Microsoft AI: Mobile Emotion Recognition Application

There is no single unifying definition or theory of emotion. Nevertheless, several characteristics of an emotional reaction are frequently mentioned. Nowadays feedback is more important for product promotion and review the article, store, etc. manual entry of emotion maybe wrong so we can automate the feedback and review.



In this blog, we can learn about how to implement Emotion Recognition in Xamarin.Forms with Microsoft Cognitive Services.

Getting Started with the Emotion API 

The Emotion API can detect anger, contempt, disgust, fear, happiness, neutral, sadness, and surprise, in a facial expression. As well as returning an emotion result for a facial expression, the Emotion API also returns a bounding box for detected faces using the Face API. If a user has already called the Face API, they can submit the face rectangle as an optional input. Note that an API key must be obtained to use the Emotion API. You can follow below steps for obtained at free API key on Microsoft.com

Step 1: Click here for Generate Subscriber Key and Click on Create



Step 2: You can agree the terms and Microsoft privacy statement > select the country and click on Next



Step 3: Login with Microsoft /LinkedIn /Facebook /Github for create your API



Step 4: You login and subscribe you should see the keys and the below information on available your subscriptions 

We will use emotion API key in the following Xamarin application. before that we can test the API key

Test on Your Emotion Key

You can navigate emotion API online console application and provide the subscription key and image url . If your provide valid key and input the image, you will get it JSon data as output or relevant error

Create New Xamarin Forms Application 

You can refer my previous article for setup and create new Xamarin forms application.

If you already installed VS on your machine, Open Visual Studio > Create New Xamarin Forms application like below



Solution will create with all the platform and PCL project



In this solution, we are going to create demo application for Article review automate based on the emotion.

Camera in Xamarin. Forms 

We can use Media.Plugin Nuget package for Capture image or select image gallery from all the mobile platform (iOS,Android,Windows)

Right Click on Solution > Type “xam.plugin” in the search box > Select all the project (PCL, iOS, Android and UWP) > Click on Install

Emotion Nuget Package

 This client library is a thin C# client wrapper for the Microsoft Emotion API.The easiest way to use this client library is to get microsoft.projectoxford.emotion package from nuget .

Right Click on Solution > Type “Microsoft Emotion” in the search box > Select all the project (PCL, iOS, Android and UWP) > Click on Install

UI Design 

After successfully install above two nuget package. Let start UI design from PCL project In Dotnet Standard Project > Open MainPage.Xaml page and design following xaml code


<?xml version="1.0" encoding="utf-8" ?>
<ContentPage xmlns="http://xamarin.com/schemas/2014/forms"
xmlns:x="http://schemas.microsoft.com/winfx/2009/xaml"
xmlns:local="clr-namespace:DevEnvExeEmotion"
x:Class="DevEnvExeEmotion.MainPage">
<StackLayout Margin="20">
<StackLayout Orientation="Horizontal" Padding="30" HorizontalOptions="CenterAndExpand" >
<StackLayout x:Name="emotion">
<Label Text="Reader is :" FontSize="Large"/>
<Label x:Name="emotionResultLabel" FontSize="Large" />
</StackLayout>
<Button Text="Share Your FeedBack" Clicked="OnFeedBackClicked"></Button>
</StackLayout>
<Image x:Name="image" Source="article.png" />
<ActivityIndicator x:Name="activityIndicator" />
</StackLayout>
</ContentPage> 

In mainPage.xaml.cs file add your logic.


public partial class MainPage : ContentPage
{
EmotionServiceClient emotionClient;
MediaFile photo;
public MainPage()
{
InitializeComponent();
emotion.IsVisible = false;
emotionClient = new EmotionServiceClient("7f495be5faf643adbeead444d5b79a13");
}
For capture image, write the following code
public async Task CaptureImage()
{
await CrossMedia.Current.Initialize();
emotion.IsVisible = false;

// Take photo
if (CrossMedia.Current.IsCameraAvailable || CrossMedia.Current.IsTakePhotoSupported)
{
photo = await CrossMedia.Current.TakePhotoAsync(new StoreCameraMediaOptions
{
SaveToAlbum = false
});
}
else
{
await DisplayAlert("No Camera", "Camera unavailable.", "OK");
}
}
For Recognize emotion, write the following code
public async Task Recognizeemotion()
{
try
{
if (photo != null)
{
using (var photoStream = photo.GetStream())
{
Emotion[] emotionResult = await emotionClient.RecognizeAsync(photoStream);
if (emotionResult.Any())
{
// Emotions detected are happiness, sadness, surprise, anger, fear, contempt, disgust, or neutral.
emotionResultLabel.Text = emotionResult.FirstOrDefault().Scores.ToRankedList().FirstOrDefault().Key;
emotion.IsVisible = true;
}
photo.Dispose();
}
}
}
catch (Exception ex)
{
Debug.WriteLine(ex.Message);
}
}

While click on Feedback button, call the above two methods

async void OnFeedBackClicked(object sender, EventArgs e)
{
//capture image
await CaptureImage();
((Button)sender).IsEnabled = false;
// Recognize emotion
await Recognizeemotion();
((Button)sender).IsEnabled = true;
}

Run the Application

Select the iOS /Android /Windows and install into the device and click on ‘Share your Feedback ‘button

The device should be camera so after click on the button, capture your face for Recognize emotion.

 In this article, you learned about how to generate emotion API key and implement Emotion Recognition in Xamarin.Forms with Microsoft Cognitive Services.If you have any questions/ feedback/ issues, please write in the comment box.