Asked — Edited

Microsoft Emotion And Face Api Plugin Too?

@DJ Thanks so much for the other MS API plugins... Cognitive vision works pretty good... I was wondering (and not to get greedy or anything) but would it be possible to create an Emotion and Face plugin as well?.... These api's are fantastic and add so much extra vision functionality to ARC... But If not, again thanks for what you have done so far... :)


Upgrade to ARC Pro

Join the ARC Pro community and gain access to a wealth of resources and support, ensuring your robot's success.


Asking ruins the surprise, doesn't it? I'm going to delay the release now... Everyone can thank Richard :P


Wait? What? delete, delete, delete! You didn't hear me say anything... :P Nooooo eek


I love the plug-in's that are released on Microsoft cognitive services. (speech recognition, computer vision, emotion), puts EZ-B on a different maturity level and very powerful.

@DJ, reading above replies almost scared to ask, could you add one more: Microsoft Face (recognition) API. (Detect, identify, analyze, organize, and tag faces in photos)

Like speech recognition, the difference between local and cloud compute power makes a stepwise change in capability/reliability, need the same for face recognition.


@Charel, removed your other duplicate thread since you're posting in this one.

To identify specific faces, use the Object Training feature of the camera control in ARC. Have fun!

Here's a link to the manual of the Camera Device:

that page can also be viewed by pressing the ? (question mark) next to the X (close) button on the camera device control.


Thanks for removing the duplicate, was indeed searching on "microsoft" i.s.o. "msft" when I found this thread, which was a better place to ask the question.

I have indeed installed/seen/tested the camera control and face recognition and they work great on local compute power. The point is the difference between local and cloud compute power makes a stepwise change in capability/reliability. Like local speech recognition works fine but cannot compete with the power of cloud based api's. Hence my request for the Microsoft Face (recognition) API.

E.g. copied from

  • "This does require additional processing and will slow the framerate. Also, detecting images within a video stream with many complicated objects will return false positives. Use the Face Tracking against a white wall or non-complicated background."
  • "HAAR Cascades. This is very cpu intensive and will most likely bring a slower computer to a halt. If the input cascade is not optimized, it will also have great affects on framerate processing performance"

I don't mean HAAR cascades. There is an actual OBJECT TRAINING and RECOGNITION. I am pretty sure that will surprise you. You can train the robot an object (such as your face) and have it detect.

We have an upcoming The Robot Program episode on just that. Stay tuned, it'll all be clear if you haven't figured it out by then :D


Thanks for the delay Richard. :P

I finally managed to get a microsoft account, time to try these plug-ins.