AutoMovie API
    Preparing search index...

    Interface IAutoMovieExpression

    A facial expression at one instant: preset, intensity and a required nullable ARKit override list.

    Preset and intensity supply coarse semantic control; blendshapes overrides expanded ARKit weights. Visible response requires matching morph targets or a host-supplied expression sink, so the names alone do not establish portability across every avatar or equivalent appearance.

    validateExpression refuses unknown names, nonfinite or out-of-range [0, 1] magnitudes and repeated channels when invoked. The viewer separately clamps finite values to [0, 1] and treats nonfinite values as zero.

    sampleMotion blends same-preset intensity and channel weights using outgoing eased progress, switching differing presets when that progress reaches 0.5. An absent side is neutral intensity zero with no channels. Cubic y controls can overshoot between admitted keys; the sampler does not independently gate expression magnitudes there.

    interface IAutoMovieExpression {
        blendshapes: IAutoMovieBlendshapeChannel[] | null;
        intensity: number;
        preset: AutoMovieExpressionPreset;
    }
    Index
    blendshapes: IAutoMovieBlendshapeChannel[] | null

    ARKit weights replacing expanded preset contributions for the same channel, or null when there are no overrides; each channel at most once.

    intensity: number

    How strongly preset applies, in [0, 1].

    Coarse emotion, viseme or eye intent.