cmvr_edge_ai/configs/detection_viewer.yaml

api_version: cmvr.edge.ai/v1

runtime:
  # Decoder, two YOLO models and viewer drawing share this bounded pool.
  thread_workers: 4
  shutdown_timeout_s: 8

endpoints:
  cmvr_es:
    transport: grpc
    # Remote cmvr-es address. Keep this aligned with the robot being viewed.
    target: 192.168.0.119:50052
    tls: false
    timeout_s: 5
    options:
      max_receive_mb: 32

pipelines:
  detection_show:
    enabled: true
    nodes:
      camera:
        uses: cmvr.grpc.camera_rgb_stream@1
        with:
          endpoint: cmvr_es
          device_id: wrist_cam
          pixel_format: BGR8
          reconnect: true
          reconnect_initial_s: 0.5
          reconnect_max_s: 10
          stream_log_interval_s: 5

      decoder:
        uses: media.video_decoder.pyav@1

      frame_fanout:
        # One shared fan-out node chooses the latest decoded frame before both
        # blocking detector queues. This keeps the two model inputs identical.
        uses: core.passthrough@1

      detector:
        uses: detection.model@1
        with:
          model: construction-ppe-yolov8@2
          detect_labels:
            - No-Boots
            - No-Ear-Protection
            - No-Glass
            - No-Glove
            - No-Helmet
            - No-Mask
            - No-Vest
          confidence: 0.50
          inference_log_interval_s: 5
          # Required so the viewer receives the exact decoded inference frame.
          attach_frame: true
          model_options:
            # Repository-relative path; launch the viewer from the repository root.
            weights: models/detection/construction-ppe-yolov8/v2/model.onnx
            providers: [CPUExecutionProvider]
            intra_op_threads: 1
            inter_op_threads: 1
            imgsz: 640
            iou: 0.70
            max_det: 100

      phone_detector:
        uses: detection.model@1
        with:
          model: yolov8n-mobile-phone@2
          detect_labels:
            - mobile_phone
          confidence: 0.50
          inference_log_interval_s: 5
          attach_frame: true
          model_options:
            weights: models/detection/yolov8n-mobile-phone/v2/model.onnx
            providers: [CPUExecutionProvider]
            intra_op_threads: 1
            inter_op_threads: 1
            imgsz: 640
            iou: 0.70
            max_det: 100

      viewer:
        uses: demo.opencv_detection_viewer@1
        with:
          window_name: CMVR PPE + Phone Detection
          window_width: 1280
          window_height: 720
          wait_key_ms: 1
          box_thickness: 2
          font_scale: 0.6
          show_stats: true
          # Render only after both results for the exact same ImageFrame arrive.
          expected_inputs: [ppe, phone]
          max_pending_frames: 8

    edges:
      # H264/H265 packets must remain contiguous until decoding.
      - from: camera.frames
        to: decoder.frames
        qos:
          profile: video_contiguous
          capacity: 8
          overflow: block

      # Drop only before the shared fan-out, so both models see the same
      # surviving decoded frame rather than independently dropping frames.
      - from: decoder.frames
        to: frame_fanout.input
        qos:
          profile: realtime_latest
          capacity: 1
          overflow: drop_oldest

      # These two blocking edges preserve the same frame set and ordering for
      # both models. Do not add independent detector max_fps limits here.
      - from: frame_fanout.output
        to: detector.frames
        qos:
          profile: telemetry
          capacity: 1
          overflow: block

      - from: frame_fanout.output
        to: phone_detector.frames
        qos:
          profile: telemetry
          capacity: 1
          overflow: block

      # The viewer joins both result streams by frame identity before drawing.
      - from: detector.detections
        to: viewer.ppe
        qos:
          profile: telemetry
          capacity: 2
          overflow: block

      - from: phone_detector.detections
        to: viewer.phone
        qos:
          profile: telemetry
          capacity: 2
          overflow: block