The events that occur in video usually span multiple frames. For example, a person, object, or logo might appear on screen and remain there for seconds or minutes. Media Server analysis tasks run analysis on single frames, but also follow events across multiple frames.
Analysis tasks can produce the following tracks:
Data tracks contain a record, describing the analysis results, for every frame of an event. For example, a face detection data track contains one record for every video frame in which the face appears. If a person remains in the scene for several seconds, this track could contain hundreds of records that represent the same face.
A data track can contain multiple records for a single frame, if multiple events occur at the same time. For example, if a video frame contains three faces, the face detection data track will contain three records related to that frame. The records for different events, in this example different faces, are assigned a different ID.
Data
and DataWithSource
tracks contain a significant amount of information, often more than you want to output from Media Server. These tracks are most useful as the input for other tasks. For example, you can use the DataWithSource
track from face detection as the input for face recognition, so that face recognition receives the location of each face in every frame and all of the source frames.
SegmentDuration
. When a record reaches the maximum duration, Media Server outputs the record and begins a new one with the same ID. This means that for every record in the result track that exceeds the maximum duration, there will be two or more records in the SegmentedResult
track. Segmented results are useful when you want to output information about an event before it finishes. The SegmentedResult
track can offer better accuracy than the data track, because Media Server can track events across frames and analyze multiple frames before providing the result.SegmentedResult
tracks. The records are the same, except that each record also includes the best source frame that was available at the time the record was generated.The following diagram shows how Media Server creates records (represented by blue rectangles) when a person appears in a video. All of the records in the following example would have the same ID because they all relate to the same detected face (the same event).
Event
track when the person appears and disappears from the scene.Data
and DataWithSource
tracks for each analyzed frame. If there were multiple people in the scene, Media Server would create multiple records in the Data
and DataWithSource
tracks for each frame.Result
and ResultWithSource
tracks for each event (in this example each detected face). These records span the event and summarize the analysis results.SegmentedResult
and SegmentedResultWithSource
tracks. In this example the person remains in the scene longer than the configured SegmentDuration
, so Media Server generates one record when the maximum duration is reached and then begins another.For information about the tracks that are produced by Media Server tasks, and the information contained in each track, refer to the Media Server Reference.
|