This is a part of the DeepWaifu project.
This repository holds the AI model that converts selfies into waifus.
Credit for the model goes to https://github.com/taki0112/UGATIT and https://github.com/t04glovern/UGATIT
The live version of the dapp can be found here.
This project consists of the following repositories:
- https://github.com/chronologic/deep-waifu-ui
- https://github.com/chronologic/deep-waifu-server
- https://github.com/chronologic/deep-waifu-payment-contract
- https://github.com/chronologic/deep-waifu-model (this repository)
Run conda env create -f environment.yml
and then conda activate deep_waifu
(to add new dependencies, change environment.yml and then run conda env update -f environment.yml)
The easiest way is to use the pretrained model provided by https://github.com/taki0112/UGATIT here.
There might be an issue with extracting the file, as mentioned here and here. 7-zip on Windows should be able to handle the file.
Once extracted, the model should be placed in the checkpoint directory.
Run python server.py to start the server
The server exposes a POST /selfie2anime endpoint that accepts Form Data with file containing an image and returns a png of a waifu.