28 Sep Akvelon uses AI to Create New Facial Recognition Emoji Game
Putting AI to Work
After you load the game just smile to begin!
As the emoji enters the screen from the right hand side you must match the face that it’s making until it disappears and the next emoji enters the screen. You gain points with each emoji you destroy and you can earn extra points by destroying a blue “Boss Emoji”. A Boss Emoji will appear after a player destroys six emoji’s in a row. To defeat a Boss Emoji you must match the emotion until the laser in the bottom right corner gets six charges (one charge=matching emotion for half a second). When you defeat the boss you will receive 24 extra points!
You can also gain bonus points by keeping a streak alive. The first emoji you destroy will earn you 1 bonus point then if you destroy 4 emoji’s in a row you will earn 2 bonus points and if you destroy 7 emoji’s in a row you will earn three bonus points! However, you have to keep the streak going because if an emoji is not destroyed your bonus point score will reset to 0. One of the easiest way to keep a streak alive is to grab a friend and use Command Mode! In Command Mode, the two players work together (instead of against each other) to destroy the emoji’s. If player one successfully destroys their emoji but player two is struggling a bit, player one can step in and destroy the emoji for them giving that team a better chance of receiving bonus points!
The Dream Team
Our team was led by Project Manager Ivan Efimov and Key Developer Aleksey Kharchenko (pictured below, in reverse). They created this game to show how powerful front-end is since it is the technology the user directly interacts with. The creators also wanted to show how this game can have the traditional feel of an app, in terms of UI and performance, but also the benefits of easy updating and single code for various platforms.
There are three modes you can play in:
Single player: Play by yourself and try to beat your own high score!
Multiplayer: Grab a buddy and compete to win!
Command: Team up with a friend and save the planet together!
Typescript, webRTC, and API being put to use
The game was written in TypeScript with the help of React, and WebRTC was used to capture the media. The media was then processed with trackin.js to detect faces and web workers transformed the face detection from a multithreaded system in single threaded JS. 3rd party API (Azure Face API and Face++) was used for the emotion detection aspect of the game and also dedicated a lot of time to UI using techniques like absolute positioning, modern grid layouts, and manual transition calculations over multiple properties to ensure smooth animation within the game.
Interested in saving a planet and trying out your best “emoji face”? Try this free game today!
AKVELON AUGMENTED REALITY PHOTO CONTEST, ENTER NOW TO WIN!
Our team has developed a fun and interactive AR download within the free SeekXR app that lets you take Akvelon Man, our company’s mascot, wherever you go! This app will bring Akvelon Man to life through AR, overlaying his image onto your surroundings through your phone’s camera. Read more.
MACHINE LEARNING PLATFORM FOR DATA ANALYSIS
Nowadays, fast-growing Machine Learning technologies allow us to solve an increasing number of real-life problems. Routine operations in such cases can be reduced by creating a system that allows algorithm developers to focus on their tasks instead of on supplying execution environments. Read more.
WHAT IS AR AND HOW DOES IT COMPARE TO VR AND MR?
Augmented Reality is changing the way we see the world every day. Peering at our phone screens, we can add digital creatures like Pokémon to our surroundings, we can alter our faces with fun filters for pictures and videos, and can even test out furniture in our rooms before we buy it. Read more.