes a novel algorithm, YOLOv9-ARC, which is designed to enhance the accuracy of detecting low-quality fake facial images. The proposed algorithm introduces an innovative convolution module, Adaptive Kernel Convolution (AKConv), which dynamically adjusts kernel sizes to effectively extract image features, thereby mitigating the challenges posed by low resolution, blurriness, and noise. Furthermore, a hybrid attention mechanism, Convolutional Block Attention Module (CBAM), is integrated to amplify salient features while suppressing irrelevant information. Extensive experiments demonstrate that YOLOv9-ARC achieves a mean average precision (mAP) of 75.1% on the DFDC (DeepFake Detection Challenge) dataset, representing a 3.5% improvement over the baseline model. The proposed YOLOv9-ARC not only addresses the challenges of low-quality deepfake detection but also demonstrates significant improvements in accuracy within this domain.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Digital Media Forensic Detection · Generative Adversarial Networks and Image Synthesis · Face recognition and analysis
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Intelligent Equipment, Shandong University of Science and Technology, Taian 271019, China
Funding
Shandong Provincial Science and Technology-Based Small and Medium-Sized Enterprises Innovation Capacity Improvement Project
References (30)
Show source-provided references
1. Generative adversarial nets
2. Deep learning for deepfakes creation and detection: A survey10.1016/j.cviu.2022.103525 ↗
3. The distinct wrong of deepfakes10.1007/s13347-021-00459-2 ↗
4. Zhao, H., Zhou, W., Chen, D., Wei, T., Zhang, W., and Yu, N. (2021, January 19–25). Multi-attentional deepfake detection. Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA.10.1109/CVPR46437.2021.00222 ↗
5. Lee, S., Tariq, S., Kim, J., and Woo, S.S. (2021, January 22–24). TAR: Generalized forensic framework to detect deepfakes using weakly supervised learning. Proceedings of the IFIP International Conference on ICT Systems Security and Privacy Protection, Oslo, Norway.10.1007/978-3-030-78120-0_23 ↗
6. Rössler, A., Cozzolino, D., Verdoliva, L., Riess, C., Thies, J., and Nießner, M. (November, January 27). FaceForensics++: Learning to detect manipulated facial images. Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, Republic of Korea.10.1109/ICCV.2019.00009 ↗
7. Wang, C., and Deng, W. (2021, January 19–25). Representative forgery mining for fake face detection. Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA.10.1109/CVPR46437.2021.01468 ↗
8. Deepfakes and beyond: A survey of face manipulation and fake detection10.1016/j.inffus.2020.06.014 ↗
9. Dolhansky, B., Bitton, J., Pflaum, B., Lu, J., Howes, R., Wang, M., and Ferrer, C.C. (2020). The Deepfake Detection Challenge (DFDC) Dataset. arXiv.
10. Zhou, Y., Liu, D., and Huang, T. (2018, January 15–19). Survey of face detection on low-quality images. Proceedings of the 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018), Xi’an, China.10.1109/FG.2018.00121 ↗
11. A survey on deepfake video detection10.1049/bme2.12031 ↗
12. DIFLD: Domain invariant feature learning to detect low-quality compressed face forgery images10.1007/s40747-023-01160-x ↗
13. Image quality assessment based fake face detection10.1007/s11042-021-11493-9 ↗
14. Forgery detection of low quality deepfake videos10.14311/NNW.2023.33.006 ↗
15. Deepfakes generation and detection: State-of-the-art, open challenges, countermeasures, and way forward10.1007/s10489-022-03766-z ↗
16. Wang, Z., Liu, D., Yang, J., Han, W., and Huang, T. (2015, January 7–13). Deep networks for image super-resolution with sparse prior. Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, Chile.10.1109/ICCV.2015.50 ↗
17. Herrmann, C. (2013, January 27–30). Extending a local matching face recognition approach to low-resolution video. Proceedings of the 2013 10th IEEE International Conference on Advanced Video and Signal Based Surveillance, Krakow, Poland.10.1109/AVSS.2013.6636683 ↗
18. Face recognition in low-quality images using adaptive sparse representations10.1016/j.imavis.2019.02.012 ↗
19. Kim, H.-I., Lee, S.H., and Ro, Y.M. (2014, January 27–30). Adaptive feature extraction for blurred face images in facial expression recognition. Proceedings of the 2014 IEEE International Conference on Image Processing (ICIP), Paris, France.10.1109/ICIP.2014.7026205 ↗
20. Lin, S., Jiang, C., Liu, F., and Shen, L. (2021, January 4–7). High quality facial data synthesis and fusion for 3D low-quality face recognition. Proceedings of the 2021 IEEE International Joint Conference on Biometrics (IJCB), Shenzhen, China.10.1109/IJCB52358.2021.9484339 ↗
21. Mu, G., Huang, D., Hu, G., Sun, J., and Wang, Y. (2019, January 15–20). LED3D: A lightweight and efficient deep approach to recognizing low-quality 3D faces. Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.10.1109/CVPR.2019.00592 ↗
22. Zhang, J., Huang, D., Wang, Y., and Sun, J. (2016, January 13–16). Lock3DFace: A large-scale database of low-cost Kinect 3D faces. Proceedings of the 2016 International Conference on Biometrics (ICB), Halmstad.
23. Wang, C.-Y., Yeh, I.-H., and Liao, H.-Y.M. (October, January 29). YOLOv9: Learning what you want to learn using programmable gradient information. Proceedings of the European Conference Computer Vision, Milan, Italy.
24. Woo, S., Park, J., Lee, J.-Y., and Kweon, I.S. (2018, January 8–14). CBAM: Convolutional block attention module. Proceedings of the European Conference Computer Vision, Munich, Germany.10.1007/978-3-030-01234-2_1 ↗
25. Zhang, X., Song, Y., Song, T., Yang, D., Ye, Y., Zhou, J., and Zhang, L. (2023). AKConv: Convolutional kernel with arbitrary sampled shapes and arbitrary number of parameters. arXiv.
26. Faster R-CNN: Towards real-time object detection with region proposal networks
27. Tan, M., and Le, Q. (2019, January 9–15). EfficientNet: Rethinking model scaling for convolutional neural networks. Proceedings of the 36th International Conference on Machine Learning, Long Beach, CA, USA.
28. Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (2016, January 27–30). You only look once: Unified, real-time object detection. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.10.1109/CVPR.2016.91 ↗
29. Bochkovskiy, A., Wang, C.-Y., and Liao, H.-Y.M. (2020). YOLOv4: Optimal speed and accuracy of object detection. arXiv.
30. Gradient-based learning applied to document recognition10.1109/5.726791 ↗