Michael Ruhl
06. September 2017
1 min

ARKit and Face-Recognition in iOS

iOS 11

Adds a lot of new posibilites for developers with new frameworks for Augmented Reality (ARKit), Object Detection (Vision) and Machine Learning (CoreML).

To demonstrate we created an augmented reality app with face-recognition:

How does it work:

  1. The app continuously grabs the camera image
  2. Detects faces on the image using Vision
  3. Extracts the faces and runs through our machine-learning model
  4. Identifies persons with the result
  5. Calculates a three dimensional coordinate according to the location on the screen
  6. Displays information in AR

We will write a more in-depth blog post, how we created the machine-learning model 😉

Meanwhile the project is available on GitHub:

https://github.com/NovaTecConsulting/FaceRecognition-in-ARKit

Update:

Here is our follow-up blog-post about the model creation:

How we created our Face-Recognition model

Comment article

Comments

  1. Bruno

    Could you create a guide on how to create the model? Thanks!

  2. dream zhang

    hello, I want to know how do you created the machine-learning model?When can you write the blog?