photoguard
timm-vis
photoguard | timm-vis | |
---|---|---|
7 | 1 | |
520 | 39 | |
0.0% | - | |
1.8 | 0.0 | |
about 1 year ago | almost 3 years ago | |
Jupyter Notebook | Jupyter Notebook | |
MIT License | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
photoguard
- PhotoGuard - сервіс для захисту зображень від нейромереж. Працює за допомогою моделей редагування фотографій на основі машинного навчання, таких як Stable Diffusion.
-
Are there any tools for "Defend Against the Dark Arts" of diffusion?
I've been searching for a tool to create Diffusion-resistant images, and I came across the Photoguard repository and tried it, the results weren't so good (very strange colorful noises on the image, which makes it unusable)
- Raising the Cost of Malicious AI-Powered Image Editing
- PhotoGuard: Defending Against Diffusion-Based Image Manipulation
-
Welcome to a community to discuss what to do about the negative effects of AI art
- PhotoGuard can help to protect images from being able to be edited by Stable Diffusion with inpainting, it can theoretically also help to protect artwork from being finetuned by a model: https://github.com/MadryLab/photoguard
- Somebody implemented a photoguard against inpainting.
timm-vis
-
[P] - timm-vis: Visualizer for PyTorch image models
Hello, thanks for bringing these points up. Currently the methods work only with inputs with 3 channels. I have not implemented grad-cam yet. The visualization method closest to grad-cam would be a saliency map. A saliency map shows the influence of each pixel with respect to the model outputs. It calculates gradients of the input image unlike grad-cam, which computes the gradients of the last activation layer. I plan to add 1 channel input support and grad-cam support in the next few days. I encourage you to take a look at the the existing methods in the python notebook to see if anything interests you meanwhile.
What are some alternatives?
web-stable-diffusion - Bringing stable diffusion models to web browsers. Everything runs inside the browser with no server support.
advertorch - A Toolbox for Adversarial Robustness Research
stable-diffusion-webui-colab - stable diffusion webui colab
neural-tangents - Fast and Easy Infinite Neural Networks in Python
fastai - The fastai deep learning library
nn - 🧑🏫 60 Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gans(cyclegan, stylegan2, ...), 🎮 reinforcement learning (ppo, dqn), capsnet, distillation, ... 🧠
Made-With-ML - Learn how to design, develop, deploy and iterate on production-grade ML applications.