Body Tracking

Track a person's body pose, dense face mesh with blendshapes, hands, and gestures.

What it does

Body Tracking runs a single machine-learning model over an image to find a person: body pose, hands and face at once. The full set of pins is always present, body keypoints, left and right hand keypoints, face keypoints, a skeleton preview and per-point confidences, so downstream wiring survives a toggle being flipped. Regions that are switched off come through empty.

Dense Face gives a 478-point face mesh plus 52 expression blendshape coefficients, Dense Hands gives finer hand landmarks, and Gestures gives a per-hand gesture label. A One-Euro smoothing filter is on by default to settle per-frame jitter, and the skeleton preview draws over the image while the node is selected.

When to use it

ParameterTypeDefault
imageOneOf([FieldOf(Raster)])
enabledBooleantrue
detector_sourceString"whole_person"
model_qualityString"balanced"
confidence_thresholdNumber0.300
hand_confidenceNumber0.200
face_confidenceNumber0.200
min_detection_confidenceNumber0.500
min_tracking_confidenceNumber0.500
dense_faceBooleanfalse
blendshapesBooleantrue
dense_handsBooleanfalse
gesturesBooleanfalse
smoothingBooleantrue
smoothing_min_cutoffNumber1
smoothing_betaNumber0.500
line_widthNumber3
joint_radiusNumber6
point_sizeNumber2
colorColorrgba(0.00, 1.00, 0.00, 1.00)
preview_regionsString"all"
opacityNumber1

Gotchas

If a Model Quality tier is not installed, the balanced model is used instead with no other sign of it. Check results when expecting the accurate tier.

Worked example

  1. Add Body Tracking and wire a Camera or Image into its input.

  2. Leave Detector Source on whole-person and Enabled on. A skeleton preview draws over the image.

  3. Turn on Dense Face and Blendshapes to get facial expression coefficients for a face rig.

  4. Turn on Dense Hands and Gestures to get finer hand tracking plus a canned gesture label per hand.

  5. If tracking looks jittery at rest or laggy in motion, adjust Smoothing Min Cutoff and Smoothing Beta to taste.

See also

Vision Pose · Object Detect · Camera Input · Points