A number of new capabilities (including Face Detection and Background blur acceleration) appear to require the addition of VideoFrame metadata:
Examples: w3c/mediacapture-extensions#57
Eye gaze correction: w3c/mediacapture-extensions#56
Face framing: w3c/mediacapture-extensions#55
Lighting correction: w3c/mediacapture-extensions#53
There are also uses for encoded chunk metadata, however, there is no registry for this metadata, nor are there any rules for carrying VideoFrame metadata over to encoded chunks (e.g. in WebCodecs encoder) or vice-versa (in WebCodecs decoder).
Related: w3c/webcodecs#559