Image Recognition, Vuforia and the HoloLens

This post has been republished via RSS; it originally appeared at: Channel 9.

Today's project comes to us via Sebastian Gambolati, who's written a great tutorial/how-to build image recognition into your HoloLens app...

How-To: Create an image recognition app with Vuforia and HoloLens

Image recognition and interpretation is a very requested task into modern application. Either you want to identify a specific image or object or either you want to interpret what the users get in front their smartphone camera.
A technology that is becoming more and more popular among developers with this task is Vuforia. This framework can recognize a specific image and track it while the image is in the scene. The same goes to Custom 3D object, Cylinders, boxes, etc.
Moreover, Vuforia is widely accepted because we can use it with almost any device, from older and cheap Android smartphone to the most recent VR/AR/MX devices. In this post, we will use the Vuforia capabilities from Unity3D because it is also multi-platform.
...
Prerequisites
First, we need to download Unity3D from here. You may try the personal (and free) plan. In the meantime, it is installing, review What is Unity3D.
What is Unity3d?...
What is Vuforia?...
Step 1: Setup Vuforia Database...
Step 2: Create Unity3D project and import Vuforia asset...
Step 3: More 3D Models...
Step 4: Deploy to Android...
Step 5: Deploy to HoloLens...
Conclusion
In this post, Upload images to a database that later can be used for tracking other images and objects. Moreover, we imported it to Unity3D and setup the ARCamera and ImageTarget component. And we imported an external 3D model to our scene. At the end, we build the project for different platform. And we didn’t make a single line of code!
We will continue in a next post showing how to recognize multiple images and how to distinguish each one showing different 3D models or specific data.