chochim / projectx

Sciencenter Wall of Inspiration

ProjectX

Sciencenter Wall of Inspiration

An interactive, voice commands driven gallery based on Kinect 2.0 SDK

Pre-requisites

Before running the code, ensure that the following pre-requisites are installed on your machine:

1. Microsoft Visual Studio
2. Kinect for Windows 2.0
3. Kinect Speech Recognition API
4. Kinect Speech Recognition Language Pack
5. Visual Studio Github plugin

How to run?

Checkout the latest code from the master branch. Create a folder on Desktop named sciencenterimages that contains all the images required to run the slide show. The images are formatted to encode extra information. For example, if the image corresponds to Sir Isaac Newton and we want to keep it there for 5 seconds, the image will be named as newton_5.jpg. The application will pick up this information and break it down to adequate data structures. The official project website is maintained here.

Coding

The project is written in C# and works on Windows platform. The code functionality is divided into three main parts:

1. GestureRecognition: The main files that deal with gesture recognition are GestureBase.cs, GestureEventArgs.cs, GestureHelper.cs, GestureRecognitionEngine.cs, GestureRecognitionResult.cs, GestureRecognitionResult.cs, GestureType.cs, SwipeToLeftGestureWithLeftHand.cs, SwipeToLeftGestureWithRightHand.cs, SwipeToRightGestureWithRightHand.cs, and SwipeToLeftGestureWithLeftHand.cs
2. VoiceSearch: The voice search code retrieves the training examples from the information contained in the image names. The relevant code for this is MainWindow.xaml.cs.
3. UserInterface: The user interface consists of the entire slide show and the entire visual elements. The UI is defined at MainWindow.xaml and the slide show code is defined at MainWindow.xaml.cs.

NOTE - You need to compile the program for x86 or x64 architectures. This code does not work for ARM.