Colorfront has integrated Transkoder with Qumulo NeuralSearch, writing on-set metadata into media files so it stays searchable through every copy and archive move.
Colorfront and Qumulo have announced a joint workflow that writes on-set metadata directly into media files, keeping camera, lens, color and framing information searchable through every copy, move and archive tier.
As anyone working in post knows, there is an increasing amount of metadata to deal with. What’s worse is that it ends up scattered across different places too. That means everything from camera make and model, lens, focal length, aperture, color intent as ASC CDL values and LUTs, framing intent, timecode, script supervisor notes and VFX flags, can be found across camera headers, sidecar files, ALEs, EDLs and spreadsheets.
That makes each hand-off in the process an opportunity for metadata to be lost. And, as Colorfront points out, hunting the right files down as a result can snowball into significant time and financial costs over the length of a production.
Its solution is to write the metadata to the file, then make the file searchable via an integration with Qumulo NeuralSearch. This is storage-native search that indexes file metadata continuously as data lands, with no external crawler and no separate search infrastructure. It also answers SQL, natural-language, and semantic queries directly against the filesystem.
It’s a neat system in itself and Colorfront Transkoder and On-Set Dailies are well placed to action it, as they tend to sit at the point in the pipeline where metadata is at its richest. The joint workflow is in active development and has been shown as a preview at the show. No word yet on when it will be finalized, but it’s definitely one to keep an eye out for.