Automating Embedded Metadata Extraction to Custom Fields

Help
1

Hi Team,

Love the platform and the API flexibility! However, I’m hitting a roadblock with metadata.

The Issue: Uploaded files contain rich embedded metadata (Shooting Date, Scene, Shot, Director, etc.), but Frame.io doesn’t automatically map these to the UI. Creating Custom Fields works, but they remain empty unless populated manually.

Technical Question: What is the most efficient way to programmatically extract embedded metadata and sync it to Custom Fields?

  1. Does the Frame.io API support native mapping for embedded XMP/sidecar data during ingest?

  2. If not, what is the recommended Automation workflow?

  3. Are there any specific metadata schemas Frame.io prioritizes for auto-population?

Looking for a scalable way to avoid manual entry. Thanks for the great product!

Best,

2

hi @huynhdumkt unfortunately there’s no way to associate a XMP/sidecar file with another asset automatically. As far as Frame.io is concerned they are two separate & independent assets, there’s no link between them (they have two unique asset_ids). If you wanted to apply custom fields to assets in Frame.io, the best place to do that is when the asset is uploaded to Frame.io via a custom script or application.

if you have something like scene / shot / take embedded in the metadata, you could write a script that reads that data, uploads the file to Frame (getting an asset_id from frame), then either creates the metadata that you want and adds it to that file, or just updates the file with the correct information via update metadata across multiple files endpoint.

There might be some 3rd party developers who could do this (Silverstack Labs or ShotPut Pro for example might be able to add metadata to assets when they get to frame, it would be worthwhile to suggest that as a feature request to those teams).