Body Tracking Example

Hey Guys,

Question regarding the Object Tracking Example - is it possible to save only the skeleton without the image?
Thanks in advance.



Yes, the skeleton information is stored inside bodies object when calling :
zed.retrieve_objects(bodies, obj_runtime_param) (L.87 of

You can then access the 3D coordinates of each keypoint in the skeleton by doing for example :

# Loop over each detected body
for body in bodies.object_list:
    # Loop over each body part in the skeleton
    for part in range(len(sl.BODY_PARTS)-1):
        # Get the 3D coordinates for each keypoint 
        kp = body.keypoint[part]       

You can find more information on how the body parts are defined in the API Reference

Oh - Thanks!

Is there an “easy” option to print this points in an opencv window?

Because the file is not really easy to understand…


You can take a look at our older version of the body tracking sample, which uses OpenCV for 2D display.

It is available here :

1 Like

Exactly what i am looking for!