Hello all, I just found a tool I think you'll love that was given me by the author. Colin Clark, an incredible person, a composer and software developer (and much more !) based in Toronto. He created a software called Movement OSC, (mac PC and linux) that does pose estimation and sends osc. Movement OSC detects body keypoints using Tensorflow.js' MoveNet models and sends data over OSC. It can also compare Movenet and MediaPipe so you can compare them ! The software formats the osc messages 4 ways Bundeled messages per axis, messages per axis, and bundled xyz array. I use just the latter then parse it in the Json parser 

Here are the keypoints and the keypoints diagram 

Enjoy !

MovementOSC provides a normalized keypoint format that is consistent across all its supported models:

33 keypoints in totalthe first 16 keypoints are the COCO setthe remaining 17 keypoints are those unique to BlazePose, in order from head to toeregardless of which model you use, all 33 keypoints will be provided; keypoints that aren't supported by the model, or which don't meet the specified minimum score, will be set to NaN

Keypoints

0: nose
1: left_eye
2: right_eye
3: left_ear
4: right_ear
5: left_shoulder
6: right_shoulder
7: left_elbow
8: right_elbow
9: left_wrist
10: right_wrist
11: left_hip
12: right_hip
13: left_knee
14: right_knee
15: left_ankle
16: right_ankle
17: left_eye_inner
18: left_eye_outer
19: right_eye_inner
20: right_eye_outer
21: mouth_left
22: mouth_right
23: left_pinky
24: right_pinky
25: left_index
26: right_index
27: left_thumb
28: right_thumb
29: left_heel
30: right_heel
31: left_foot_index
32: right_foot_index

@joe_pavey123

Are you referring to PoseOSC? If so all the info you need is on the Github page, including a link to the keypoint diagram.

There seem to be only 38 (max) tracking points.
Each seems to have a x,y, and 'score' value, so I think this is only 3*38 OSC values isn't it? Not thousands ☺

HTH

Mark (not that Mark)

@Armando OH MY GOD !!!! I tried to connect a broadcaster to skeleton output property of the rokokko actor and putting a listener in data mode and it works ! It is just that listener type must t be set on data. when like that its value output becomes "skeleton" if attached to a skeleton decoder. The only confusing part is that the broadcaster value input doesn't show data not skeleton but just three dashes ! It works !!!!

Link to video

@woland

Thank you for your time. I had installed the patch you mentioned. v1.0.1

What I found was the problem to be was - in the Isadora Plugins Folder, under the properties for the RokokoStudio Live Watcher, Skelton Decoder etc, the various .dll and .qtr files, their Attributes had been blocked by Windows 10 because "it was downloaded from another computer". I checked  on the unblock option and hit apply and - magic -  they are talking to each other. 

Big thanks Isadora Community. 

As an aside Rokoko tech support advised to use Json in Rokoko's Livestreaming. That was wrong for me. JsonLz4 worked not Json.They also advised to avoid Mac.....??

Thanks Again

@jtoenjes said:

OpenCV python script

 At this time this is not possible. Python scripting is something we are working on, but is not yet available.

@dusx yes, I have done this, libraries are not needed to do the maths, but on the end we can only draw with Euler.