Open navigation

Smart Tags - Setting Up for Images, Videos and Text


Restriction: You need to be signed in using an administrator's or contributor's account in order to set up smart tags.


Settings > Configuration Options > Recognition > Images / Videos / Text


The Smart Tags feature uses AWS Rekognition to recognize objects, properties, and details in images and creates tags that are automatically assigned to the assets.

For more information and technical details about the AWS Rekognition Service,
see https://aws.amazon.com/rekognition.
Note: If the Smart Tags settings are not displayed, contact your Canto Account Manager!


How to Set up Smart Tags for Images

To set up Smart Tags:

  1. Select the Setup smart tags service to detect objects and scenes in images and/or Setup Facial Recognition to identify people in images.
  2. Select a value between 0 and 99% for Min Confidence.
    This value indicates the reliability of the recognition. Canto will not accept tags or recognized faces with a probability lower than the specified value. The lower this value is, the more (but possibly less relevant) smart tags will be provided.
  3. Select a value for the maximum number of tags that will be provided (maximum 30) and/or for the maximum number of people that will be recognized (maximum 50).
  4. Click Save to apply the settings.
    You also have the option to Recreate Smart Tags (your entire library would be assigned with smart tags and old entries will be overwritten) and/or Reidentify Faces (this would overwrite all already identified faces).


How to Deactivate Smart Tags During Uploads

When the Smart Tags feature is enabled, smart tags are added to new assets by default when they are uploaded. However, you can turn off the smart tagging process during upload, either to speed up the upload process or simply because you don't need smart tags for some of your assets.

  1. In the Add Metadata dialog that appears when you upload images to your library scroll down until you see the Smart Tags for Images section.
  2. Uncheck the box.


If you do not see the option above in your metadata window upon upload,  you will need to enable this feature by going into Settings > Configuration Options > Custom Configuration > Metadata Editor (Upload) > add Smart Tags for Images


How to Set up Smart Tags for Videos

  1. Check Setup smart tags service to detect objects and scenes in videos and/or Setup Facial Recognition for videos and/or Setup Video Transcription service to automatically create subtitles in a video.
  2. Select a value between 0 and 99% for Min. Confidence.
    This value indicates the reliability of the recognition. Canto will not provide tags or recognized people with a probability lower than the specified value. The lower this value is, the more (but possibly less relevant) smart tags will be provided.
  3. Select a value for the maximum number of tags that will be delivered (maximum 30) or for the maximum number of people that will be detected (maximum 50).
  4. Click Save to apply the settings.

The time the recognition software needs for the process depends on the length of the video.


Similar to Smart Tags for Images you can control the Confidence Score and control the Maximum Tag amount for each video. The length of the video will affect the time it takes for the recognition software to process.


For video recognition, you will have a preset number of hours of content you can identify. You can view the number of hours you have and the number of hours you have used. The number of hours can be adjusted after reaching out to your account manager. Your hour usage is combined with the Facial Recognition feature. 


How to Set up Smart Tags for Text

You can activate Text Recognition and/or Optical Character Recognition.

Text Recognition:

  1. Check Setup text recognition to identify text in an image.
  2. Choose between Read text in an image to the Canto Description Field and Read text in an image to the Canto Tags Field.

Optical Character Recognition (OCR):

  1. Check Setup Optical Character Recognition (OCR) for scanned PDFs.

Did you find it helpful? Yes No

Send feedback
Sorry we couldn't be helpful. Help us improve this article with your feedback.