AForge.NET

  :: AForge.NET Framework :: Articles :: Forums ::

Movement Recognition

The forum is to discuss topics from different artificial intelligence areas, like neural networks, genetic algorithms, machine learning, etc.

Movement Recognition

Postby Julia.O » Fri Apr 12, 2013 8:07 pm

Hi all,

I am currently working on my Master's thesis and I could need help on a problem related to machine learning. I am developing a game for patients suffering from vertigo (dizziness) with a balance board as input device. I am using the XNA Framework (C#) for the creation of the prototype.

I never got in touch with the topic of machine learning during my studies, so I'm rather new to this. My aim is to detect specific movement patterns in the player's performance (e.g. one leg stand, bend down). In order to define how these patterns look like, I developed a testing software which collects performance data of healthy people. So I have sample data sets of "correct" movements in form of XML files containing (x,y)-coordinates (center of gravity) at different points of time.

I am now looking for a library that enables me to let the software learn from this data and decide if the player performed a given movement correctly. Is this possible with AForge.NET? If yes, which approach should I choose towards this issue? Is there many any sample I could use as a reference point?

Best regards,
Julia
Julia.O
 
Posts: 1
Joined: Fri Apr 12, 2013 7:24 pm



Return to Artificial Intelligence

cron