Hundreds of features to explore
Games made in Construct
Your questions answered
Popular & trusted by schools and Universities world-wide
Construct 3 runs in the browser & works offline
Students do not need accounts with us
Our educational partners
Free education resources to use in the classroom
World class complete documentation
Official and community submitted guides
Learn and share with other game developers
Upload and play games from the Construct community
Game development stories & opinions
Integrates face-api.js for face detection, landmarks.
Example using webcam, works in preview, included weights in project files.
Add face points to video also.
An updated version added 18.104.22.168, works in preview, see the example for changes required (weights now included in project files.)
Hi Mikal, this updated version is not working in mobile device now at all. previous one was though slow but working.
Do you have any console logs, so we can see the error? Is this when you build an apk? Does it work if you try remote preview?
I tried creating this link but dots are not mapped to the face.
Can you please help me what is the problem? i'm using ur example file and changed only layout size 640x960 instead of 1280x720.
I think this comment will be useful for all of us. For example you can define different dots just for the eye or for the mouth. Let's do the creation like dot 1 for the eyes, let's say dot 2 for the nose.
Hello and congratulations. This is very useful to me. but I couldn't run it. I want to move the points that I have set to the face, face and eye points on a photo. Is it possible?
It works for me, but you need to use the 22.214.171.124 beta version of the addon (listed above) and use the faceapiexample02.c3p example (listed above.) Try it in preview. A photo may be possible, but you would need to figure out how to get a uri that points to your photo and replace 'UserMedia.SnapshotURL' with your uri.
Hi! Thank you for your answer. I can run it now. But from export html5 mod and I run the browser 'chrome' I get the this error code "reference error: g_aAe not defined" thanks for reply.
any idea how to store value of any point in a global variable?
C3 changed one of the APIs to load project files, I think this may be why it's not working. I will update the plugin for this.
This plugin is no longer working with latest update of c3. Earlier it was working fine. Now no dots are drawn by plugin so face values can't be extracted now.
Hi The plugin looks good. Really appreciate the efforts but it not looking useful, bcz we can't do anything with it. It takes screenshot continuously and place it above another. So if we want to detect mouth is open or eyes are blinked or not. That won't be possible with this plugin.It would have been great if dots can move along with detected face area in real time instead of screenshot.
Thanks for the comments. The model in this example does not run at real time speed on most systems, it takes a lot of compute. I am testing a new version which may run faster. You can save the points captured and compare past points vs current points to do a comparison.
Thanks, Mikal. for the update. Will wait for newer version eagerly. Will try to be the first one to test this as I am looking for face detection especially blink and smile from a long time.
I tried to run, but failed) I will wait for stable releases!
Just checking for more info, in case I can fix it. Did you try the example project in preview? It will not work in preview (due to how the model files are loaded.) It has only be tested on export and hosted on a local (or web based) webserver.
Yes, I saw a warning :)
I was deployed html5 build to NOW by ZEIT and got an error: "ReferenceError: Can't find variable: g_auk". Then I got access to the camera and clicked FaceDetect but nothing happened