Decensoring Hentai with Deep Neural Networks.
A deep learning-based tool to automatically replace censored artwork in hentai with plausible reconstructions.
Before DeepCreamPy can be used, the user must color censored regions in their hentai green with an image editing program (e.g. GIMP, Photoshop). DeepCreamPy takes the green colored images as input, and a neural network automatically fills in the censored regions.
You can download the latest release for Windows 64-bit here.
For users interested in compiling DeepCreamPy themselves, DeepCreamPy can run on Windows, Mac, and Linux.
Please before you open a new issue check closed issues and check the table of contents.
- Decensoring images of any size
- Decensoring of ANY shaped censor (e.g. black lines, pink hearts, etc.)
- Decensoring of mosaic decensors
- Limited support for decensoring black and white/monochrome images
- Generate multiple variations of decensors from the same image
The decensorship is for color hentai images that have minor to moderate censorship of the human reproductive organs. If an organ is completely censored out, decensoring will be ineffective.
It does NOT work with:
- Hentai with screentones (e.g. printed hentai)
- Real life pornographic material
- Censorship of nipples
- Censorship of lower orifice of the alimentary canal
- Animated gifs/videos
Setup:
Usage:
Miscellaneous:
- Resolve all Tensorflow compatibility problems
- Finish the user interface
- Add error log
If you want to make a pull request to DeepCreamPy, you must first sign our Contributor License Agreement (the "CLA"). Then I can accept your pull requests.
Special thanks to ccppoo, IAmTheRedSpy, 0xb8, deniszh, Smethan, harjitmoe, itsVale, StartleStars, and SoftArmpit for their contributions!
Source code and official releases/binaries are distributed under the GNU Affero General Public License v3.0.
Example mermaid image by Shurajo & AVALANCHE Game Studio under CC BY 3.0 License. The example image is modified from the original, which can be found here.
Neural network code is modified from Forty-lock's project PEPSI, which is the official implementation of the paper PEPSI : Fast Image Inpainting With Parallel Decoding Network. PEPSI is licensed under the MIT license.
Training data is modified from gwern's project Danbooru2017: A Large-Scale Crowdsourced and Tagged Anime Illustration Dataset and other sources.
See ACKNOWLEDGEMENTS.md for full license text of these projects.