It’s impossible to avoid human input when classifying media documents. It’s possible for text and audio files, but not for images and videos. Daminion can extract all the technical information from the files, but you will need to manually describe the content of your pictures and videos:
- Who is showing in a picture
- What is showing in a picture
- Where was picture taken (This can be partially solved by using GPS info)
- What place is shown in a picture?
- What event is depicted in the picture?
- Who are the document creators (third-party images often come with some copyright info on IPTC that can be easily detected and imported by Daminion)
Using tags drastically reduces the time needed to classify documents. Daminion includes about 30 predefined tags that can be shown on the tag panel; that is usually enough in most cases. Some tags examples are: Creation Date/time, File Folder, Author, People, Event, Keywords, Categories, Featured Place, Rating, Color Label, Camera, Lens, and Media Format. A bit later, we’ll add the ability to create and assign custom tags. For example, tags for a Construction Company could be: Device, Department, System, and Chassis.
|The “Shown Place” tag is composite and contains 4 named levels:*
Each level is mapped to separate IPTC/XMP metadata field. * We’ll probably add a 5th level named “Region” according to XMP specification.
Below is the prototype of the Quick Assigner window that is not yet implemented. It will provide quick access to recently assigned tags.