I recently read your paper "CNN-generated images are surprisingly easy to spot... for now" and found your work on DeepFake detection very insightful. I appreciate that you have provided the download links of the models for the "Blur+JPEG (0.5)" and "Blur+JPEG (0.1)" versions.

Would it be possible for you to also share the download links of the remaining models ("No aug", "Blur only", and "JPEG only") used in your experiments? Having access to these models would be extremely helpful for me in further analysis and comparison.

I would greatly appreciate your help, and I look forward to your response. Thank you for your time and consideration.

Best regards,
Imran Hasan
Email: [email protected]