Need clarification about the min_pose_presence_confidence in pose landmark detection

[KunHanKH]

Have I written custom code (as opposed to using a stock example script provided in MediaPipe)

None

OS Platform and Distribution

Ubuntu 18.04

MediaPipe Tasks SDK version

No response

Task name (e.g. Image classification, Gesture recognition etc.)

Human Pose Landmark

Programming Language and version (e.g. C++, Python, Java)

Python

Describe the actual behavior

The min_pose_presence_confidence seems not work in the demo

Describe the expected behaviour

Show different results with different setting

Standalone code/steps you may have used to try to get what you need

When I pass the parameter to the PoseLandmarkOption, there is a `min_pose_presence_confidence` parameter, and I\'m not sure of the usage of this parameter. Then I tried the online demo https://mediapipe-studio.webapps.google.com/demo/pose_landmarker, and adjusted the `min_pose_presence_confidence`. The results are the same. 

I want to know:
1. Is this `min_pose_presence_confidence` applied on each landmark or the mean of all landmarks
2. What happens if the actual confidence is below `min_pose_presence_confidence`? What happens to the detection result for the frame? Will I still get the detected landmark results or not?
3. What exactly is the `min_pose_presence_confidence` used for?

Thanks

Other info / Complete Logs

No response

[kuaashish]

Duplicate of #4859.

[google-ml-butler[bot]]

Are you satisfied with the resolution of your issue? Yes No