AutoMovie API
    Preparing search index...

    Interface IAutoMovieGrammarSubjectObservation

    One subject's deterministic measurements at a shot's two edited boundaries.

    The production film-grammar pass measures each scene node a shot contract requires readable through the same nodeSubjectExtent, nodeSubjectBox and framedBoxOf reads the camera solve uses, so the edit is judged on the box the camera was aimed at. readFilmGrammar throws when the id is blank or repeated in the shot, a point is not finite, height is not finite and positive, radius is negative or not finite, or an eyeline target id is blank.

    interface IAutoMovieGrammarSubjectObservation {
        end: IAutoMovieVector3;
        eyeline:
            | { end: IAutoMovieVector3; start: IAutoMovieVector3; target: string }
            | null;
        height: number;
        id: string;
        radius: number;
        start: IAutoMovieVector3;
    }
    Index

    World bottom-centre point of the subject at the last shown frame.

    eyeline:
        | { end: IAutoMovieVector3; start: IAutoMovieVector3; target: string }
        | null

    Resolved gaze over the shot, or null when it is not observed.

    Type Declaration

    height: number

    World-space subject height in metres; finite and positive.

    id: string

    Scene-node or formation id, non-blank and unique within the shot.

    radius: number

    Half the subject's horizontal diagonal in metres, or 0 when nothing horizontal could be measured.

    The same number IAutoMovieFramedBox.radius states, read here for the reason the framing solve reads it: a subject wider than the frame can hold at its height is placed by its width, so a shot size measured from the height alone would report a looser framing than the camera delivered. In a landscape raster a standing figure's radius decides nothing; a 60 m facade's decides the size.

    World bottom-centre point of the subject at the first shown frame.