NEW!  Version 3 of iPi Motion Capture is available. iPi Motion Capture 3 delivers new features for improved workflow including support of Kinect 2, improved arms tracking, simplified calibration and others. Read more >>

This documentation corresponds to version 2 of iPi Motion Capture. The latest documentation for version 3 can be found at https://docs.ipisoft.com/

Quick Start Guide for Dual Depth Sensor Configuration

From iPiSoft Wiki
(Difference between revisions)
Jump to: navigation, search
(Calibration)
(Calibration)
Line 61: Line 61:
  
 
== Calibration ==
 
== Calibration ==
 +
 +
=== Preparations ===
  
 
There are two possible arrangements of the two sensors:
 
There are two possible arrangements of the two sensors:
Line 90: Line 92:
 
       <td><center>too many yellow (no depth data) points<br/>too small<br/>too close to human body</center></td></tr>
 
       <td><center>too many yellow (no depth data) points<br/>too small<br/>too close to human body</center></td></tr>
 
</table></center>
 
</table></center>
 +
 +
 +
'''Important!''' Yellow color in the depth map means "depth here is unknown". Thus it is important to minimize amount of yellow points using appropriate materials and clothes.
 +
 +
=== Recording ===
 +
 +
After preparations and sensors setup record the calibration video:
 +
* Start recording by pressing the "Start Video Recording" button.
 +
* Initial two seconds of video should contain only background and static objects. Presence of actor or any moving objects in the initial two seconds of recorded video is unacceptable.
 +
* Go to the capture volume with prepared flat rectangular veneer/plywood/cardboard/pasteboard.
 +
* Turn the veneer/plywood/cardboard/pasteboard to the left and to the right thus view angle is changing but the veneer/plywood/cardboard/pasteboard is still visible in both sensors.
 +
* Stop the recording by pressing the "Stop Video Recording" button.
 +
 +
<table border=0><tr>
 +
<td>[[Image:Dual-kinect-calibration-pos-0.jpg|center|200px]]</td>
 +
<td>[[Image:Dual-kinect-calibration-pos-1.jpg|center|200px]]</td>
 +
<td>[[Image:Dual-kinect-calibration-pos-2.jpg|center|200px]]</td>
 +
<td>[[Image:Dual-kinect-calibration-pos-3.jpg|center|200px]]</td>
 +
</tr><tr><td height=10/></tr><tr>
 +
<td>[[Image:Dual-kinect-calibration-pos-2.jpg|center|200px]]</td>
 +
<td>[[Image:Dual-kinect-calibration-pos-1.jpg|center|200px]]</td>
 +
<td>[[Image:Dual-kinect-calibration-pos-0.jpg|center|200px]]</td>
 +
<td>[[Image:Dual-kinect-calibration-pos-4.jpg|center|200px]]</td>
 +
</tr><tr><td height=10/></tr><tr>
 +
<td>[[Image:Dual-kinect-calibration-pos-5.jpg|center|200px]]</td>
 +
<td>[[Image:Dual-kinect-calibration-pos-6.jpg|center|200px]]</td>
 +
<td>[[Image:Dual-kinect-calibration-pos-4.jpg|center|200px]]</td>
 +
<td>[[Image:Dual-kinect-calibration-pos-5.jpg|center|200px]]</td>
 +
</tr></table>
 +
 +
 +
'''Note''' that is important to have variety in angles from one hand and good visibility of the flat in each camera from the other hand. It is not so easy to reach good results for 90 degrees angle thus 60-80 degrees angles could be more practical.
 +
 +
<center><table border=0>
 +
  <tr><td width="45%"><center>[[Image:Dual-kinect-calibration-pos-good.jpg|center|400px]]</center></td>
 +
      <td width="10%"/>
 +
      <td width="45%"><center>[[Image:Dual-kinect-calibration-pos-bad.jpg|center|400px]]</center></td></tr>
 +
  <tr><td><center style="color:DarkGreen">'''Good'''</center></td><td/>
 +
      <td><center style="color:DarkRed">'''Bad'''</center></td></tr>
 +
  <tr><td><center>the flat is good visible in both sensors</center></td><td/>
 +
      <td><center>the flat is not visible in the left sensor</center></td></tr>
 +
</table></center>
 +
 +
 +
Open resulted video in the '''iPi Studio''' and check the following:
 +
* At least first two seconds of video contains only background and static objects (no actor, no calibration flat)
 +
* The floor (ground) is good visible on the depth maps from both sensors
 +
* Calibration flat is good visible on the depth maps most the time
 +
* Calibration flat are moving and has at least 30 degrees variety in the horizontal view angle
 +
* The amount of yellow points is rather small
 +
 +
=== Processing ===
  
 
== Tracking ==
 
== Tracking ==

Revision as of 18:00, 12 October 2011


  BETA STATUS  


Contents

Prerequisites


Note that before starting of working with two Kinects it's highly recommended to get appropriate results with single Kinect solution: Quick Start Guide with MS Kinect Sensor.


Recording with 2 Kinects

Important! Each Kinect Sensor should be plugged through different USB controllers. If both Kinects are plugged through one and the same USB controller then the second Kinect will not work. How it looks in the Device Manager:

Dual-kinect-device-manager-good.png
Dual-kinect-device-manager-bad.png
Good
Bad
Both Kinects are plugged correctly
Both Kinects are plugged through one and the same USB controller

Run iPi Recorder and mark both MS Kinect Sensors in the Camera Selector dialog:
Click to enlarge

Press “Ok” button. As a result you should see something like this:
Click to enlarge


Make sure that you can see correct colored “depth video” from both sensors and

  1. that you can change motor position for each sensor using appropriate slider and “Motor of” combobox
  2. that you can switch between RGB and “depth” video using “View RGB/Depth” button
  3. that actual frame rate is 30 fps (or 29.9) and frames drop is in range from 0 to 5 for each sensor

Note that you can turn off RGB streaming using appropriate checkbox in the Advanced Options section of Camera Selector dialog. Also you can change correspondence between cameras and motors in case when they are in the wrong order:
Click to enlarge

Don't forget to set output directory for recorded video:
Click to enlarge

Calibration

Preparations

There are two possible arrangements of the two sensors:

  1. angle between sensors is between 60 and 90 degrees;
  2. angle between sensors in near to 180 degrees that is sensors are opposite to each other.


At present time we strongly recommend to use the first configuration:

Click to enlarge
Click to enlarge
Top view
Side view


For calibration purposes flat rectangular veneer/plywood/cardboard/pasteboard is used:

  • horizontal size should be at least 0.5m (1m - 1.3m is recommended);
  • vertical size should be at least 0.7m (1m - 1.5m is recommended);
  • hold it vertically in front of yourself on outstretched arms;
  • make sure that the veneer/plywood/cardboard/pasteboard is good visible in the both sensors (the amount of yellow is low in the area of veneer/plywood/cardboard/pasteboard).
Dual-kinect-calibration-good.png
Dual-kinect-calibration-bad.png
Good
Bad
good size
good depth data
too many yellow (no depth data) points
too small
too close to human body


Important! Yellow color in the depth map means "depth here is unknown". Thus it is important to minimize amount of yellow points using appropriate materials and clothes.

Recording

After preparations and sensors setup record the calibration video:

  • Start recording by pressing the "Start Video Recording" button.
  • Initial two seconds of video should contain only background and static objects. Presence of actor or any moving objects in the initial two seconds of recorded video is unacceptable.
  • Go to the capture volume with prepared flat rectangular veneer/plywood/cardboard/pasteboard.
  • Turn the veneer/plywood/cardboard/pasteboard to the left and to the right thus view angle is changing but the veneer/plywood/cardboard/pasteboard is still visible in both sensors.
  • Stop the recording by pressing the "Stop Video Recording" button.
Dual-kinect-calibration-pos-0.jpg
Dual-kinect-calibration-pos-1.jpg
Dual-kinect-calibration-pos-2.jpg
Dual-kinect-calibration-pos-3.jpg
Dual-kinect-calibration-pos-2.jpg
Dual-kinect-calibration-pos-1.jpg
Dual-kinect-calibration-pos-0.jpg
Dual-kinect-calibration-pos-4.jpg
Dual-kinect-calibration-pos-5.jpg
Dual-kinect-calibration-pos-6.jpg
Dual-kinect-calibration-pos-4.jpg
Dual-kinect-calibration-pos-5.jpg


Note that is important to have variety in angles from one hand and good visibility of the flat in each camera from the other hand. It is not so easy to reach good results for 90 degrees angle thus 60-80 degrees angles could be more practical.

Dual-kinect-calibration-pos-good.jpg
Dual-kinect-calibration-pos-bad.jpg
Good
Bad
the flat is good visible in both sensors
the flat is not visible in the left sensor


Open resulted video in the iPi Studio and check the following:

  • At least first two seconds of video contains only background and static objects (no actor, no calibration flat)
  • The floor (ground) is good visible on the depth maps from both sensors
  • Calibration flat is good visible on the depth maps most the time
  • Calibration flat are moving and has at least 30 degrees variety in the horizontal view angle
  • The amount of yellow points is rather small

Processing

Tracking

Next Steps

Personal tools