THIS PROJECT IS KIND OF EMOTION RECOGNITION. IN THIS PROJECT, WE FOLLOW ONE-MODEL-FOR-ONE-USER PRINCIPLE,THAT' S, RECOGNITION IS NOT GENERAL-PURPOSED. THAT IS WHAT THIS MAKES THE MAJOR DIFFERENCE FROM OTHERS.
Windows
- [Install Anaconda](https://www.anaconda.com/download/)
$ conda install python=3.5
$ conda install -c menpo opencv3
$ conda install -c conda-forge dlib=19.4
$ python -m pip install neurolab imutils
Ubuntu
- Install dlib, opencv, boost
$ sudo apt-get install build-essential cmake libgtk-3-dev libboost-all-dev libopencv-dev
$ wget https://bootstrap.pypa.io/get-pip.py
$ sudo python get-pip.py
$ sudo pip install numpy scipy scikit-image dlib
- Install neurolab
$ sudo pip install neurolab
OSX
- Install homebrew
$ /usr/bin/ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"
$ brew update
$ brew tap homebrew/science
$ sudo nano ~/.bash_profile
# Homebrew
export PATH=/usr/local/bin:$PATH
$ source ~/.bash_profile
- [Install Anaconda & Downgrade python](https://www.continuum.io/downloads/)
$ conda install python=3.5
- Install opencv
$ conda install numpy
$ conda config --add channels conda-forge
$ conda install opencv
- Install dlib
$ conda install dlib
- Install neurolab, imutils
$ sudo pip install neurolab imutils
-
Download Landmark.dat
$ ./download_data.sh -
Train
$python record.py -
Test
$python test.py
You can download necessary files such as models from the following urls.
THE CODE IS NOT ALLOWED FOR COMMERCIAL PURPOSE. PLEASE CONTACT THE AUTHOR FIRST BEFORE YOU USE IT COMMERCIALLY.