Attributing Fake Images to GANs: Analyzing Fingerprints in Generated Images
By Ning Yu, Larry S. Davis, Mario Fritz. 2018
43 citations recorded by the source corpus.
Subfields: Generative Models, Safety, Robustness & Interpretability
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Generative Models
- Richard Zhang
- Jun-Yan Zhu
- Eli Shechtman
- Luisa Verdoliva
- Alexei A. Efros
- Oliver Wang
- Daniel Cohen-Or
- Bolei Zhou
In Safety, Robustness & Interpretability
- Luisa Verdoliva
- Hany Farid
- Richard Zhang
- Alexei A. Efros
- Francesco Marra
- Diego Gragnaniello
- Andrew Owens
- Oliver Wang
Citations in the corpus
- CNN-Generated Images Are Surprisingly Easy to Spot… for Now (2019, methodology): “Researchers have also proposed methods for identifying which, of several, known GANs generated a given image [30, 47].”
- Analyzing and Improving the Image Quality of StyleGAN (2019, background): “At present, classifier-based methods can quite reliably detect generated images, regardless of their exact origin [24, 39, 35, 45, 36].”
- Detecting and Simulating Artifacts in GAN Fake Images (2019, methodology): “GAN fake image classifier is to collect a large number of GAN generated images from one or multiple pre-trained GAN models and train a binary classifier [4], [5].”
- Incremental learning for the detection and classification of GAN-generated images (2019, background): “These artificial fingerprints have been studied in [11], [12], they have been shown to characterize not only a given architecture but also a specific instantiation of it, e.”
- DeepFakes and Beyond: A Survey of Face Manipulation and Fake Detection (2020, background): “Later on, Yu et al. analysed in [29] the existence and uniqueness of GAN fingerprints in order to detect fake images.”
- Learning Self-Consistency for Deepfake Detection (2020, background): “, JPEG compression patterns [1], compression rates) or image synthesis models [58].”
- Source Generator Attribution via Inversion (2019, methodology): “[24].”
- Do GANs Leave Artificial Fingerprints? (2018, background): “Recently, it has also been shown [21] that a deep network can reliably discriminate images generated with different architectures.”
- Protecting World Leaders Against Deep Fakes (2019, background): “Other forensic techniques exploit low-level pixel artifacts introduced during synthesis [17, 1, 21, 24, 32, 13, 25, 19].”