Plotting Tracking Tools Data in Matlab

Post Reply
Posts: 3
Joined: Wed Jun 20, 2012 9:10 am

Plotting Tracking Tools Data in Matlab

Post by vrtroopers » Fri Jun 29, 2012 9:13 am


I am currently working on a project to use Tracking Tools and the data from rigid bodies to create a 3D model for future students. I would be grateful if you could answer my questions.

When using Trackables I am getting 8 columns of data. I looked around and couldn't figure out what they mean. I understand that the first three must be x,y,z and three others could be yaw,pitch, roll but I am confused as to which is which.

I am also interested in knowing if there is a Matlab code out there that will read data in from a Tracking Tools csv file, and create a live plot of the positions of the points in a 3D graph. Also, if there is a better way to keep track of or plot points in 3D, I would like to know.

I am currently working with Tracking Tools version 2.3.3 Final
on a computer with a NVidia Tesla and extremely sufficient hardware.


NaturalPoint - Mike
Posts: 1896
Joined: Tue Feb 01, 2011 8:41 am
Location: Corvallis, OR

Re: Plotting Tracking Tools Data in Matlab

Post by NaturalPoint - Mike » Thu Jul 05, 2012 1:27 pm

If you are referencing the CSV file, it outputs data in the following format:

Frame line:
Index, time stamp, trackable count, (trackable id, centroid x,y,z, qx,qy,qz,qw,yaw,pitch,roll) per trackable, marker count (marker x,y,z) per marker

Trackable line (one per trackable):
index, time stamp, name, id, frames since last tracked, marker count, (marker x,y,z) per trackable marker, (marker x,y,z) per point cloud marker, (is tracked?) per marker, (quality) per marker, mean error

The difference between trackable marker and point cloud marker (in the trackable line) is due to the fact that the trackable is a rigid set of definitions and the point cloud can fluctuate. The trackable marker locations are the expected locations of the markers, based on their position relative to the calculated centroid. The point cloud markers are the locations of the actual markers in the point cloud used to calculate the position and orientation of the trackable. Slight variance here is expected.

Post Reply