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.
Please before you open a new issue check [closed issues](https://github.com/gguilt/DeepCreamPy/issues?q=is%3Aissue+is%3Aclosed) and check the [table of contents](https://github.com/gguilt/DeepCreamPy#table-of-contents).
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.
If you want to make a pull request to DeepCreamPy, you must first sign our [Contributor License Agreement](https://github.com/deeppomf/contributing/blob/master/sign-cla.md#sign-the-cla) (the "CLA"). Then I can accept your pull requests.
Example mermaid image by Shurajo & AVALANCHE Game Studio under [CC BY 3.0 License](https://creativecommons.org/licenses/by/3.0/). The example image is modified from the original, which can be found [here](https://opengameart.org/content/mermaid).
Neural network code is modified from Forty-lock's project [PEPSI](https://github.com/Forty-lock/PEPSI), which is the official implementation of the paper [PEPSI : Fast Image Inpainting With Parallel Decoding Network](http://openaccess.thecvf.com/content_CVPR_2019/html/Sagong_PEPSI__Fast_Image_Inpainting_With_Parallel_Decoding_Network_CVPR_2019_paper.html). [PEPSI](https://github.com/Forty-lock/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](https://www.gwern.net/Danbooru2017) and other sources.
