The MediaPipe Holistic Landmarker task lets you combine components of the pose, face, and hand landmarkers to create a complete landmarker for the human body. You can use this task to analyze full-body gestures, poses, and actions. This task uses a machine learning (ML) model on a continuous stream of images. The task outputs a total of 543 landmarks (33 pose landmarks, 468 face landmarks, and 21 hand landmarks per hand) in real-time.
An upgraded version of this MediaPipe Solution is coming soon! The MediaPipe Legacy Solution for this task is available on GitHub.