Personal video recorder

1. A method of creating a bitstream in which a sequence of picture slices is encoded as reference slices and non-reference slices, wherein the reference slices include a plurality of B slices, said method comprising:

for each B slice, forming at least one data packet containing data values derived from brightness and color information pertaining to the slice, and

for each B slice, including in the at least one data packet a header element indicating whether a B slice is a reference slice,

said header element containing a value that indicates where in a hierarchy of discardability the B slice is,

said hierarchy of discardability comprising a ranking of the plurality of B slices by how much degradation to the sequence of picture slices will be experienced by a discarding of each B slice.

2. The method according to claim 1 , wherein, for a picture that includes at least two reference B slices, the method comprises setting the header element of each reference B slice to the same value.

3. The method according to claim 2 , further comprising decoding the bitstream and selectively discarding pictures that contain reference B slices in accordance with the value of the header element of the or each reference B slice.

4. The method according to claim 1 , comprising encoding a sequence of pictures each containing at least one picture slice, including each reference B slice in a reference B picture, and setting the header elements of all reference B slices in a given reference B picture to the same value.

5. The method according to claim 1 , comprising encoding a sequence of pictures each containing at least one picture slice, including each reference B slice in a reference B picture, and limiting reference B pictures to three levels of importance.

6. The method according to claim 5 , comprising encoding the reference B pictures in a manner such that a reference B picture at a lower level of importance is not used as reference for a reference B picture at a higher level of importance and is not used as reference for a P picture.

7. The method according to claim 1 , wherein the header element is a two-bit syntax element and the method comprises:

assigning a first value to the syntax element in the event that the B slice is a non-reference slice,

assigning a second value to the syntax element in the event that the B slice is a reference slice and is a slice of a picture having a first level of discardability in a hierarchy of discardability of reference B pictures,

assigning a third value to the syntax element in the event that the B slice is a reference slice and is a slice of a picture having a second level of discardability in a hierarchy of discardability of reference B pictures, and

assigning a fourth value to the syntax element in the event that the B slice is a reference slice and is a slice of a picture having a third level of discardability in a hierarchy of discardability of reference B pictures,

said hierarchy of discardability comprising a ranking of the plurality of B slices by how much degradation to the sequence of picture slices will be experienced by a discarding of each B slice.

8. Apparatus for creating a bitstream in which a sequence of picture slices is encoded as reference slices and non-reference slices, wherein the reference slices include a plurality of B slices, said apparatus comprising a means for performing a method comprising:

for B each slice, forming at least one data packet containing data values derived from brightness and color information pertaining to the slice, and

for each B slice, including in the at least one packet a header element reflecting whether the B slice is a reference slice,

said header element containing a value that indicates where in a hierarchy of discardability the B slice is,

said hierarchy of discardability comprising a ranking of the plurality of B slices by how much degradation to the sequence of picture slices will be experienced by a discarding of each B slice.

9. A method of decoding a bitstream conveying a sequence of pictures each containing at least one picture slice, wherein each picture slice is encoded as a reference slice or a non-reference slice, the reference slices include a plurality of B slices, bits of each reference B slice are conveyed by at least one data packet that includes a header element having a value, and said method comprises:

receiving a trick play command calling for pictures to be discarded without being presented, and selecting, using an advanced video coding (AVC) decoder, a picture to be discarded based on the value of the header element

said value indicating where in a hierarchy of discardability a B slice is,

said hierarchy of discardability comprising a ranking of the plurality of B slices by how much degradation to the sequence of picture slices will be experienced by a discarding of each B slice.

10. The method according to claim 9 , wherein the header element is a two-bit syntax element.

11. Apparatus for decoding a bitstream conveying a sequence of pictures each containing at least one picture slice, wherein each picture slice is encoded as a reference slice or a non-reference slice, the reference slices include B slices, and bits of each reference B slice are conveyed by at least one data packet that includes a header element having a value, the apparatus comprising a means for performing a method comprising:

receiving a trick play command calling for pictures to be discarded without being presented,

selecting a reference B picture to be discarded based on the value of the header element,

said value indicating where in a hierarchy of discardability a B slice is,

said hierarchy of discardability comprising a ranking of the B slices by how much degradation to the sequence of picture slices will be experienced by a discarding of each B slice.

12. A non-transitory computer readable medium containing instructions that, when executed by a computing machine receiving a bitstream in which a sequence of picture slices is encoded as reference slices and non-reference slices, wherein the reference slices include a plurality of B slices, cause the computing machine to perform a method comprising:

for each B slice, forming at least one data packet containing data values derived from brightness and color information pertaining to the slice, and

for each B slice, including in the at least one data packet a header element indicating whether the B slice is a reference slice,

said value indicating where in a hierarchy of discardability a B slice is,

said hierarchy of discardability comprising a ranking of the plurality of B slices by how much degradation to the sequence of picture slices will be experienced by a discarding of each B slice.

13. A method of assisting in trick play operation of a video recording and playback device, comprising:

at a transmitting location,

encoding, using an advanced video coding (AVC) encdoder, a sequence of pictures, each containing at least one picture slice, selectively in either a normal mode or a trick play assist mode, and

in the trick play assist mode, encoding each slice as one of at least two different types depending on prediction needed to decode the slice, for each slice of at least one type, forming at least one data packet containing data values derived from image information pertaining to the slice, creating at least one data value for each slice of said one type,

said value indicating where in a hierarchy of discardability the B slice is,

said hierarchy of discardability comprising a ranking of the B slices by how much degradation to the sequence of picture slices will be experienced by a discarding of each B slice,

at the receiving location,

extracting data packets from the transport stream and temporarily storing the data packets for later playback in normal operation or trick play operation, and

in trick play operation, detecting, using an advanced video coding (AVC) decoder, whether the sequence of picture slices was encoded in the trick play assist mode and, if so, selectively discarding pictures depending on said data values.

14. The method according to claim 13 , comprising encapsulating the data packets in an MPEG transport stream wherein a descriptor indicates whether the sequence of pictures has been encoded in the trick play assist mode.

15. The method according to claim 13 , comprising, in the trick play assist mode, generating supplementary assist information that indicates that the sequence of pictures has been encoded in the trick play assist mode and communicated the trick play assist information to the receiving location.

16. The method according to claim 13 , wherein each slice is either an I slice, a P slice or a B slice, each B slice is a reference B slice or a non-reference B slice, each reference B slice is included in a reference B picture, each reference B picture has one of three levels of importance, and the method comprises encoding the reference B pictures in a manner such that a reference B picture at a lower level of importance is not used as reference for a reference B picture at a higher level of importance and is not used as reference for a P picture.

17. A method of creating a bitstream in which a sequence of picture slices is encoded as slices of at least two different types depending on prediction needed to decode the slice, said method comprising:

for each slice of at least one type, forming at least one data packet containing data values derived from brightness and color information pertaining to the slice, and

for a slice of said one type, including in the at least one data packet a header element indicating whether the slice is a reference slice, and

wherein in the event that said slice is a reference slice the header element has a value,

said value indicating where in a hierarchy of discardability the reference slice is,

said hierarchy of discardability comprising a ranking of the reference slices of the sequence by how much degradation to the sequence of picture slices will be experienced by a discarding of each B slice.

18. The method according to claim 17 , wherein the slices of said one type are B slices.

19. The method according to claim 17 , comprising encoding a sequence of pictures as intra-coded pictures and predictive coded pictures, wherein the predictive coded pictures include reference pictures, and the method includes limiting reference pictures of at least one type of predictive coded picture to three levels of importance.