SkillAgentSearch skills...

FaceLabeling

Multi-Objective Convolutional Network for Face Labeling

Install / Use

/learn @Liusifei/FaceLabeling
About this skill

Quality Score

0/100

Supported Platforms

Universal

README

Please check https://github.com/Liusifei/Face_Parsing_2016 for latest HELEN (11 classes) model.

Since we change the matlab interface, you need to adapt the Makefile.config to your own machine.

Compile caffe and matlab respectively using "make all -j8" and "make matcaffe" at FaceLabeling/;

Run Face3_demo.m (cd FaceLabeling/matlab/caffe/lfw_cvpr15) to obtain the 3-classes parsing result.

If you want to test other facial images, please include a shape file (e.g. shape.txt) that contains the face's five points positions. There are many existing tools to generate the 5pts, e.g. http://mmlab.ie.cuhk.edu.hk/archive/CNN_FacePoint.htm

The Helen model will be released soon.

Please cite this work in your publications if it helps your research:

@inproceedings{liu2015multi,
  Author = {Sifei Liu and Jimei Yang and Chang Huang and Ming-Hsuan Yang},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  Title = {Multi-Objective Convolutional Learning for Face Labeling},
  Year = {2015}
}
View on GitHub
GitHub Stars56
CategoryDevelopment
Updated4y ago
Forks15

Languages

C++

Security Score

65/100

Audited on Jul 21, 2021

No findings