APDrawingGAN++ Jittor Implementation

We provide Jittor implementations for our TPAMI 2020 paper "Line Drawings for Face Portraits from Photos using Global and Local Structure based GANs". [Paper]

It is a journal extension of our previous CVPR 2019 work APDrawingGAN.

This project generates artistic portrait drawings from face photos using a GAN-based model.

PyTorch implementation

Prerequisites

• Linux or macOS
• Python 3
• CPU or NVIDIA GPU + CUDA CuDNN

Sample Results

Up: input, Down: output

Installation

• To install the dependencies, run

pip install -r requirements.txt

Apply pretrained model

• 1. Download pre-trained models from BaiduYun(extract code: 9w83) and rename the folder to checkpoints.
• 2. Test for example photos: generate artistic portrait drawings for example photos in the folder ./samples/A_img/example using models in checkpoints/apdrawinggan++_author

python test.py

Results are saved in ./results/portrait_drawing/apdrawinggan++_author_150/example

• 3. To test on your own photos: First run preprocess here). Then specify the folder that contains test photos using option --input_folder, specify the folder of landmarks using --lm_folder, the folder of foreground masks using --mask_folder, and the folder of compact masks using --cmask_folder, and run the test.py again.

Train

• 1. Download the APDrawing dataset (augmented using histogram matching) from BaiduYun(extract code: sq62) and put the folder to data/apdrawing++.
• 2. Train our model (150 epochs)

python apdrawing_gan++.py

Models are saved in folder checkpoints/apdrawing++

• 3. Test the trained model

python test.py --which_epoch 150 --model_name apdrawing++

Results are saved in ./results/portrait_drawing/apdrawing++_150/example

Citation

If you use this code or APDrawing dataset for your research, please cite our paper.

@inproceedings{YiXLLR20,
  title     = {Line Drawings for Face Portraits from Photos using Global and Local Structure based {GAN}s},
  author    = {Yi, Ran and Xia, Mengfei and Liu, Yong-Jin and Lai, Yu-Kun and Rosin, Paul L},
  booktitle = {{IEEE} Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},
  doi       = {10.1109/TPAMI.2020.2987931},
  year      = {2020}
}