Adds support for adding properties relevant to images based derived using machine learning. Whenever a media item is saved it will have properties set on the media items if they are present on the media type. Will work on any media item with an upload property called umbracoFile. Currently Azure Computer Vision API and Google's Vision API are both supported.
To install via nuget:
Either
Install-Package Our.Umbraco.IntelligentMedia.Google
For Google Vision API or
Install-Package Our.Umbraco.IntelligentMedia.Azure
for Azure Vision API, or both to combine data from both
You'll need to configure AppSettings:
If you want to override the media's name property with one generated by the Vision API:
<add key="IntelligentMedia:OverwriteName" value="true" />For Google you'll need the following:
<add key="IntelligentMedia:Google:ApiKey" value="" />You'll need to insert your Google Vision API key
For Azure these:
<add key="IntelligentMedia:Azure:SubscriptionKey" value="" />
<add key="IntelligentMedia:Azure:Region" value="westeurope" />You'll need to insert your Azure Computer Vision Subscription key and region
For the data to be populated on the media you'll need to add properties to the media types.
On the image media type: Add a new tab called "AI" Add to that tab the following properties:
| Name | Alias | Editor |
|---|---|---|
| Tags | imTags | Tags (CSV) |
| Description | imDescription | Textarea |
| Categories | imCategories | Tags (CSV) |
| Number of Faces | imNumberOfFaces | Numeric |
| Primary Colour | imPrimaryColour | Textstring |
| Background Colour | imBackgroundColour | Textstring |
| Populated | imPopulated | Checkbox |
Icons made by Kiranshastry from www.flaticon.com