Emoji8 app: Rate your facial expressions against emojis
Want to know how much your expression is in place? Microsoft recently released the latest Emoji8 app, which allows you to get accurate expression scores. The Emoji8 app is a super fun Windows app that provides emoticons for users to imitate, interpreting and inputting accuracy percentages and scores by analyzing video input from a webcam in real time. What’s more interesting is that users can also directly share the GIF of the expression to the social networking site.
Microsoft project manager Killian McCoy said in a blog post that Emoji8 uses “Windows Machine Learning” to evaluate the user’s facial expressions while simulating randomly selected emojis. It can be played on the app even if there is no internet connection.
Basically, some common emoji expressions such as anger, crying, surprise, happiness, etc. can be scored through the Emoji8 app. This app is for users who are updating with Windows 10 October 2018 and can be downloaded for free in the Microsoft Store.
Teach the machine to feel emoji with the help of deep learning
The chat robot Dango is based on neural network development, developers have used millions of examples to train the neural network, let the program better understand the meaning of emoji .
Recently, neural networks have become an option for solving a series of computer science problems: Facebook uses neural networks to identify faces in images, and Google uses them to identify everything in an image. Apple uses them to understand what you say to Siri, and IBM is used to manipulate the synergy of business units.
This is too impressive. But what about the actual problem? Can the neural network help you find any emoji you need when you need it?Oh, yes. They can.
This article will outline some of the engineering principles behind Dango that allow us to learn automatically from the use of hundreds of millions of real-world symbolic expressions, resulting in a lightweight and fast tool for real-time on mobile phones. Predict emoji for you.
What is Dango?
Dango is a floating assistant that runs on your phone and predicts emoji, textures and GIF images based on what you and your friends are writing in any app. This allows you to have the same rich conversations in any application: Messenger, Kik, Whatsapp, Snapchat, etc. (Implementing such functionality in all applications is a big challenge, but it’s not discussed here).
It’s difficult to recommend emoji: Dango must understand the meaning of the statement you are writing, and then recommend the expression you want to use. In the core, Dango’s predictions are implemented by neural networks. A neural network is a computational structure with millions of tunable parameters that are connected to each other in a manner similar to that of human brain neurons.
Randomly initialize these parameters to train the neural network, then enter millions of real-world emoji use cases captured from the Internet.
At first, the network was just a random guess, but with the input of the new training example, it slightly adjusted the millions of parameters, so it would perform better on that example. After training for a few days on the top GPU, the network began to output more meaningful recommendations.