90 Lines of code to convert your face movement into keyboard commands.
This is a basic face movement tracking that can convert face movement into keyboard commands like UP - DOWN - LEFT - RIGHT. I used facial landmarks to detect face and get the nose out of it for better referencing. I have created two versions of it, v1 is using a fixed reference boundary which not work expected properly because we need to come at the same position after each movement. To save this I created V2 which uses position change with respect to the previous position. This is more dynamic and easy to control the moves. No need to set position again and again.
In version1, I used a fixed reference boundary. If nose reference is out of boundary then I calculate the direction of movement. After getting direction I am converting it into keyboard commands using the keyboard library.
In version2, I am using reference change with respect to the previous position in a particular time window and then calculating the direction vector to get direction and converting it to keyboard command.
This is the list of dependencies for running this application. Use pip to install them.
- opencv
- keyboard
$ pip install -r requirements.txt
-
Download or clone this repository.
-
Extract to some location.
-
First, run
movement-v1.py(for fix boundary) or runmovement-v2.py(for dynamic movement)
NOTE: If you are getting 215 assertion failed!! on line 81 check this (surya-veer#4 (comment)) -
open any online atari game like Subway surfers or temple run.
-
Start doing movements to play game. It will press up-down-left-right based on your movements.
Open any online game on the browser which needs UP-DOWN-LEFT-RIGHT movements following games, you can find many games if you search on google.
- Subway surfer https://www.kiloo.com/subway-surfers/
- Temple run https://m.plonga.com/adventure/Temple-Run-2-Online-Tablet