[HW Accel Support]: New Coral install on NUC running HAOS

[RCG-SW]

Describe the problem you are having

Frigate crashes trying to detect USB Coral

2024-09-01 02:50:31.384884989  [2024-08-31 20:50:28] frigate.detectors.plugins.edgetpu_tfl INFO    : Attempting to load TPU as auto
2024-09-01 02:50:31.391335138  Process detector:coral:
2024-09-01 02:50:31.391338580  Traceback (most recent call last):
2024-09-01 02:50:31.391339899    File "/usr/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
2024-09-01 02:50:31.391340839      self.run()
2024-09-01 02:50:31.391341932    File "/usr/lib/python3.9/multiprocessing/process.py", line 108, in run
2024-09-01 02:50:31.391342991      self._target(*self._args, **self._kwargs)
2024-09-01 02:50:31.391344070    File "/opt/frigate/frigate/object_detection.py", line 102, in run_detector
2024-09-01 02:50:31.391347636      object_detector = LocalObjectDetector(detector_config=detector_config)
2024-09-01 02:50:31.391366380    File "/opt/frigate/frigate/object_detection.py", line 53, in __init__
2024-09-01 02:50:31.391367711      self.detect_api = create_detector(detector_config)
2024-09-01 02:50:31.391368735    File "/opt/frigate/frigate/detectors/__init__.py", line 18, in create_detector
2024-09-01 02:50:31.391389365      return api(detector_config)
2024-09-01 02:50:31.391390587    File "/opt/frigate/frigate/detectors/plugins/edgetpu_tfl.py", line 43, in __init__
2024-09-01 02:50:31.391391996      self.interpreter = Interpreter(
2024-09-01 02:50:31.391393071    File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 348, in __init__
2024-09-01 02:50:31.391404673      _interpreter_wrapper.CreateWrapperFromFile(
2024-09-01 02:50:31.391405759  ValueError: Model provided has model identifier 'l ve', should be 'TFL3'
2024-09-01 02:50:31.391406587  
2024-09-01 02:50:31.391651885  [2024-08-31 20:50:31] frigate.detectors.plugins.edgetpu_tfl INFO    : TPU found
2024-09-01 02:50:31.391817218  [2024-08-31 20:50:31] frigate.detectors.plugins.edgetpu_tfl ERROR   : No EdgeTPU was detected. If you do not have a Coral device yet, you must configure CPU detectors.

Version

Frigate (Full Access) Current version: 0.14.1

In which browser(s) are you experiencing the issue with?

N/A

Frigate config file

mqtt:
  enabled: true
  host: lestoucans.local
  port: 1883
  user: xxx
  password: xxx

ffmpeg:
  hwaccel_args: preset-vaapi

birdseye:
  enabled: false
  width: 1280
  height: 720

detect:
  # Optional: width of the frame for the input with the detect role (default: use native stream resolution)
  width: 1280
  # Optional: height of the frame for the input with the detect role (default: use native stream resolution)
  height: 720
  # Optional: desired fps for your camera for the input with the detect role (default: shown below)
  # NOTE: Recommended value of 5. Ideally, try and reduce your FPS on the camera.
  fps: 3
  # Optional: enables detection for the camera (default: True)
  enabled: true
  # Optional: Number of consecutive detection hits required for an object to be initialized in the tracker. (default: 1/2 the frame rate)
  min_initialized: 2
  # Optional: Number of frames without a detection before Frigate considers an object to be gone. (default: 5x the frame rate)
  max_disappeared: 25
  # Optional: Configuration for stationary object tracking
  stationary:
    # Optional: Frequency for confirming stationary objects (default: same as threshold)
    # When set to 1, object detection will run to confirm the object still exists on every frame.
    # If set to 10, object detection will run to confirm the object still exists on every 10th frame.
    interval: 50
    # Optional: Number of frames without a position change for an object to be considered stationary (default: 10x the frame rate or 10s)
    threshold: 50
    # Optional: Define a maximum number of frames for tracking a stationary object (default: not set, track forever)
    # This can help with false positives for objects that should only be stationary for a limited amount of time.
    # It can also be used to disable stationary object tracking. For example, you may want to set a value for person, but leave
    # car at the default.
    # WARNING: Setting these values overrides default behavior and disables stationary object tracking.
    #          There are very few situations where you would want it disabled. It is NOT recommended to
    #          copy these values from the example config into your config unless you know they are needed.
    max_frames:
      # Optional: Default for all object types (default: not set, track forever)
      default: 3000
      # Optional: Object specific values
      objects:
        person: 1000
  # Optional: Milliseconds to offset detect annotations by (default: shown below).
  # There can often be latency between a recording and the detect process,
  # especially when using separate streams for detect and record.
  # Use this setting to make the timeline bounding boxes more closely align
  # with the recording. The value can be positive or negative.
  # TIP: Imagine there is an event clip with a person walking from left to right.
  #      If the event timeline bounding box is consistently to the left of the person
  #      then the value should be decreased. Similarly, if a person is walking from
  #      left to right and the bounding box is consistently ahead of the person
  #      then the value should be increased.
  # TIP: This offset is dynamic so you can change the value and it will update existing
  #      events, this makes it easy to tune.
  # WARNING: Fast moving objects will likely not have the bounding box align.
  annotation_offset: 0


objects:
  # Optional: list of objects to track from labelmap.txt (default: shown below)
  track:
    - person
    - car
    - dog
    # - cat
    # - sheep
live:
  # Optional: Set the name of the stream that should be used for live view
  # in frigate WebUI. (default: name of camera)
  stream_name: camera_name
  # Optional: Set the height of the jsmpeg stream. (default: 720)
  # This must be less than or equal to the height of the detect stream. Lower resolutions
  # reduce bandwidth required for viewing the jsmpeg stream. Width is computed to match known aspect ratio.
  height: 720
  # Optional: Set the encode quality of the jsmpeg stream (default: shown below)
  # 1 is the highest quality, and 31 is the lowest. Lower quality feeds utilize less CPU resources.
  quality: 20

detectors:
  coral:
    type: edgetpu
    # device: usb
  ov:
    type: openvino
    device: GPU
    # model:
    #   path: /openvino-model/ssdlite_mobilenet_v2.xml

model:
  width: 300
  height: 300
  input_tensor: nhwc
  input_pixel_format: bgr
  labelmap_path: /openvino-model/coco_91cl_bkgr.txt
  path: /openvino-model/ssdlite_mobilenet_v2.xml

go2rtc:
  streams:
    Barn1: rtsp://xxx:[email protected]:554/cam/realmonitor?channel=1&subtype=00
    Driveway: rtsp://xxx:[email protected]:554/cam/realmonitor?channel=1&subtype=00

cameras:
  Barn1: # <--- this will be changed to your actual camera later
    enabled: true
    ffmpeg:
      inputs:
        - path: rtsp://xxx:[email protected]:554/cam/realmonitor?channel=1&subtype=00
          roles:
            - record
    onvif:
      host: 192.168.1.223
      port: 80
      user: xxx
      password: xxx
    #   autotracking:
    #     # Optional: enable/disable object autotracking. (default: shown below)
    #     enabled: true
    #     # Optional: calibrate the camera on startup (default: shown below)
    #     # A calibration will move the PTZ in increments and measure the time it takes to move.
    #     # The results are used to help estimate the position of tracked objects after a camera move.
    #     # Frigate will update your config file automatically after a calibration with
    #     # a "movement_weights" entry for the camera. You should then set calibrate_on_startup to False.
    #     calibrate_on_startup: true
    #     # Optional: the mode to use for zooming in/out on objects during autotracking. (default: shown below)
    #     # Available options are: disabled, absolute, and relative
    #     #   disabled - don't zoom in/out on autotracked objects, use pan/tilt only
    #     #   absolute - use absolute zooming (supported by most PTZ capable cameras)
    #     #   relative - use relative zooming (not supported on all PTZs, but makes concurrent pan/tilt/zoom movements)
    #     zooming: disabled
    #     # # Optional: A value to change the behavior of zooming on autotracked objects. (default: shown below)
    #     # # A lower value will keep more of the scene in view around a tracked object.
    #     # # A higher value will zoom in more on a tracked object, but Frigate may lose tracking more quickly.
    #     # # The value should be between 0.1 and 0.75
    #     # zoom_factor: 0.3
    #     # # Optional: list of objects to track from labelmap.txt (default: shown below)
    #     track:
    #       - person
    #       - sheep
    #     # Required: Begin automatically tracking an object when it enters any of the listed zones.
    #     required_zones:
    #       - barn_inside
    #     # Required: Name of ONVIF preset in camera's firmware to return to when tracking is over. (default: shown below)
    #     return_preset: Home
    #     # Optional: Seconds to delay before returning to preset. (default: shown below)
    #     timeout: 10
    #     # Optional: Values generated automatically by a camera calibration. Do not modify these manually. (default: shown below)
    #     # movement_weights: []
    zones:
      barn_inside:
        coordinates: 0,720,1280,720,1280,0,0,0
    motion:
      threshold: 30
      contour_area: 20
      # improve_contrast: true
  Driveway: # <--- this will be changed to your actual camera later
    enabled: true
    detect:
      enabled: true
      width: 1280
      height: 720
      fps: 5
    ffmpeg:
      inputs:
        - path: rtsp://xxx:[email protected]:554/cam/realmonitor?channel=1&subtype=00
          roles:
            - detect
            - record
    record:
      enabled: true
      retain:
        days: 2
        mode: motion
      events:
        retain:
          default: 30
          mode: motion
    motion:
      threshold: 30
      contour_area: 25
      # improve_contrast: 'true'
    zones:
      Main:
        coordinates: 
          0,0,0.678,0,0.686,0.191,1,0.452,1,1,0.328,0.995,0.384,0.733,0.271,0.723,0.139,1,0,1
        loitering_time: 0
        inertia: 3
    review:
      alerts:
        required_zones: Main
      detections:
        required_zones: Main

  Barn: # <--- this will be changed to your actual camera later
    enabled: true
    detect:
      enabled: true
      width: 1280
      height: 720
      fps: 5
    ffmpeg:
      inputs:
        - path: rtsp://xxx:[email protected]:554/cam/realmonitor?channel=1&subtype=00
          roles:
            - detect
            - record
    record:
      enabled: true
      retain:
        days: 2
        mode: motion
      events:
        retain:
          default: 30
          mode: motion
    motion:
      threshold: 33
      contour_area: 17
      # improve_contrast: true
    zones:
      Main:
        coordinates: 0,0,0.749,0,0.752,0.098,1,0.09,1,1,0.328,0.995,0.367,0.338,0.152,0.358,0,0.89
        loitering_time: 0
        inertia: 3
    review:
      alerts:
        required_zones: Main
      detections:
        required_zones: Main
  Barn2: # <--- this will be changed to your actual camera later
    enabled: true
    ffmpeg:
      inputs:
        - path: rtsp://xxx:[email protected]:554/cam/realmonitor?channel=1&subtype=00
          roles:
            - record
    onvif:
      host: 192.168.1.157
      port: 80
      user: xxx
      password: xxx
    detect:
      enabled: false
    record:
      enabled: false
      retain:
        days: 0
        mode: motion
      events:
        retain:
          default: 2
          mode: motion
    motion:
      threshold: 38
      contour_area: 61
      # improve_contrast: 'true'
  Barn3: # <--- this will be changed to your actual camera later
    enabled: true
    ffmpeg:
      inputs:
        - path: rtsp://xxx:[email protected]:554/cam/realmonitor?channel=1&subtype=00
          roles:
            - record
    onvif:
      host: 192.168.1.159
      port: 80
      user: xxx
      password: xxx
    detect:
      enabled: false
    record:
      enabled: false
      retain:
        days: 1
        mode: motion
      events:
        retain:
          default: 14
          mode: motion
  Barn4: # <--- this will be changed to your actual camera later
    enabled: true
    detect:
      enabled: false
    ffmpeg:
      inputs:
        - path: rtsp://xxx:[email protected]:554/cam/realmonitor?channel=1&subtype=00
          roles:
            - record
    onvif:
      host: 192.168.1.161
      port: 80
      user: xxx
      password: xxx
    record:
      enabled: false
      retain:
        days: 1
        mode: motion
      events:
        retain:
          default: 14
          mode: motion
  nw: # <--- this will be changed to your actual camera later
    enabled: true
    ffmpeg:
      inputs:
        - path: rtsp://xxx:[email protected]:554/cam/realmonitor?channel=1&subtype=00
          roles:
            - record
    onvif:
      host: 192.168.1.69
      port: 80
      user: xxx
      password: xxx
    detect:
      enabled: false
    record:
      enabled: false
      retain:
        days: 1
        mode: motion
      events:
        retain:
          default: 14
          mode: motion

docker-compose file or Docker CLI command

N/A

Relevant Frigate log output

2024-09-01 02:50:31.384884989  [2024-08-31 20:50:28] frigate.detectors.plugins.edgetpu_tfl INFO    : Attempting to load TPU as auto 2024-09-01 02:50:31.391335138  Process detector:coral: 2024-09-01 02:50:31.391338580  Traceback (most recent call last): 2024-09-01 02:50:31.391339899    File "/usr/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap 2024-09-01 02:50:31.391340839      self.run() 2024-09-01 02:50:31.391341932    File "/usr/lib/python3.9/multiprocessing/process.py", line 108, in run 2024-09-01 02:50:31.391342991      self._target(*self._args, **self._kwargs) 2024-09-01 02:50:31.391344070    File "/opt/frigate/frigate/object_detection.py", line 102, in run_detector 2024-09-01 02:50:31.391347636      object_detector = LocalObjectDetector(detector_config=detector_config) 2024-09-01 02:50:31.391366380    File "/opt/frigate/frigate/object_detection.py", line 53, in __init__ 2024-09-01 02:50:31.391367711      self.detect_api = create_detector(detector_config) 2024-09-01 02:50:31.391368735    File "/opt/frigate/frigate/detectors/__init__.py", line 18, in create_detector 2024-09-01 02:50:31.391389365      return api(detector_config) 2024-09-01 02:50:31.391390587    File "/opt/frigate/frigate/detectors/plugins/edgetpu_tfl.py", line 43, in __init__ 2024-09-01 02:50:31.391391996      self.interpreter = Interpreter( 2024-09-01 02:50:31.391393071    File "/usr/lib/python3/dist-packages/tflite_runtime/interpreter.py", line 348, in __init__ 2024-09-01 02:50:31.391404673      _interpreter_wrapper.CreateWrapperFromFile( 2024-09-01 02:50:31.391405759  ValueError: Model provided has model identifier 'l ve', should be 'TFL3' 2024-09-01 02:50:31.391406587   2024-09-01 02:50:31.391651885  [2024-08-31 20:50:31] frigate.detectors.plugins.edgetpu_tfl INFO    : TPU found 2024-09-01 02:50:31.391817218  [2024-08-31 20:50:31] frigate.detectors.plugins.edgetpu_tfl ERROR   : No EdgeTPU was detected. If you do not have a Coral device yet, you must configure CPU detectors.

Relevant go2rtc log output

N/A

FFprobe output from your camera

N/A

Operating system

HassOS

Install method

HassOS Addon

Network connection

Wired

Camera make and model

Amcrest

Screenshots of the Frigate UI's System metrics pages

Firgate crashes trying to detect

Any other information that may be helpful

No response

[NickM-27]

You can't mix and match openvino and coral