none
Interactive Floor RRS feed

Answers

  • You can detect someone's position using Kinect as shown in SkeletalViewer sample (installed to your computer with SDK, at C:\Users\Public\Documents\Microsoft Research KinectSDK Samples\NUI\SkeletalViewer), or also http://channel9.msdn.com/Series/KinectSDKQuickstarts/Skeletal-Tracking-Fundamentals from Channel9 Coding4Fun.

    However, Kinect sensor range is 0.8-4.0 meters away from Kinect sensor device. The video shown is of a hallway that seems significantly longer than 4 meters, so you would not be able to accomplish this experience with one single kinect device.

    About making objects move based on motion detected, that is a matter of animating 2-D images or 3D models based on perceived movement, and that is a question for a different forum, sorry. If you want to use DirectX technologies, you could ask a direct X forum such as those listed under http://forums.create.msdn.com/forums/default.aspx?GroupID=7. If you want to use WPF technologies for rendering, you could ask WPF forum (http://social.msdn.microsoft.com/Forums/en/wpf/threads). If you want to use a different rendering technology for the animation, you should do some research to see what is available out there.

    Hope this helps,
    Eddy


    I'm here to help
    Monday, August 22, 2011 8:17 PM

All replies

  • You can detect someone's position using Kinect as shown in SkeletalViewer sample (installed to your computer with SDK, at C:\Users\Public\Documents\Microsoft Research KinectSDK Samples\NUI\SkeletalViewer), or also http://channel9.msdn.com/Series/KinectSDKQuickstarts/Skeletal-Tracking-Fundamentals from Channel9 Coding4Fun.

    However, Kinect sensor range is 0.8-4.0 meters away from Kinect sensor device. The video shown is of a hallway that seems significantly longer than 4 meters, so you would not be able to accomplish this experience with one single kinect device.

    About making objects move based on motion detected, that is a matter of animating 2-D images or 3D models based on perceived movement, and that is a question for a different forum, sorry. If you want to use DirectX technologies, you could ask a direct X forum such as those listed under http://forums.create.msdn.com/forums/default.aspx?GroupID=7. If you want to use WPF technologies for rendering, you could ask WPF forum (http://social.msdn.microsoft.com/Forums/en/wpf/threads). If you want to use a different rendering technology for the animation, you should do some research to see what is available out there.

    Hope this helps,
    Eddy


    I'm here to help
    Monday, August 22, 2011 8:17 PM
  • Dear Eddy, 

    Thanks for your good reply but I think using the depth image will be more better coz the Kinect's position will be on the top of the floor so the Skeletal tracking will not be a good solution here. I asked the question here to know whether there is a standard way to do it and if there is any available implementation like this using Kinect SDK?


    Bo_Ansari
    Monday, August 22, 2011 10:57 PM
  • I'm not aware of any standard way of implementing interactive floors with kinect. Kinect hasn't been out in the market too long and people are mostly still experimenting with that functionality in creative ways (i.e.: however they can get their scenario work given their constraints) for lots of different kinds of applications, rather than having settled for any specific standard way that is always followed.

    We also don't have a sample for this scenario, so you will either have to search the web for a sample close enough to your needs, or write the code yourself from scratch, learning from the more general-purpose samples installed with Kinect SDK.

    The main pieces of advice I can give you are that:

    1) Yes, if kinect will be on the top of the floor then using depth image directly will probably yield better results than trying to use the default skeletal tracking.
    2) given the size of the hallway shown in video, you will most likely have to use multiple kinect sensors, because each one can sense depth for up to 4 meters, and that hallway is longer than that.

    Good luck!
    Eddy


    I'm here to help
    Wednesday, August 24, 2011 9:55 PM
  • Hi,

    I'm working in a similar project and this thread were so helpful for me.

    It's my first project with Kinect, and before start working I want to have some ideas clear.

    I will use depth image as a movement sensor. I will compare each frame from depth source with one first calibration frame, so if I get differents depth points in an area, an action will begin in the same area.

    The question is about how to compare the area provided by Kinect and the area of my software, because I have differents dimensions:

    The projectors use a 1280x800px dimension, so my software will work with this dimension. But frames from kinect have different dimensions. I've seen SDK can transform between color camara an depth camara with coordinate mapper, but neither type of frames have the dimension I need (1280x800px).

    So I have to resize the frame from kinect to the dimension I need and then compare areas?

    Thanks in advance.

    Monday, April 14, 2014 10:48 AM