We provide PyTorch implementations for our TPAMI paper "Quality Metric Guided Portrait Line Drawing Generation from Unpaired Training Data". paper

Our method can (1) learn to generate high quality portrait drawings in multiple styles using a single network and (2) generate portrait drawings in a “new style” unseen in the training data.

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

• To install the dependencies, run

pip install -r requirements.txt

• 1. Download pre-trained models from BaiduYun(extract code:g8is) or GoogleDrive and rename the folder to checkpoints/.
• 2. Test for example photos: generate artistic portrait drawings for example photos in the folder ./examples using

python test_seq_style3.py

The test results will be saved to html files here: ./results/QMUPD_model/test_200/indexstyle*.html. The result images are saved in ./results/QMUPD_model/test_200/imagesstyle*, where real, fake, correspond to input face photo, synthesized drawing of a certain style, respectively.

You can contact email ranyi@sjtu.edu.cn for any questions.

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

@article{YiLLR22,
  title     = {Quality Metric Guided Portrait Line Drawing Generation from Unpaired Training Data},
  author    = {Yi, Ran and Liu, Yong-Jin and Lai, Yu-Kun and Rosin, Paul L},
  journal   = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year      = {DOI (identifier)  10.1109/TPAMI.2022.3147570, 2022},
}

Our code is inspired by pytorch-CycleGAN-and-pix2pix.