Detection of Synthetic Face Images: Accuracy, Robustness, Generalization
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908541531455488 |
|---|---|
| author | Petrzelkova, Nela Cech, Jan |
| author_facet | Petrzelkova, Nela Cech, Jan |
| contents | An experimental study on detecting synthetic face images is presented. We collected a dataset, called FF5, of five fake face image generators, including recent diffusion models. We find that a simple model trained on a specific image generator can achieve near-perfect accuracy in separating synthetic and real images. The model handles common image distortions (reduced resolution, compression) by using data augmentation. Moreover, partial manipulations, where synthetic images are blended into real ones by inpainting, are identified and the area of the manipulation is localized by a simple model of YOLO architecture. However, the model turned out to be vulnerable to adversarial attacks and does not generalize to unseen generators. Failure to generalize to detect images produced by a newer generator also occurs for recent state-of-the-art methods, which we tested on Realistic Vision, a fine-tuned version of StabilityAI's Stable Diffusion image generator. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_17547 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Detection of Synthetic Face Images: Accuracy, Robustness, Generalization Petrzelkova, Nela Cech, Jan Computer Vision and Pattern Recognition An experimental study on detecting synthetic face images is presented. We collected a dataset, called FF5, of five fake face image generators, including recent diffusion models. We find that a simple model trained on a specific image generator can achieve near-perfect accuracy in separating synthetic and real images. The model handles common image distortions (reduced resolution, compression) by using data augmentation. Moreover, partial manipulations, where synthetic images are blended into real ones by inpainting, are identified and the area of the manipulation is localized by a simple model of YOLO architecture. However, the model turned out to be vulnerable to adversarial attacks and does not generalize to unseen generators. Failure to generalize to detect images produced by a newer generator also occurs for recent state-of-the-art methods, which we tested on Realistic Vision, a fine-tuned version of StabilityAI's Stable Diffusion image generator. |
| title | Detection of Synthetic Face Images: Accuracy, Robustness, Generalization |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2406.17547 |