dualFace

dualFace: Two-Stage Drawing Guidance for Freehand Portrait Sketching (CVMJ)

We provide python implementations for our CVM 2021 paper "dualFace:Two-Stage Drawing Guidance for Freehand Portrait Sketching". This project provide sketch support for artistic portrait drawings with a two-stage framework. [arXiv][PDF][Project][Video]

User Interface

Prerequisites

Window
Conda (Python 3.6)
CPU or NVIDIA GPU + CUDA CuDNN

Getting Started

Installation

Install PyTorch 1.3.1 and torchvision 0.4.1 from http://pytorch.org and other dependencies (e.g., visdom and dominate). You can install all the dependencies by

bat
call conda remove -n py36df
call conda create -n py36df python=3.6 
call conda activate py36df
call conda install pytorch==1.3.1 -c pytorch
pip install cmake
pip install -r requirements.txt

Quick Start (Apply a Pre-trained Model)

Download a pre-trained model from (https://drive.google.com/open?id=1cQx9hPOJ18sU5HPGkbRTJ-e6cYqqHUHh)

cd sse
sse.exe "-i index_file -v vocabulary -f filelist -n 8"
call conda activate py36df
python demo.py

Acknowledgments

Our code has depended on the following opensource codes.

MaskGAN(https://github.com/switchablenorms/CelebAMask-HQ)
faceParsing(https://github.com/zllrunning/face-parsing.PyTorch)
APDrawingGAN(https://github.com/yiranran/APDrawingGAN)
OpenSSE(https://github.com/zddhub/opensse)

Please contact xie@jaist.ac.jp for any comments or requests.

Citation

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

@misc{huang21dualface,
  title     = {dualFace: Two-Stage Drawing Guidance for Freehand Portrait Sketching},
  author    = {Zhengyu Huang and Yichen Peng and Tomohiro Hibino and Chunqi Zhao and Haoran Xie and Tsukasa Fukusato and Kazunori Miyata},
  year      = {2021},
  eprint    = {2104.12297},
  archivePrefix = {arXiv},
  primaryClass  ={cs.GR}
}

tamwaiban/dualFace