Auto Recognize

This section describes features that are coming in 4.0

QName: f:auto-recognize

With this feature in place, a content instance will automatically have its default binary attachment processed through recognition services to detect and discover meaning contained in the underlying image. The recognition process will detect the following kinds of elements:

  • Faces
  • Celebrities
  • Labels
  • Text
  • Moderation Labels (bad language, adult content)

To use this service, you will first need to set up a Recognition Service. The service can either be configured as the default Recognition Service for your Project or it can be referenced from the f:auto-recognize feature. This allows different feature configurations to use different services if that is your preference.

The feature is defined as shown here:

{
    "_qname": "f:auto-recognize",
    "_type": "d:feature",
    "type": "object",
    "title": "Auto Recognize",
    "description": "Indicates that a node should have recognition automatically applied to an attachment",
    "systemBootstrapped": true
}

Configuration

<thead>
    <tr>
        <th>Property</th>
        <th>Type</th>
        <th>Default</th>
        <th nowrap>Read-Only</th>
        <th>Description</th>
    </tr>
</thead>
<tbody>
    <tr>
        <td>serviceId</td>
        <td>text</td>
        <td></td>
        <td></td>
        <td>
            The `_doc` identifier of the Recognition Service configuration to use.
        </td>
    </tr>
</tbody>

Image Recognition

With this feature in place, an image recognition process will launch whenever you create or update the default attachment for the node. The image recognition process will analyze your binary content and detect any of the following structured elements:

  • Faces - the positional coordinates and analysis of a face within the image (such as whether the face is Female or Male, Happy or Sad, etc)
  • Celebrities - the positional coordinates and analysis of a celebrity's face within the image (including who the celebrity is)
  • Labels - the positional coordinates and analysis of any identified objects or elements within the image (such as a Toothbrush or a Hat)
  • Text - the positional coordinates of any identified text within the page
  • Moderation Labels - an analysis of the moderation characteristics of the image (such as whether it has Nudity or Suggestive Sexual elements)

The extracted content is then stored onto your original Node so that you can work with it from a post-processing perspective. In addition, the Cloud CMS user interface makes use of this extracted content to provide your editorial users with in-context insights and awareness of what was extracted.

For more information, please see our formal documentation on how Cloud CMS works with Recognition.

Example

If you have a default Recognition Service configured for your project, all you need to do is this:

{
    "title": "My Document",
    "_features": {
        "f:auto-recognize": {
        }
    }
}

Otherwise, you will need to point to a configuration:

{
    "title": "My Document",
    "_features": {
        "f:auto-recognize": {
            "serviceId": "ec0a7bfb2d2f971b262e"
        }
    }
}