The Interactive Art group at CITAR, recently bought a Neurosky Mindset device (it's a headset that reads the brainwave activity) and, because there were no available Processing libraries to fetch the brainwave data I decided to implement my own.

So... here's the MindsetProcessing library. 

This class only takes care of getting the access token, the rest is up to you. It uses the WebBrowser component from the Lobo Browser library to display the authentication page (I tested it on Windows and Mac OS X).  Here you find an example Processing project with all the components. 

I implemented this as part of an example for the Multimedia Programming course of the School of Arts.

Tenho sido contactado várias vezes sobre a existência de cursos sobre Processing. Aqui fica a lista que eu conheço... Se souberem de mais ou quiserem corrigir alguma informação estejam à vontade.

I couldn't resist to play a bit more with the Face Detection library for Processing... see my previous post.

 This time the eyes follow the mouse :)

After seeing Bryan Chung's face detection Processing library for windows I decided to have a go with it.

First, I had trouble running the original example because of this error (see these comments on Bryan Chung's blog):

Exception in thread “Thread-2″ java.lang.UnsatisfiedLinkError: D:\processing-0135-expert\libraries\pFaceDetect\library\PFaceDetect.dll: This application has failed to start because the application configuration is incorrect. Reinstalling the application may fix this problem
at java.lang.ClassLoader$NativeLibrary.load(Native Method)
at java.lang.ClassLoader.loadLibrary0(Unknown Source)
at java.lang.ClassLoader.loadLibrary(Unknown Source)
at java.lang.Runtime.loadLibrary0(Unknown Source)
at java.lang.System.loadLibrary(Unknown Source)
at pFaceDetect.PFaceDetect.(PFaceDetect.java:10)

I ended up having to recompile the C++ code and replacing the original dll with my own, but Bryan's tested on several computers and he had no problems, so this was probably something wrong with my setup. Here's my pFaceDetect.dll, if yours doesn't work either.

I adapted slightly the original example so that it would track a single face (but detects all faces in the image) and replace it with a smiley.

In the video you can note some of the library's limitations (actually, OpenCV's haarcascade tracking limitations): sometimes it finds faces where there are none. Sometimes it doesn't find a face where there is one.

You can't see this in the video, but the maximum framerate I got in my laptop was 10 fps. This is low, but it doesn't stop you from doing something cool with this library, and there are probably ways to make it faster.

PS: I was chewing gum in the video...

This code is still very messy, but I've decided to make public anyway..

This is the code for the last example. 

edit: You need Java 1.5 installed to run the examples. Install the JDK 1.5, or higher, and install Processing (without Java).

Another video that demonstrates the use of background subtraction and silhouette extraction.

The spikes in the silhouette represent the normal of the silhouette segments, that are used to calculate the reflected angle when balloons touch it.

I have been working on some examples for my Interactive Video Art course at the School of Arts of the Portuguese Catholic University. 

Here's a video of an example that performs background subtraction and silhouette extraction. I'll not post the code because it's still a work in progress...

The video is low quality so you can't see the silhouette lines...