Input
ExecutionInitiate Execution
AI/ML/ONNX/Face
Detect faces and extract embeddings, gender and age using a face_id analyzer
Scores range from 0 to 10. Higher values mean more impact, exposure, or operational weight.
Initiate Execution
Face analyzer handle
Cache ID for the analyzer
Per-call detector confidence threshold (sessions are shared across threshold variants)
Per-call detector NMS threshold (sessions are shared across threshold variants)
Input Image
Maximum number of faces to embed and analyze
Done with the Execution
Analyzed faces
Face bounding box in pixels (shared object-detection type; carries the detection score)
5-point landmarks in pixels [[x, y], ...], if the detector produced them
512-dimensional, L2-normalized identity embedding
Estimated gender ("Male" / "Female")
Estimated age in years
Number of detected faces