feat: Full Body Tracking (FBT) via MediaPipe Pose #38
Labels
No labels
architecture
bug
duplicate
enhancement
help wanted
invalid
performance
quality
question
wontfix
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
lavarius/ProjectOverlay#38
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Overview
Extend the existing MediaPipe face capture pipeline to support Full Body Tracking (FBT) using MediaPipe's PoseLandmarker. Characters already have the RIG + IK effector hierarchy set up (IK/Arms, IK/Legs) — this issue wires it all up.
Architecture
Python Server (
PythonFaceCaptureServer)PoseLandmarker(model already present:models/pose_landmarker.task)ENABLE_POSE_TRACKINGtoggle totracking_config.pypose|world|{idx}|{x}|{y}|{z}over existing UDP channelUnity Client
FaceTrackingManager: addpose|message parsing +OnPoseTrackingbroadcast eventBodyTrackingClient.cs: MonoBehaviour added to each character model that:IK/Arms,IK/Legs) using key landmarkstrackHeadtoggle (default off — face capture already handles head)Landmark → IK Mapping
Head Overlap
Face capture already drives head/neck/chest via
MultiAimConstraint → HeadTracker. FBT will not override head by default (pose head data skipped). AtrackHeadinspector bool allows opting in if face tracking is disabled.Files Changed
PythonFaceCaptureServer/tracking_config.py— add pose togglePythonFaceCaptureServer/combinerTracker_udp.py— add PoseLandmarker + pose UDP messagesAssets/_SRC/code/PipeServer/FaceTrackingManager.cs— add pose parsing + eventAssets/_SRC/code/PipeServer/BodyTrackingClient.cs— new per-character FBT componentAssets/_SRC/Prefabs/Character Inside.prefab— add BodyTrackingClient to each model