Facial Recognition - customize your mirror for every user!

[rvdgeer]

Very nice module.
However I keep ruinning into the same error when I try capture.py to convert them to pgm files:
Could not detect single face!

I’ve taken ‘selfies’ with different camera’s in different lighting conditions, but not even one is good enough to convert into a pgm file…
Can you please give me some clues on what conditions my pictures have to meet?
Am I doing something wrong?
All help and advice appreciated!

[rvdgeer]

@rajbadri Have you added the classes: (with a value) to all the other modules in the config.js? If you get a plain black screen without errors it might be that no modules were loaded.

[rvdgeer]

@rvdgeer OK, replying to my own post for future reference:

I got it working by doing the following:

• cropped the pictures to only show my face;
• copied the cropped pictures and transformed them to greyscale;
• copied the grey pictures and upped the contact;
• moved all pictures (originals and copies) to a direct sub folder of the facetrainer folder;
• started the capture.py, pressed 2, entered my name, entered the name of the sub folder (only the sub folder, not the entire path) and waited.

This way I got it to recognize about 10 of the pictures…
Not bad!

Getting it to recognize me is still not easy, but that because of the lighting conditions in my ManCave where I’m putting it together.

[shashank]

@paviro Hi,

I am getting this error while running facerecognition.py

and also need to know in //Facerecognition Test Tool Readme
In tools/facerecognition_tester/lib/config.py add add your training labels from tools/facetrainer under personen = [] ( what need to enter in this personen= [])

[rvdgeer]

@shashank If you run the facetrainer (capture.py and train.py) first, you should know what to enter here…

[shashank]

@rvdgeer hi,

After running capture.py a folder created by name ‘sasi’ in the /home/pi/MagicMirror/modules/MMM-Facial-Recognition-Tools/facetrainer/training_data

then by running train.py a training.xml created in the path
/home/pi/MagicMirror/modules/MMM-Facial-Recognition-Tools/facetrainer

and also in /home/pi/MagicMirror/modules/MMM-Facial-Recognition-Tools/facerecognition_tester/lib i mentioned name in personen = ["sasi"]

my config.js file as below

{
    module: 'MMM-Facial-Recognition',
    position: 'top_left',
    config: {
        // 1=LBPH | 2=Fisher | 3=Eigen
        recognitionAlgorithm: 1,
        // Threshold for the confidence of a recognized face before it's considered a
        // positive match.  Confidence values below this threshold will be considered
        // a positive match because the lower the confidence value, or distance, the
        // more confident the algorithm is that the face was correctly detected.
        lbphThreshold: 50,
        fisherThreshold: 250,
        eigenThreshold: 3000,
        // force the use of a usb webcam on raspberry pi (on other platforms this is always true automatically)
        useUSBCam: false,
        // Path to your training xml
        trainingFile: 'modules/MMM-Facial-Recognition/training.xml',
        // recognition intervall in seconds (smaller number = faster but CPU intens!)
        interval: 2,
        // Logout delay after last recognition so that a user does not get instantly logged out if he turns away from the mirror for a few seconds
        logoutDelay: 15,
        // Array with usernames (copy and paste from training script)
        users: ['sasi'],
        //Module set used for strangers and if no user is detected
        defaultClass: "default",
        //Set of modules which should be shown for every user
        forAllClass: "for_all"
    }
},

{
    module: 'clock',
    position: 'top_left',
    //Set your classes here seperated by a space.
    //Shown for all users
    classes: 'for_all'
},

when i run pm2 start mm i am struck at below message

can someone help me on this error

[zachlan]

Potentially stupid question here, apologies in advance if I’ve missed something obvious. I have the mirror and the face recognition working (Yay!) but when it’s running, on the left side of the screen it shows “MMM-Facial-Recognition” and module-10-mmm-facial-recognition underneath. I though about removing the position tag, but that’s there in the readme version so did not. How do I remove the text showing the running facial recognition module?
Thanks!Zach

[paviro]

@rvdgeer that sucks sorry about that. Recognition can be tricky sometimes :/

@shashank check the trainingFile in your config, does it really exist at that location?

Update: I see you figured it out, was a wrong location the problem? Just remove the position part is isn’t relevant anymore sorry :)

[paviro]

This post is deleted!

[shashank]

@paviro

yes , i copied the training file from //MMM-Facial-Recognition-Tools/facetrainer

to // Path to your training xml
trainingFile: ‘modules/MMM-Facial-Recognition/training.xml’,