Kinect

Microsoft Kinect V2

The Kinect is a motion sensing input device. Based around a webcam-style add-on peripheral, it enables users to control and interact with their computer by retrieving joint information. Microsoft enables a non-commercial Kinect SDK for gestures and spoken commands.

Kinect V2

The device features an RGB camera, depth sensor and multi-array microphone running proprietary software, which provide full-body 3D motion capture, facial recognition, voice recognition and acoustic source localization capabilities.

The depth sensor consists of an infrared laser projector combined with a monochrome CMOS sensor, which captures video data in 3D. Kinect is capable of simultaneously tracking up to six people. sensors output video at a frame rate of 15 Hz(low light) to 30 Hz. The default RGB video stream uses 8-bit VGA resolution (512 x 424 pixels) with a Bayer color filter, but the hardware is capable of resolutions up to 1920x1080 @30fps colour image. The Kinect sensor has a practical ranging limit of 1.2–4.5 m distance. The Kinect can process 2 gigabytes of data per second, so it uses USB 3. The SDK runs only on Windows 8.

Look at the color depth :)

Color Depth Image

You can retrieve this joint information as well as finger information.

Joint Information Screen

Presentation software

 begin;

 array{
   ellipse_graphic {
      ellipse_width = 1;
      ellipse_height = 1;
      color = 255, 0, 0;
   };
   ellipse_graphic {
      ellipse_width = 1;
      ellipse_height = 1;
      color = 0, 255, 0;
   };
 }arrayJoints;

 picture {
	text { font = "Arial"; font_size = 20; caption = "move a bit:)"; } cap;
	x = 0; y = -400;
 } pic;

 begin_pcl;

 array<vector> data[0]; # in centimeters
 array<int> body_ids[0];
 int counter = 0, dataBody;
 string toScreen;

 sub change_caption( string caption ) begin
	cap.set_caption( caption, true );
	pic.present()
 end;

 sub showJoint( int joint , int partNr) begin
	pic.set_part_x(partNr+1,data[joint].x() * 10);
	pic.set_part_y(partNr+1,data[joint].y() * 10);
	if partNr <= 3 then
		arrayJoints[partNr].set_dimensions(data[joint].z(),data[joint].z());
		arrayJoints[partNr].redraw();
	end;
 end;

 kinect k = new kinect();
 depth_data dd = k.depth();
 dd.RES_512x424;

 body_tracker tracker = k.body();
 tracker.set_seated( true );
 tracker.start();

 pic.present();
 body_data bd;
 loop until !is_null( bd ) begin
	bd = tracker.get_new_body();
 end;

 #add the joints to picture
 loop
	int count = 1;
 until count > 6 begin
	pic.add_part(arrayJoints[count],0,0);
	count = count + 1;
 end;

 dd.start();
 loop until false begin
   if (dd.new_data()) then
		counter = counter + 1;
		bd.get_positions( data, body_ids);
		if data.count() > 0 then
			showJoint(bd.HAND_RIGHT,1);
			showJoint(bd.HAND_LEFT,2);
			showJoint(bd.HEAD,3);
			showJoint(bd.WRIST_RIGHT,4);
			showJoint(bd.HAND_TIP_RIGHT ,5);
			showJoint(bd.THUMB_RIGHT,6);
			toScreen = "dataCount: " + string(counter) + "\n";
			toScreen.append("data X: " + string(data[bd.HAND_RIGHT].x()) + "\n");
			toScreen.append("data Y: " + string(data[bd.HAND_RIGHT].y()) + "\n");
			toScreen.append("dataZ: " + string(data[bd.HAND_RIGHT].z()) + "\n");
			change_caption( toScreen );
		end;
   end;
 end;