Table of Contents
In this project, we tackle the problem of generating color photorealistic images of human faces from corresponding hand-drawn sketches. We aggregate and align datasets CUHK and FS2K of facial sketches and corresponding real facial photos for training and evaluation. For our baseline we tried to follow 4 papers, sketch2face [Julia Gong et al.], pix2pix [Philip Isola et al.], and Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks [Jun-Yan Zhu, Alexei A. Efros et al.]
For more details, please see Report
- PyTorch
conda install -c pytorch pytorch
- NumPy
conda install numpy
- OpenCV
conda install -c conda-forge opencv
- Clone the repo
git clone https://github.com/kanthprashant/Facial-Sketch-to-Colored-Images.git
Distributed under the MIT License. See LICENSE.txt
for more information.
- Philip Isola, Jun-Yan Zhu, Tinghui Zhou, Alexei A. Efros: Image-to-Image Translation with Conditional Adversarial Networks
- Julia Gong, Matthew Mistele: sketch2face: Conditional Generative Adversarial Networks for Transforming Face Sketches into Photorealistic Images.
- Shu-Yu Chen, Wanchao Su, Lin Gao, Shihong Xia, Hongbo Fu: DeepFaceDrawing: Deep Generation of Face Images from Sketches.
- Jun-Yan Zhu, Taesung Park, Philip Isola, Alexei A. Efros: Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks.
- Udemy: Computer Vision A-Z, Practical Deep Learning with Pytorch