Using Kinect with Emgu CV

Question

With EmguCV, to capture an image from a web-cam we use :

Capture cap = new Capture(0);

Image < Bgr, byte > nextFrame = cap.QueryFrame();

...

...

But I don't know how to capture images from my Kinect, I have tried kinectCapture class but it didn't work with me. Thanks

Answer 1

Basically , you need to capture and Image from the ColorStream and convert to a EmguCV Image class :

Conversion to EmguCV Image from Windows BitMap (Kinect ColorStream):

You have a Windows Bitmap variable, where holds Kinect Frame.

Bitmap bmap = new Bitmap(weightFrame,HeightFrame,System.Drawing.Imaging.PixelFormat.Format32bppRgb);

...

//Here is the code where you capture the image in the ColorFrameReady....

...

Image<Bgr,Byte> frameActualKinect = bmap.ToOpenCVImage<Bgr, Byte>();

Make the detection:

Resize

currentFrame = frameActualKinect.Resize(320, 240, Emgu.CV.CvEnum.INTER.CV_INTER_CUBIC);

//Convert it to Grayscale

gray = currentFrame.Convert<Gray, Byte>();

//Face Detector

MCvAvgComp[][] facesDetected = gray.DetectHaarCascade(face, 1.2, 10, Emgu.CV.CvEnum.HAAR_DETECTION_TYPE.DO_CANNY_PRUNING,new System.Drawing.Size(20, 20));

P.D (The helper method) :

public static Image<TColor, TDepth> ToOpenCVImage<TColor, TDepth>(this Bitmap bitmap)
        where TColor : struct, IColor
        where TDepth : new()
    {
        return new Image<TColor, TDepth>(bitmap);
    }

Answer 2

When using EmguCV, you usually use another library to access the Kinect. For example, The Kinect For Windows SDK, or OpenNI. Then after accessing the camera using OpenNI or the SDK, you can edit the image that you project on the screen using EmguCV's tools. Here's some links of how to use EmguCV with OpenNI and the SDK

• https://groups.google.com/forum/#!topic/openni-dev/I0InsNgIIz4
• http://www.geek-press.com/?p=23 (beta 2 of SDK)
• Emgu CV and the official Microsoft Kinect SDK?

Hope this helps!