Image recognition combined with AR.
Image detection combined with AR.
It was a bonus for AR_DOA School Project at Bonch-Bruevich Saint Petersburg State University of Telecommunications (SPBSUT) .
Russian: Санкт-Петербургский государственный университет телекоммуникаций (СПбГУТ) .
The goal was to provide a bonus to our school project and, also test other function from ViroAR Library.
Demo of the fonctionnal application:
Personnal conclusion:
The image fetched was made by myself, the library has some trouble to find the picture if it’s upside down or if the picture is mainly of 1 color.
Libraries used:
ViroAR by ViroMedia
Prerequisites:
Android -> An ARCore supported device .
iOS -> iOS Device with A9 chip or higher and running iOS 11 or higher .
Make sure to have npm and node installed on your pc .
MACOS:
https://blog.teamtreehouse.com/install-node-js-npm-mac
LINUX:
https://blog.teamtreehouse.com/install-node-js-npm-linux
WINDOWS:
https://blog.teamtreehouse.com/install-node-js-npm-windows
Register into ViroMedia and generate a new API Key
Edit App.js under your project.
You can also change the image detected by the application by switching logo.png to another picture in the js/res/ folder.
Download the ViroMedia testbed app:
iOS:
https://itunes.apple.com/us/app/viro-media/id1163100576?mt=8
Android:
https://play.google.com/store/apps/details?id=com.viromedia.viromedia
Now, make sure ngrok is also installed and run the command npm install .
After the installation is completed you can run npm start, the ngrok link should appear in the console.
You just have to copy it to the ViroMedia testbed app.