Human face detection and tracking.

Unspecified.

Label detection.

Safe search detection.

Shot change detection.

Detect frame-level labels.

Unspecified.

Detect both shot-level and frame-level labels.

Detect shot-level labels.

Frame-level.

Unspecified.

Segment-level.

Shot-level.

Video-level.

Likely.

Possible.

Unknown likelihood.

Unlikely.

Very likely.

Very unlikely.

Video annotation progress.

Video annotation request.

Video annotation response.

Bounding box.

Face annotation.

Face location.

Label annotation.

Label location.

Safe search annotation (based on per-frame visual signals only).

Annotation progress for a single video.

Annotation results for a single video.

Video context and/or feature-specific parameters.

Video segment.

Video annotation feature.

Label detection mode.

Label level (scope).

Bucketized representation of likelihood.