AI metadata detection
Tagging large amounts of media by hand takes time, and the result is rarely consistent. With AI metadata detection, MediaLab analyzes your media and fills in metadata automatically, so files are easier to find and organize from the moment they are uploaded.
MediaLab offers two options:
- AWS Rekognition: ready-to-use image analysis.
- Enterprise AI metadata workflows: a workflow built on demand around your own AI models, vocabularies or metadata structure.
Both are available on request. Contact us to enable AWS Rekognition for your account or to discuss an enterprise workflow.
AWS Rekognition
AWS Rekognition analyzes uploaded images and detects labels (objects, scenes and concepts), text, celebrities, face attributes and potentially unsafe content. You choose which results are saved in your metadata, and administrators can fine-tune what is detected.
See AWS Rekognition image analysis for everything it detects and how to set it up.
Enterprise AI metadata workflows
When an off-the-shelf analysis isn't enough, we build an AI metadata workflow around your requirements, for example:
- Analysis of video and audio as well as images.
- Your own AI models or AI providers.
- Your own vocabularies, taxonomies and metadata fields, so results match the way your teams search.
- Detection rules that combine AI results with existing metadata, followed by automated actions such as review steps or delivery.
These workflows are set up on demand. Contact us to discuss your use case.
Good to know
- AI detection fills in metadata automatically, but it is not perfect. Users with edit rights can correct or remove detected values like any other metadata.
- Recognizing your own people by name is a separate feature: see Face detection.