Input
ExecutionInitiate Execution
AI/ML/ONNX
Detect human poses and keypoints using ONNX models. Download models from: YOLOv8-Pose (https://docs.ultralytics.com/models/yolov8/), MoveNet (https://tfhub.dev/google/movenet/), HRNet (https://github.com/OAID/TengineKit)
Scores range from 0 to 10. Higher values mean more impact, exposure, or operational weight.
Initiate Execution
ONNX Model Session
Cache ID for Session
Image Object
Minimum keypoint confidence threshold
Maximum number of poses to detect
Done with the Execution
Detected poses with keypoints
All keypoints in this pose
A single keypoint in a pose skeleton
Keypoint index in the skeleton
X coordinate (in image space)
Y coordinate (in image space)
Confidence score for this keypoint (0.0-1.0)
Optional keypoint name (e.g., "left_shoulder", "right_knee")
Overall confidence score for the pose detection
Optional bounding box (x1, y1, x2, y2) around the detected person
Skeleton connections (which keypoints connect to which)
Connection between two keypoints in a skeleton
Index of the first keypoint
Index of the second keypoint