Error recovery for video delivery via a segmentation process
A client device may receive encoded video via a transport stream based on a video coding protocol. When errors result in receiving the encoded video, the client device may use an Internet Protocol (IP) connection to recover from the error. For example, an encoder may insert markers inband in the transport stream, and a segmenter then segments the video using the markers. The content remains in the form of a continuous transport stream that is compatible with existing transport stream delivery mechanisms. When an error occurs, the client device can then determine a locator for a segment that can be used to recover from the error and requests the segment from a server through the IP connection. The server sends the segment to the client device at the level of the transport stream layer without adding another protocol layer to encapsulate the segment.
1. A method comprising:
receiving, by a computing device, configuration information describing markers that delineate segments of a video;
generating, by the computing device, an encoded version of the video;
inserting, by the computing device, the markers inband in the encoded version of the video in a transport stream;
sending, by the computing device, the transport stream including the encoded version of the video to a server, wherein the server segments the encoded version of the video based on the markers and identifies the segments; and
sending, by the computing device, the transport stream including the encoded version of the video to a client device at a transport layer, wherein the client device requests a segment of the encoded version of the video from the server using an identifier for the segment upon determining an error with the transport stream, and receives the segment at the transport layer from the server,
wherein the configuration information includes a locator descriptor specifying a location of a recovery presentation description that describes a naming convention for the identifier of the segment, the method further comprising:
including the locator descriptor in a mapping table wherein the mapping table includes a frequency descriptor that describes a frequency of the markers to define a segment size; and
sending the mapping table to the client device.
2. The method of claim 1 , wherein the markers are encoder boundary point markers inserted inband with the encoded version of the video in the transport stream.
3. The method of claim 1 , wherein each marker includes a timestamp for a time in the video.
4. The method of claim 1 , wherein the locator descriptor specifies which markers should be used to generate the segments.
5. The method of claim 1 , wherein the locator descriptor specifies a universal resource locator (URL) for requesting the recovery presentation description.
6. The method of claim 1 , wherein the client requests the recovery presentation description from the server and uses the recovery presentation description to generate the identifier for the segment with a timestamp associated with a marker for the segment.
7. A method comprising:
receiving, by a computing device, a transport stream of an encoded version of a video at a transport layer, the transport stream also being sent to a server; determining, by the computing device, markers that are located inband in the encoded version of the video in the transport stream;
determining, by the computing device, an error associated with a segment in the encoded version of the video, wherein the segment is identified with a marker located inband in the transport stream;
generating, by the computing device, a locator for the segment based on the marker;
sending, by the computing device, a request for the segment using the locator to the server;
receiving, by the computing device, the segment from the server via the transport layer where the server segmented the transport stream into segments based on the markers; and
receiving a mapping table that includes a frequency descriptor that describes a frequency of the markers to define a segment size.
8. The method of claim 7 , wherein: the encoded version of the video is received via a broadcast delivery network, and the segment is received via an IP-based network.
9. The method of claim 7 , further comprising: receiving a mapping table that includes a locator descriptor specifying a location of a recovery presentation description that describes a naming convention for locators of segments for the encoded version of the video.
10. The method of claim 9 , further comprising sending a request to the server for the recovery presentation description, wherein the naming convention is used to generate the locator for the segment along with a time stamp for the marker.
11. The method of claim 7 , further comprising:
buffering a set of segments from the encoded version of the video; and
adding to the segment to the buffered set of segments to recover from the error.
12. The method of claim 7 , wherein the error is detected based on a marker not being received in the encoded version of video.
13. The method of claim 7 , wherein the locator is based on the naming convention and a timestamp for the segment.