This is the readme for the robot ticket checker project made by Yi Zhang, Yinglei Song, Tao Ma, Chongyu Zhang. 

/*******************************************************************************************/
Dependencies
/*******************************************************************************************/
1.numpy
2.face_recognition

How to install numpy:
$ pip install numpy

How to install face_recognition:
$ pip install face-recognition

/*******************************************************************************************/
Packages of the whole task
/*******************************************************************************************/
Packages:

control
face

/*******************************************************************************************/
Network setup
/*******************************************************************************************/
You may need a separate laptop to run face recognition part.
It is important to setup your ROS IP and ROS Master.
For master(usually desktop machine), open .bashrc, and add:
export ROS_IP=localhost
export ROS_MASTER_URI=http://localhost:11311

For slave(usually laptop), open .bashrc, and add:
export ROS_IP=localhost
export ROS_MASTER_URI=http://IP_of_Master:11311

localhost is IP address of own machine.
For slave machine, you need to use the IP address of master in ROS_MASTER_URI.

Detailed reference: http://wiki.ros.org/ROS/NetworkSetup

/*******************************************************************************************/
How to run the code
/*******************************************************************************************/
On desktop machine
1. open one ternimal, run: $ cd project
2. run: $ catkin build
3. run: $ source devel/setup.bash
4. run: $ roslaunch control start.launch 

On laptop
1. open one ternimal, run: $ cd project
2. run: $ catkin build
3. run: $ source devel/setup.bash
4. run: $ rosrun face face_detect.py

/*******************************************************************************************/
Links to demo videos
/*******************************************************************************************/
Full Demo 1
https://youtu.be/rvDgHgA3xF8
Full Demo 2
https://youtu.be/iJaAUak_v8Y
Invalid cases
https://youtu.be/gdtXTLhnkf8

In case 1, Mike used Jack's ticket and said he was Jack, but face recognition found the problem.

In case 2, Mike used Jack's ticket and said he was Mike. The name on the ticket and the name from the speech did not match.

In case 3, everything was valid.