Plans to Use PoseWorldLandmark for Controlling Multiple Annotations in a Room via Task APIs?

[isoroku56]

Feature Description

Plugin Version or Commit ID

v0.14.4

Unity Version

2022.3.25f1

Description

I particularly enjoy using the scene at MediaPipeUnity/Samples/Scenes/Tasks/PoseLandmark/PoseLandmarkDetection, which allows simultaneous pose estimation for multiple people using NumPoses. While the scene at MediaPipeUnity/Samples/Scenes/PoseTracking/PoseTracking does not support multiple people, it does allow placing poses in a 3D space using WorldLandmark within a Room. I would like to know if there are any plans to implement a similar feature in the Tasks scene, where we could set up a Room and observe poses moving in 3D space.

Thank you very much for your consideration.

Current Behaviour/State

No response

Additional Context

No response

[homuler]

I thought it wasn't necessary, so I deleted it. The origin of the world coordinate system is the hip, so there's no point in displaying more than one person, and I don't think I'll implement this feature. Of course, it's possible to move and display based on the hip position from the Landmarks, but I think it's too complex as a sample.

cf. https://ai.google.dev/edge/mediapipe/solutions/vision/pose_landmarker/python

The output contains the following world coordinates (WorldLandmarks):

• x, y, and z: Real-world 3-dimensional coordinates in meters, with the midpoint of the hips as the origin.