Fluent Help Center

    Calibrate a device for Video Analytics

    You can calibrate certain video devices for Video Analytics using the Customer Website, or the Fluent app. The calibration process involves taking five still images of a person or object in different locations visible to the camera.

    Important: The video device should be re-calibrated whenever it is moved to another location.

    The following video devices require calibration before a user can create a Video Analytics recording rule. For more information about creating a Video Analytics rule, see Configure a Video Analytics rule.

    Video devices that require calibration include:

    • 1080p Outdoor Wi-Fi Camera (ADC-V723)
    • 1080p Indoor Wi-Fi Camera (ADC-V523)

    Calibrate the camera

    Users can calibrate their cameras using the MobileTech app, the Customer app, or the Customer Website. Using a mobile device to complete the calibration is recommended because the user can take the calibration images while standing where they need to be.

    Open the calibration page
    To open the calibration page using the Fluent app:
    1. Log into the Fluent app.
    2. Tap Menu.
    3. Tap Video.
    4. Tap Configure.
    5. Tap Calibrate Cameras.
    6. Tap > next to the device to calibrate.
    To open the calibration page using the Customer Website:
    1. Log into the Customer Website.
    2. Click Video.
    3. On the Video banner, click Settings.
    4. Using the Video Device dropdown menu, select the video device to calibrate.
    5. Click Video Analytics Calibration.
    Complete the camera calibration

    Important: If this is the first time the camera is being set up for Video Analytics, the camera's existing Video Motion Detection (VMD) rules will be removed, but they can be added back later.

    1. On the Camera Calibration page, click Continue.

      Camera Calibration tips.PNG
       
    2. Click Take Photo.

      Take photo.png 
    3. Verify that the bounding boxes are as close around the object as possible, and the entire object is in view.
      1. If the snapshot needs to be retaken, click Retake Photo.

        snapshot step.PNG
        • A person's entire body must be in view of the camera, head to toe.
        • Each snapshot should be as different as possible to ensure good calibration.
      2. Edit the bounding box to exactly fit around the detected body.
    4. Click Next.
    5. After five images are captured, click Submit.
    6. Once the calibration is complete, click Done.

      success page.PNG 
    7. If the calibration is unsuccessful, review the snapshots and correct them as necessary.

    Did this help resolve your question?

    Send feedback

    Can’t find what you’re looking for?

    Our award-winning customer care team is here for you.

    Contact Support