mp.tasks.vision.HolisticLandmarkerResult

The holistic landmarks result from HolisticLandmarker, where each vector element represents a single holistic detected in the image.

face_landmarks Detected face landmarks in normalized image coordinates.
pose_landmarks Detected pose landmarks in normalized image coordinates.
pose_world_landmarks Detected pose world landmarks in image coordinates.
left_hand_landmarks Detected left hand landmarks in normalized image coordinates.
left_hand_world_landmarks Detected left hand landmarks in image coordinates.
right_hand_landmarks Detected right hand landmarks in normalized image coordinates.
right_hand_world_landmarks Detected right hand landmarks in image coordinates.
face_blendshapes Optional face blendshapes.
segmentation_mask Optional segmentation mask for pose.

Methods

__eq__

face_blendshapes None
segmentation_mask None