PSGAN can be used to apply makeup to a face image.
[Instruction]
The main process is:
Prepare training images -> training -> apply makeup on images
For the source of the training dataset, please see the Source Code URL in readme.txt.
1. 1_train.ipynb
Start training the images in the data/train folder.
2. 2_inference.ipynb
Make up the image.
The sample image for makeup needs to be placed in the data/test/makeup folder. If you want to replace the unmakeup image, please put it in the data/test/non-makeup folder and modify the file name of the source_path parameter (change xfsy_0106.png) .
Parameter Description:
--device: cpu or cuda. cuda can add: to specify gpu, for example: cuda:0.
--model_path: Inferred model path.
--source_path: The path of the face image being makeup.
--reference_dir: The path of the sample image folder for makeup.
Note:
For the test image without makeup, please try to have only the face area like a headshot, and do not include too much background, so as not to reduce the makeup effect.
This SDK is built in AppForAI - AI Dev Tools.
Purchase license separately: USD 600, permanent authorization, single APP authorization, single machine authorization, one-year activation, one-year download, one-year update, one-year email technical support.