How to get the position (x,y) from a Kinect depth array?

Question

While working with the kinect I found out, that the bitmap and its depth information are unreliable and for some reason much more disturbed than the data from the actual byte array.

I realised this when I tried get the min and max by accessing the bitmap like this

for (var y = 0; y < height; y++)
{
  var heightOffset = y * width;
  for (var x = 0; x < width; x++)
  {
    var index = ((width - x - 1) + heightOffset) * 4;
    var distance = GetDistance(depth[depthIndex], depth[depthIndex + 1]);

But on the other hand I achieved much better results when I directly accessed the depth byte array (as a Stackoverflow member Chris pointed out, thank you, Chris) with this method:

int min2 = int.MaxValue, max2 = int.MinValue;
for (int i = 0; i < depth.Length; i += 2)
{
  int dist = GetDistance(depth[i], depth[i + 1]);
  if (dist < min2 && dist > 0) 
    min2 = dist;

  if (dist > max2) 
    max2 = dist;
}

My question is:

I want to count all the points within a certain distance intervall.
Then I want to get their actual position (x,y).

How can I get the position or directly add those points to a list - for further action?

My ultimate goal is to identify certain predefined objects by counting the number of pixels within certain intervall (to distinct a ball from a block for example).

for (int i = 0; i < depth.Length; i += 2)
{
  int dist = GetDistance(depth[i], depth[i + 1]);

  if (dist < z2 && dist > z1) 
    a++;
}

Chris · Accepted Answer · 2013-11-14T16:29:59.320

Assuming depth is a byte array with 2 * width bytes per row, and height rows, try something like this:

var points = new List<Point>();
for( int y = 0; y < height; y++ )
    for( int x = 0; x < width; x++ )
    {
        int i = y * width * 2 + x * 2;
        int dist = GetDistance( depth[i], depth[i + 1] );

        if( dist < z2 && dist > z1 )
            points.Add( new Point( x, y ) );
    }

You'll obviously have to replace Point with some type to represent an int x/y pair!

You could also do it with a simple for (int i = 0; i < depth.Length; i += 2) as before, and then calculate the x/y values from i. (Hint: the modulus operator will give you the x value, and a simple integer division will get you y, more or less)

How to get the position (x,y) from a Kinect depth array?

1 Answers1

Linked