About MetaRex

The Metadata Resource EXpress project is based on an idea that would allow broadcast metadata to persist from the studio to the mastering stage. The goal was to reduce the cost of mastering international versions of content to extend reach and increase the value of dubbing / subtitling & accessibility.

Time passed and virtual production went from a luxury to a necessity during Covid. Despite the need for a media metadata framework, there didn't seem to be one. A hotch-potch of point solutions was created whichworked, but turned every production into a custom-engineering project just to preserve metadata. As a result, a lot of metadata is discarded and re-created later in the production chain. Oftern re-created means "an AI doing some directed guessing". This seems far from optimal and burns a lot more CPU cycles than just remembering what you did in the first place.

MetaRex has only 3 key goal

1. Preserve media metadata
2. Identify media metadata
3. Enable the use of media metadata

Preserve media metadata

Simply put - transport it to where it can be used or stored. Avoid trascoding it into another format if that would involve loss of information

Identify media metadata

Give each type of metadata an ID and give each instance of it an ID. Associate the metadata with some time / date stream where possible.

IDs might be ad-hoc or might be registered if that metadata is likely to be widely used (e.g. identifying lens metadata from major manufacturers.)

Enable the use of media metadata

The needs change for each different type of metadata. The common case is to enable the metadata ID to turn into a human/machine readable decoding tool. For example the id MRX.123.456.789.abc might point to a tool that allows the metadata to be turned into markers on and Avid Media Composer timeline or an OTIO (Open Timeline IO) timeline