Mid-GOP fast channel-change
A system comprising a decoder to receive an encoded video stream including video frames including reference and non-reference frames, and decode the reference frames, a processor to receive a request from an end-user rendering device requiring a first reference frame as part of decoding the encoded video stream of the first channel from the middle of a group of pictures of the encoded video stream, wherein a second reference frame is a reference frame for the first reference frame in the received encoded video stream, an encoder to encode the decoded first reference frame yielding a re-encoded first reference frame so that decoding the re-encoded first reference frame is not dependent upon the second reference frame, and a transmitter to transmit the re-encoded first reference frame to the end-user rendering device. Related apparatus and methods are also described.
1. A server comprising:
a decoder to:
receive an encoded video stream for a first channel, the encoded video stream including a plurality of video frames including a plurality of reference frames and a plurality of non-reference frames; and
decode the reference frames of the received encoded video stream for the first channel yielding the reference frames in decoded format;
a processor to receive a request from an end-user rendering device external to the server, the end-user rendering device requiring a first one of the reference frames as part of decoding the encoded video stream of the first channel from the middle of a group of pictures of the encoded video stream of the first channel, wherein a second one of the reference frames is a reference frame for the first reference frame in the received encoded video stream;
an encoder to encode the decoded first reference frame yielding a re-encoded first reference frame so that decoding the re-encoded first reference frame is not dependent upon the second reference frame; and
a transmitter to transmit the re-encoded first reference frame to the end-user rendering device so that the end-user rendering device can start decoding the encoded video stream for the first channel from the middle of the group of pictures, wherein:
the video frames of the group of pictures of the encoded video stream of the first channel include a plurality of first temporal references starting at zero and incrementing by one such that each of the video frames of the group of pictures has a different one of the first temporal references;
the encoder is operative to provide a new video sequence including a plurality of video frames such that:
the new video sequence includes the re-encoded first reference frame but excludes the second reference frame; and
the video frames of the new video sequence include a plurality of second temporal references so that: a different one of the second temporal references is provided for each of the video frames of the new video sequence; and the second temporal references of the video frames in the new video sequence plus the first temporal references of the video frames of the group of pictures of the encoded video stream of the first channel received by the end-user rendering device cover all the integers from zero to x, inclusive of zero and x, where x is the largest temporal reference of the first temporal references of the video frames in the group of pictures; and
the transmitter is operative to transmit the new video sequence to the end-user rendering device so that the end-user rendering device can start decoding the encoded video stream for the first channel from the middle of the group of pictures.
2. The server according to claim 1 , wherein:
the end-user rendering device requires the first reference frame and a third one of the reference frames as part of decoding the encoded video stream of the first channel from the middle of the group of pictures; and
the encoder is operative to encode the decoded first reference frame and the decoded third reference frame in the new encoded video sequence which includes the re-encoded first reference frame and a re-encoded third reference frame, the new video sequence being encoded so that decoding the new video sequence is not dependent upon the second reference frame.
3. The server according to claim 2 , wherein the encoder is operative to encode the decoded first reference frame and the decoded third reference frame so that in the new encoded video sequence, the first reference frame is a reference frame for the third reference frame.
4. The server according to claim 1 , wherein:
in the received encoded video stream, the first reference frame is not a random access point; and
the re-encoded first reference frame is a random access point.
5. The server according to claim 1 , wherein the processor is operative to determine which of the reference frames is required by the end-user rendering device in order to decode the encoded video stream of the first channel from the middle of the group of pictures.
6. The server according to claim 5 , wherein the request from the end-user rendering device includes a frame reference of a frame from which decoding of the encoded video stream of the first channel received by the end-user rendering device is going to commence.
7. The server according to claim 5 , wherein the request from the end-user rendering device includes a most recently received timestamp in the channel tuned to before the first channel or a current system clock value.
8. The server according to claim 1 , wherein the request from the end-user rendering device includes a list of which one or ones of the reference frames are required by the end-user rendering device in order to decode the encoded video stream of the first channel from the middle of the group of pictures.
9. The server according to claim 1 , wherein the encoder is operative to provide the new video sequence such that at least two of the video frames in the new video sequence include no new video data.
10. The server according to claim 1 , wherein the request relates to the end-user rendering device changing channels to the first channel in the middle of a group of pictures of the encoded video stream of the first channel.
11. A method performed in a server comprising:
receiving an encoded video stream for a first channel, the encoded video stream including a plurality of video frames including a plurality of reference frames and a plurality of non-reference frames;
decoding the reference frames of the encoded video stream for the first channel yielding the reference frames in decoded format;
receiving a request from an end-user rendering device which is external to the server, the end-user rendering device requiring a first one of the reference frames as part of decoding the encoded video stream of the first channel from the middle of a group of pictures of the encoded video stream of the first channel, wherein a second one of the reference frames is a reference frame for the first reference frame in the received encoded video stream;
encoding the decoded first reference frame yielding a re-encoded first reference frame so that decoding the re-encoded first reference frame is not dependent upon the second reference frame; and
transmitting the re-encoded first reference frame to the end-user rendering device so that the end-user rendering device can start decoding the encoding video stream for the first channel from the middle of the group of pictures, wherein:
the video frames of the group of pictures of the encoded video stream of the first channel include a plurality of first temporal references starting at zero and incrementing by one such that each of the video frames of the group of pictures has a different one of the first temporal references;
the encoding including providing a new video sequence including a plurality of video frames such that:
the new video sequence includes the re-encoded first reference frame but excludes the second reference frame; and
the video frames of the new video sequence include a plurality of second temporal references so that: a different one of the second temporal references is provided for each of the video frames of the new video sequence; and the second temporal references of the video frames in the new video sequence plus the first temporal references of the video frames of the group of pictures of the encoded video stream of the first channel received by the end-user rendering device cover all the integers from zero to x, inclusive of zero and x, where x is the largest temporal reference of the first temporal references of the video frames in the group of pictures; and
the transmitting includes transmitting the new video sequence to the end-user rendering device so that the end-user rendering device can start decoding the encoded video stream for the first channel from the middle of the group of pictures.
12. The method according to claim 11 , wherein the first reference frame and a third one of the reference frames is required as part of decoding the encoded video stream of the first channel from the middle of the group of pictures, and further comprising encoding the decoded first reference frame and the decoded third reference frame in the new encoded video sequence which includes the re-encoded first reference frame and a re-encoded third reference frame, the new video sequence being encoded so that decoding the new video sequence is not dependent upon the second reference frame.
13. The method according to claim 12 , wherein encoding the decoded first reference frame and the decoded third reference frame is performed so that in the new encoded video sequence, the first reference frame is a reference frame for the third reference frame.
14. The method according to claim 11 , wherein:
in the received encoded video stream, the first reference frame is not a random access point; and
the re-encoded first reference frame is a random access point.
15. The method according to claim 11 , further comprising determining which of the reference frames is required by the end-user rendering device in order to decode the encoded video stream of the first channel from the middle of the group of pictures.
16. The method according to claim 15 , wherein the request from the end-user rendering device includes a frame reference of a frame from which decoding of the encoded video stream of the first channel received by the end-user rendering device is going to commence.
17. The method according to claim 15 , wherein the request from the end-user rendering device includes a most recently received timestamp in the channel tuned to before the first channel or a current system clock value.
18. The method according to claim 11 , wherein the request from the end-user rendering device includes a list of which one or ones of the reference frames are required by the end-user rendering device in order to decode the encoded video stream of the first channel from the middle of the group of pictures.
19. The method according to claim 11 , further comprising providing the new video sequence such that at least two of the video frames in the new video sequence include no new video data.
20. The method according to claim 11 , wherein the request relates to the end-user rendering device changing channels to the first channel in the middle of a group of pictures of the encoded video stream of the first channel.