Read the statement by Michael Teeuw here.
Facial Recognition - customize your mirror for every user!
-
@eliddell you must be twins :D
-
Actually I found a problem in the config.py
If I comment out the line i= min(0, i-1) and return i-1 in the user_label method it works… Now just want to make sure we don’t get an index out of bounds…
Not terribley familiar with python. So what does that min(int, int) do exactly
-
Hi, I seem to be having an issue when trying to implement the face recognition tester. When I run the tester on the file I am getting an error, I cannot seem to resolve it.
This is what displays in terminal:
Loading training data... ALGORITHM: LBPH Training data loaded! Picam selected... Traceback (most recent call last): File "facerecognition.py", line 61, in label, confidence = model.predict(crop) TypeError: 'int' object is not iterableMy picamera stays active but I can see that it has stopped on my face in the background, Do you have any idea how to resolve this issue, as I am at a loss here.
-
Hello
This module looks freaking amazing!! I want to give this a try, but I have a few questions that I haven’t been able to find answers for.
-
Do you recommend I used the NoIR camera, or just the regular Pi Camera? The mirror will be in my front hall by the doorway, under what should be adequate light whenever someone will want to use it. No expectation that this should work in the dark
-
Where is the best place to mount the camera? My mirror will be based on a 32" TV mounted such that the top of the frame would be slightly higher than my eye level (I’m 188 cm tall), and my wife’s eye level would be at about the halfway point of the mirror. I also have two young boys, aged 8 and 6 and I’d want the camera to recognize them as well. I’m thinking of either mounting the camera on the frame at the top of the mirror, aimed slightly downward, or on one side of the frame at about the vertical center of the mirror. I’ve also read about people mounting the camera behind the mirror, but with mixed results. If that is an option, I could theoretically mount the camera right in the center of the mirror. What’s the best way to route the cable in this case? I don’t plan on having anything in the middle of the screen from the top down to about the center, so I could run the cable from the camera straight up and over the top of the screen. Suppose I would need to colour the cable black or purchase a black cable. Anyone have insight here?
-
Given that the Pi Camera has a relatively narrow field of view (approx 35 mm) would you suggest I use one of those phone camera wide angle attachments so that the cam can pick up a wider image and have a better chance of getting the face in the image, or do I just increase the likelihood that several faces will be in the picture, defeating the facial recognition routine?
Thanks!
-Jeff -
-
Hey! I managed to ‘complete the training’ and copied
training.xmlinto the modules folder, it recognized my face during testing and everything, however, I get this error when doingnpm startand I have no clue what it means:Starting MagicMirror: v2.1.1 Loading config ... Loading module helpers ... Initializing new module helper ... Module helper loaded: MMM-Facial-Recognition No helper found for module: clock. No helper found for module: compliments. All module helpers loaded. Starting server on port 8080 ... Starting server op port 8080 ... Server started ... Connecting socket for: MMM-Facial-Recognition Staring module helper: MMM-Facial-Recognition Sockets connected & modules started ... Launching application. [MMM-Facial-Recognition] Facerecognition started... [MMM-Facial-Recognition] Loading training data... [MMM-Facial-Recognition] ALGORITHM: LBPH [MMM-Facial-Recognition] Training data loaded! [MMM-Facial-Recognition] -------------------- [MMM-Facial-Recognition] Webcam ausgewählt... Whoops! There was an uncaught exception... { Error: TypeError: Required argument 'device' (pos 1) not found at PythonShell.parseError (/home/pi/MagicMirror/modules/MMM-Facial-Recognition/node_modules/python-shell/index.js:183:17) at terminateIfNeeded (/home/pi/MagicMirror/modules/MMM-Facial-Recognition/node_modules/python-shell/index.js:98:28) at ChildProcess. (/home/pi/MagicMirror/modules/MMM-Facial-Recognition/node_modules/python-shell/index.js:88:9) at emitTwo (events.js:106:13) at ChildProcess.emit (events.js:191:7) at Process.ChildProcess._handle.onexit (internal/child_process.js:215:12) ----- Python Traceback ----- File "modules/MMM-Facial-Recognition/facerecognition/facerecognition.py", line 63, in camera = config.get_camera() File "/home/pi/MagicMirror/modules/MMM-Facial-Recognition/facerecognition/config.py", line 57, in get_camera return webcam.OpenCVCapture(device_id=0) File "/home/pi/MagicMirror/modules/MMM-Facial-Recognition/facerecognition/webcam.py", line 31, in __init__ self._camera.open() traceback: 'Traceback (most recent call last):\n File "modules/MMM-Facial-Recognition/facerecognition/facerecognition.py", line 63, in \n camera = config.get_camera()\n File "/home/pi/MagicMirror/modules/MMM-Facial-Recognition/facerecognition/config.py", line 57, in get_camera\n return webcam.OpenCVCapture(device_id=0)\n File "/home/pi/MagicMirror/modules/MMM-Facial-Recognition/facerecognition/webcam.py", line 31, in __init__\n self._camera.open()\nTypeError: Required argument \'device\' (pos 1) not found\n', executable: 'python', options: null, script: 'modules/MMM-Facial-Recognition/facerecognition/facerecognition.py', args: [ '{"recognitionAlgorithm":1,"lbphThreshold":50,"fisherThreshold":250,"eigenThreshold":3000,"useUSBCam":true,"trainingFile":"modules/MMM-Facial-Recognition/training.xml","interval":2,"logoutDelay":15,"users":["Alex"],"defaultClass":"default","everyoneClass":"everyone","welcomeMessage":true}' ], exitCode: 1 } MagicMirror will not quit, but it might be a good idea to check why this happened. Maybe no internet connection? If you think this really is an issue, please open an issue on GitHub: https://github.com/MichMich/MagicMirror/issues -
@j.e.f.f - I played around with this module months ago. I wasn’t able to get it working satisfactorily for my purposes, but I think what I learned can help.
I don’t think the NOIR camera is necessary for your intended setup. Especially if you aren’t going to mount the camera behind the mirror itself. I do not recommend mounting the camera behind the mirror glass. I tried it and without a strong light source, I think you will just be frustrated in your results.
Given that you’re using a 32" TV, I’m going to assume you’re locked into a portrait orientation. With little kids in the house, I suspect mounting the camera either off to one side or on the top of the frame is your best bet.
The magic mirror concept works best when you’re fairly close to it. The best effect after all is that the mirror shows you information while you’re looking at your own reflection. This is helpful as it encourages people to come close to the mirror to marvel at it, which aids in the facial recognition. The camera has a decent field of view already. Sufficient to get a good look at a person standing in front of the mirror from arms-length to about two meters away. Beyond that, the facial recognition starts to get spotty because there isn’t enough detail in your face, but most people won’t be looking at the mirror from two meters away, right?
I would skip the wide-angle attachments. I don’t think it’s going to be necessary.
-
@bhepler Thanks for your insight! This helps a lot.
-
Hey
When i run npm start in “/home/pi/MagicMirror” i am struck at below message:
pi@raspberrypi:~/MagicMirror $ npm start > magicmirror@2.1.1 start /home/pi/MagicMirror > sh run-start.sh Starting MagicMirror: v2.1.1 Loading config ... Loading module helpers ... Initializing new module helper ... Module helper loaded: MMM-Facial-Recognition No helper found for module: alert. Initializing new module helper ... Module helper loaded: updatenotification No helper found for module: clock. No helper found for module: compliments. No helper found for module: currentweather. No helper found for module: weatherforecast. Initializing new module helper ... Module helper loaded: newsfeed All module helpers loaded. Starting server on port 8080 ... Starting server op port 8080 ... Server started ... Connecting socket for: MMM-Facial-Recognition Staring module helper: MMM-Facial-Recognition Connecting socket for: updatenotification Connecting socket for: newsfeed Starting module: newsfeed Sockets connected & modules started ... Launching application. Create new news fetcher for url: http://www.20min.ch/rss/rss.tmpl?type=channel&get=4 - Interval: 300000 [MMM-Facial-Recognition] Facerecognition started... [MMM-Facial-Recognition] Loading training data... [MMM-Facial-Recognition] ALGORITHM: LBPH [MMM-Facial-Recognition] Training data loaded! [MMM-Facial-Recognition] -------------------- [MMM-Facial-Recognition] PiCam ausgewählt.../* Magic Mirror Config Sample * * By Michael Teeuw http://michaelteeuw.nl * MIT Licensed. */ var config = { port: 8080, ipWhitelist: ["127.0.0.1", "::ffff:127.0.0.1", "::1"], // Set [] to allow all IP addresses. language: "en", timeFormat: 24, units: "metric", modules: [ { module: "MMM-Facial-Recognition", classes: "default everyone", config: { recognitionAlgorithm: 1, lbphThreshold: 50, fisherThreshold: 250, eigenThreshold: 3000, useUSBCam: false, trainingFile: "modules/MMM-Facial-Recognition/training.xml", interval: 2, logoutDelay: 15, users: ["Stefan"], defaultClass: "default", everyoneClass: "everyone", welcomeMessage: true } }, ] }; /*************** DO NOT EDIT THE LINE BELOW ***************/ if (typeof module !== "undefined") {module.exports = config;}python facerecongnition.py also not work:
pi@raspberrypi:~/MagicMirror/modules/MMM-Facial-Recognition-Tools $ python facerecognition.py ['Stefan'] Using FACE_ALGORITM: 1 Loading training data... Training data loaded! Strong:Stefan No windowing system, writing face.jpg image Traceback (most recent call last): File "facerecognition.py", line 153, in clean_shutdown() TypeError: clean_shutdown() takes exactly 2 arguments (0 given) -
j.e.f.f Project Sponsor Module Developerlast edited by j.e.f.f May 9, 2017, 1:44 PM May 8, 2017, 3:57 PM
So I’ve managed to get this working with a Logitech C170 USB webcam, and I’m pretty stoked!
I have a few questions:
-
Facial recognition is generally pretty slow… I need to stand in front of the camera for about 10 seconds before I am recognized. Is this normal or a bi-product of the fact that this is a fairly low quality camera… e.g.: if I were to upgrade say to the C920 instead, should I expect better performance? I’ve left the polling setting in config.js at the default 2 seconds – if the recognition were to happen this fast I’d be pretty happy!
-
There is a reference earlier in the thread that the capture resolution can be increased by modifying the module code. Where do I modify this? The c170 can capture natively at 640x480, so I’d like to experiment with capturing both the training images and the real-time facial recognition at this resolution, in an effort to see if I can get better confidence numbers. I’m currently seeing high 40s to mid 50s confidence. At the default threshold of 50, I am often not recognized, but it works more frequently if I increase the threshold to 60. (Training is based on about 50 images all captured from the c170 in representative lighting.) EDIT: one day later and I’m now often seeing confidence in the high 60’s so I needed to increase the threshold to 70.
-
I’d like to experiment with the other recognition algorithms, but I can only get LBPH to work. If I try to use Fisher or Eigen I get an error that the training data has not yet been computed. Training.xml was generated using the appropriate algorithm, and config.js was set to the matching algorithm.
-
With MMM-Facial-Recognition-Tools module, I’m having the same issue as the previous post where I get an error when trying to run tests using facerecognition.py (same error as above). So I need to do my tests with the actual MMM-Facial-Recognition module running on MagicMirror, which as you can imagine is fairly time consuming. Anyone have any insight on getting the test script working?
For context, I’m opting for a webcam instead of the picam because webcams generally look nicer. I’ve spent HOURS scouring the web for a picam enclosure that didn’t look like a high school engineering project, without finding a single one. The closest I could find is this one by Adafruit but it has two problems:
- I don’t like that the cover is clear – I don’t want to see the electronics
- I don’t like how it routes the ribbon cable
This first point is easily remedied with a can of spray paint, but the second is a little harder to deal with. If anyone has found a nice sleek enclosure for the picam, please share a link!
Secondly, I like that webcams generally also have a microphone built-in, so that I can also use it for voice control. I REALLY wish I could use this one but I contacted their support and they told me it wouldn’t work at all with the Raspberry Pi. Too bad… was hoping the 4-microphone array would really with not having to yell at my mirror to get it to hear me and I like how sleek it is. Would look quite unassuming sitting on top of my mirror.
Thanks for the help
-Jeff -
-
Also one more thought: I’m thinking of turning the camera 90 degrees so that it takes a vertical image – better chance of capturing faces of people of different heights that – thinking myself as a tall adult vs my kids, 9 and 6 years old. I’m thinking a 2:3 image (or 9:16 in the case of an HD cam) would really work well here to turn a wide FoV to a “tall” FoV – perfect for people standing right in front of the mirror… If I just turn the camera 90 degrees, and use it to train and capture in that orientation, will the facial recognition algorithms work as-is (i.e.: Can they recognize a face turned sideways) or would the module need to be changed to rotate the capture first before feeding into the algorithm?
