Skip to content
master
Go to file
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
 
 
 
 
 
 

README.md

Hotdog or Not

This is a small app demonstrating how to use CoreML with Xamarin to build an image detection app. ARKit is used to preview the camera and Vision is used to execute the model.

The model it uses was trained with Microsoft's Custom Vision service.

iOS

All the interesting code is in ViewController.cs.

Android

For Android all analysis is done using Custom Vision services. You must setup a model and then fill in the API Keys inside of HotDogOrNotPage.xaml.cs

Guid projectId = Guid.Parse("");
const string predictionKey = "";

IMPORTANT

Note: Ensure in [Microsoft's Custom Vision service] that a default is set under the Performance Tab after training. (Beside the Prediction URL under the nav bar ). Otherwise the Line within the Predict Method :

var result = await endpoint.PredictImageAsync(projectId, stream);  

Will return a http response of "Not Found". This is due to the Prediction URL containing the iteration id appended to the end of the url. This is removed once default is set.

Screenshot of the app detecting a hotdog

About

A minimal Xamarin CoreML app to detect hotdogs

Resources

License

Releases

No releases published

Packages

No packages published

Languages

You can’t perform that action at this time.