634 |
635 | This program is free software: you can redistribute it and/or modify
636 | it under the terms of the GNU Affero General Public License as published
637 | by the Free Software Foundation, either version 3 of the License, or
638 | (at your option) any later version.
639 |
640 | This program is distributed in the hope that it will be useful,
641 | but WITHOUT ANY WARRANTY; without even the implied warranty of
642 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
643 | GNU Affero General Public License for more details.
644 |
645 | You should have received a copy of the GNU Affero General Public License
646 | along with this program. If not, see .
647 |
648 | Also add information on how to contact you by electronic and paper mail.
649 |
650 | If your software can interact with users remotely through a computer
651 | network, you should also make sure that it provides a way for users to
652 | get its source. For example, if your program is a web application, its
653 | interface could display a "Source" link that leads users to an archive
654 | of the code. There are many ways you could offer source, and different
655 | solutions will be better for different programs; see section 13 for the
656 | specific requirements.
657 |
658 | You should also get your employer (if you work as a programmer) or school,
659 | if any, to sign a "copyright disclaimer" for the program, if necessary.
660 | For more information on this, and how to apply and follow the GNU AGPL, see
661 | .
662 |
--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 | # Diffusion models meet image counter-forensics
6 |
7 | [Link to code](https://www.github.com/mtailanian/diff-cf)
8 |
9 | ### Link to download paper:
10 |
11 | [](https://openaccess.thecvf.com/content/WACV2024/papers/Tailanian_Diffusion_Models_Meet_Image_Counter-Forensics_WACV_2024_paper.pdf)
12 | [](https://arxiv.org/abs/2311.13629)
13 |
14 |
15 |
16 | ### Citation:
17 | ```
18 | @InProceedings{Tailanian_2024_WACV,
19 | author = {Tailani\'an, Mat{\'\i}as and Gardella, Marina and Pardo, Alvaro and Mus\'e, Pablo},
20 | title = {Diffusion Models Meet Image Counter-Forensics},
21 | booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
22 | month = {January},
23 | year = {2024},
24 | pages = {3925-3935}
25 | }
26 | ```
27 | ## Abstract
28 | _From its acquisition in the camera sensors to its storage, different operations are performed to generate the final image. This pipeline imprints specific traces into the image to form a natural watermark. Tampering with an image disturbs these traces; these disruptions are clues that are used by most methods to detect and locate forgeries. In this article, we assess the capabilities of diffusion models to erase the traces left by forgers and, therefore, deceive forensics methods. Such an approach has been recently introduced for adversarial purification, achieving significant performance. We show that diffusion purification methods are well suited for counter-forensics tasks. Such approaches outperform already existing counter-forensics techniques both in deceiving forensics methods and in preserving the natural look of the purified images. The source code is publicly available at https://github.com/mtailanian/diff-cf._
29 |
30 |
31 |
32 |
109 |
110 |
113 |
114 |
121 |
122 |
125 |
126 |