| .github/ISSUE_TEMPLATE | ||
| decensor_input | ||
| decensor_input_original | ||
| decensor_output | ||
| docs | ||
| models | ||
| readme_images | ||
| README.md | ||
DeepCreamPy
Decensoring Hentai with Deep Neural Networks.
DeepCreamPyV2--a major upgrade over DeepCreamPyV1--is under construction.
Please bear with me. Many, many things will be broken.*
All available binaries are outdated. Wait for the next release.
A deep learning-based tool to automatically replace censored artwork in hentai with plausible reconstructions.
The user colors censored regions green in an image editing program like GIMP or Photoshop. A neural network fills in the censored regions.
DeepCreamPy has a pre-built binary for Windows 64-bit available here. DeepCreamPy's code works on Windows, Mac, and Linux.
Please before you open a new issue check closed issues and check the table of contents.
Features
- Decensoring images of ANY size
- Decensoring of ANY shaped censor (e.g. black lines, pink hearts, etc.)
- Higher quality decensors
- Support for mosaic decensors
Limitations
The decensorship is for color hentai images that have minor to moderate censorship of the penis or vagina. If a vagina or penis is completely censored out, decensoring will be ineffective.
It does NOT work with:
- Black and white/Monochrome image
- Hentai with screentones (e.g. printed hentai)
- Real life porn
- Censorship of nipples
- Censorship of anus
- Animated gifs/videos
Table of Contents
Setup:
Usage:
Miscellaneous:
To do
- Resolve all Tensorflow compatibility problems
- Finish the user interface
- Add support for black and white images
- Add error log
Follow me on Twitter @deeppomf (NSFW Tweets) for project updates.
Contributions are welcome! Special thanks to ccppoo, IAmTheRedSpy, 0xb8, deniszh, Smethan, mrmajik45, harjitmoe, itsVale, StartleStars, and SoftArmpit!
License
Acknowledgements
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.
Donations
If you like the work I do, you can donate to me via Paypal. The funds will mainly go towards purchasing better GPUs to accelerate training. 