A Meta-Learning Approach for Few-Shot Face Forgery Segmentation and Classification
The technology for detecting forged images is good at detecting known forgery methods. It trains neural networks using many original and corresponding forged images created with known methods. However, when encountering unseen forgery methods, the technology performs poorly. Recently, one suggested...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/7/3647 |
| _version_ | 1797606994450841600 |
|---|---|
| author | Yih-Kai Lin Ting-Yu Yen |
| author_facet | Yih-Kai Lin Ting-Yu Yen |
| author_sort | Yih-Kai Lin |
| collection | DOAJ |
| description | The technology for detecting forged images is good at detecting known forgery methods. It trains neural networks using many original and corresponding forged images created with known methods. However, when encountering unseen forgery methods, the technology performs poorly. Recently, one suggested approach to tackle this problem is to use a hand-crafted generator of forged images to create a range of fake images, which can then be used to train the neural network. However, the aforementioned method has limited detection performance when encountering unseen forging techniques that the hand-craft generator has not accounted for. To overcome the limitations of existing methods, in this paper, we adopt a meta-learning approach to develop a highly adaptive detector for identifying new forging techniques. The proposed method trains a forged image detector using meta-learning techniques, making it possible to fine-tune the detector with only a few new forged samples. The proposed method inputs a small number of the forged images to the detector and enables the detector to adjust its weights based on the statistical features of the input forged images, allowing the detection of forged images with similar characteristics. The proposed method achieves significant improvement in detecting forgery methods, with IoU improvements ranging from 35.4% to 127.2% and AUC improvements ranging from 2.0% to 48.9%, depending on the forgery method. These results show that the proposed method significantly improves detection performance with only a small number of samples and demonstrates better performance compared to current state-of-the-art methods in most scenarios. |
| first_indexed | 2024-03-11T05:25:03Z |
| format | Article |
| id | doaj.art-ab6f715f076643b0aaaf3ac43fc6f659 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T05:25:03Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-ab6f715f076643b0aaaf3ac43fc6f6592023-11-17T17:35:40ZengMDPI AGSensors1424-82202023-03-01237364710.3390/s23073647A Meta-Learning Approach for Few-Shot Face Forgery Segmentation and ClassificationYih-Kai Lin0Ting-Yu Yen1Department of Computer Science and Artificial Intelligence, National Pingtung University, No. 4-18 Minsheng Road, Pingtung City 90003, TaiwanDepartment of Computer Science and Artificial Intelligence, National Pingtung University, No. 4-18 Minsheng Road, Pingtung City 90003, TaiwanThe technology for detecting forged images is good at detecting known forgery methods. It trains neural networks using many original and corresponding forged images created with known methods. However, when encountering unseen forgery methods, the technology performs poorly. Recently, one suggested approach to tackle this problem is to use a hand-crafted generator of forged images to create a range of fake images, which can then be used to train the neural network. However, the aforementioned method has limited detection performance when encountering unseen forging techniques that the hand-craft generator has not accounted for. To overcome the limitations of existing methods, in this paper, we adopt a meta-learning approach to develop a highly adaptive detector for identifying new forging techniques. The proposed method trains a forged image detector using meta-learning techniques, making it possible to fine-tune the detector with only a few new forged samples. The proposed method inputs a small number of the forged images to the detector and enables the detector to adjust its weights based on the statistical features of the input forged images, allowing the detection of forged images with similar characteristics. The proposed method achieves significant improvement in detecting forgery methods, with IoU improvements ranging from 35.4% to 127.2% and AUC improvements ranging from 2.0% to 48.9%, depending on the forgery method. These results show that the proposed method significantly improves detection performance with only a small number of samples and demonstrates better performance compared to current state-of-the-art methods in most scenarios.https://www.mdpi.com/1424-8220/23/7/3647digital forensicsface forgery detectionU-Netsegmentationmeta-learningfew-shot learning |
| spellingShingle | Yih-Kai Lin Ting-Yu Yen A Meta-Learning Approach for Few-Shot Face Forgery Segmentation and Classification Sensors digital forensics face forgery detection U-Net segmentation meta-learning few-shot learning |
| title | A Meta-Learning Approach for Few-Shot Face Forgery Segmentation and Classification |
| title_full | A Meta-Learning Approach for Few-Shot Face Forgery Segmentation and Classification |
| title_fullStr | A Meta-Learning Approach for Few-Shot Face Forgery Segmentation and Classification |
| title_full_unstemmed | A Meta-Learning Approach for Few-Shot Face Forgery Segmentation and Classification |
| title_short | A Meta-Learning Approach for Few-Shot Face Forgery Segmentation and Classification |
| title_sort | meta learning approach for few shot face forgery segmentation and classification |
| topic | digital forensics face forgery detection U-Net segmentation meta-learning few-shot learning |
| url | https://www.mdpi.com/1424-8220/23/7/3647 |
| work_keys_str_mv | AT yihkailin ametalearningapproachforfewshotfaceforgerysegmentationandclassification AT tingyuyen ametalearningapproachforfewshotfaceforgerysegmentationandclassification AT yihkailin metalearningapproachforfewshotfaceforgerysegmentationandclassification AT tingyuyen metalearningapproachforfewshotfaceforgerysegmentationandclassification |