Incremental learning for the detection and classification of GAN-generated images

By Francesco Marra, Cristiano Saltori, Giulia Boato, Luisa Verdoliva. 2019

160 citations recorded by the source corpus.

Current developments in computer vision and deep learning allow to automatically generate hyper-realistic images, hardly distinguishable from real ones. In particular, human face generation achieved a stunning level of realism, opening new opportunities for the creative industry but, at the same time, new scary scenarios where such content can be maliciously misused. Therefore, it is essential to develop innovative methodologies to automatically tell apart real from computer generated multimedia, possibly able to follow the evolution and continuous improvement of data in terms of quality and realism. In the last few years, several deep learning-based solutions have been proposed for this problem, mostly based on Convolutional Neural Networks (CNNs). Although results are good in controlled conditions, it is not clear how such proposals can adapt to real-world scenarios, where learning needs to continuously evolve as new types of generated data appear. In this work, we tackle this problem by proposing an approach based on incremental learning for the detection and classification of GAN-generated images. Experiments on a dataset comprising images generated by several GAN-based architectures show that the proposed method is able to correctly perform discrimination when new GANs are presented to the network.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Xiaogang Wang
3. Fang Wen
4. Xiaoou Tang
5. Dacheng Tao
6. Dong Chen
7. Ming-Hsuan Yang
8. Jianmin Bao

Citations in the corpus

• S-Prompts Learning with Pre-trained Transformers: An Occam's Razor for Domain Incremental Learning (2022, methodology): “…[6] 79.75 ± 0.84* DyTox [14] 79.21 ± 0.10 L2P [66] 81.07 ± 0.13* EWC [32] 0/class 49.19 SimCLR [10]-CaSSLe [18] 0/class 44.2* BYOL [20]-CaSSLe [18] 49.7* Barlow Twins[72]-CaSSLe [18] 48.9* Supervised Contrastive [30]-CaSSLe [18] 50.9* L2P [66] 40 image/language prompting for S-iPrompts/S-liPrompts.”
• DeepFakes and Beyond: A Survey of Face Manipulation and Fake Detection (2020, methodology): “In general, a poor generalisation capacity is observed in most cases even for state-of-the-art detection systems [17], [32].”
• Global Texture Enhancement for Fake Face Detection in the Wild (2020, unknown): “[25] detected fake images with incremental learning.”
• A Continual Deepfake Detection Benchmark: Dataset, Methods, and Essentials (2022, methodology): “Only CoReD [36], GANfake [53] and our CDDB study continual fake detection benchmarks.”
• Face X-Ray for More General Face Forgery Detection (2019, methodology): “To handle new generated images, an incremental learning strategy is introduced in [27].”
• Deep learning for deepfakes creation and detection: A survey (2019, background): “, in [104, 116, 160, 169, 170], but more work needs to be done in this direction.”
• Deepfakes generation and detection: state-of-the-art, open challenges, countermeasures, and way forward (2021, methodology): “This approach [158] is robust to unseen GAN-generated samples but unable to perform well if the information on the fake content generation method is not available.”
• Media Forensics and DeepFakes: An Overview (2020, methodology): “In [183] a method based on incremental learning is proposed to”