Method of multiplexing H.264 elementary streams without timing information coded

1. A method for multiplexing high definition video elementary streams that do not have timing information coded to create a transport stream (TS), the method comprising:

constructing ‘n’ number of access units (AU) in a list, each AU capable of looking ahead to obtain information on timing for the next AU, including:

generating the timing information including presentation time stamp (pts) and decode time stamp (dts) and deriving variable frame rate (tc) for each AU,

generating the dts for a current AU from the dts of a preceding AU, and

generating the pts from the dts of the current AU;

initiating a pointer to indicate start of an AU from the list;

iteratively checking start of an AU from the list;

constructing a header by calculating timing information from the topmost AU of the list;

writing the header in the TS; and

filling the remaining bytes of the TS with data from a packetized elementary stream (PES), wherein the step of constructing ‘n’ number of access units comprises:

reading the elementary stream from the start while converting the elementary stream to the PES until a single AU is formed and noting its start and end points;

pushing the just formed single AU into the list;

checking primary picture type coded in the AU pushed into the list;

calculating

dts ( n )= dts ( n− 1)+ tc , and

pts ( n )= dts ( n )

if the primary picture type coded is a B frame, and

decoding B frames and presenting at the same time without including any offset between pts and dts of the B frames;

calculating

dts ( n )= dts ( n− 1)+ tc , and

pts ( n )= dts ( n )+ pts _offset* tc

if the primary picture coded type is an I or P frame, and checking next access units for the number of B frames (pts_offset) that precede before the next I or P frames occurs; and

constructing a next AU and pushing the next AU into the list for repeating the steps of calculating timing information.

2. The method as in claim 1 wherein picture type (B, P, or I) information is used to generate the pts and dts values of the Packetized Elementary Stream (PES).

3. The method as in claim 1 wherein variable tc is frame rate or field rate and is derived as tc=num_units_in_tick/time_scale.

4. The method as in claim 1 wherein PES packets are broken into fixed 188 byte TS packets.

5. An apparatus comprising circuitry for multiplexing high definition video elementary streams that do not have timing information coded to create a transport stream (TS) including circuitry configured to:

construct ‘n’ number of access units (AU) in a list, each AU looking ahead to obtain information on timing for the next AU, including the steps of

create the timing information including presentation time stamp (ptos) and decode time stamp (dts) and deriving variable frame rate (tc) for each AU;

create the dts for a current AU from the dts of a preceding AU, and

create the pts from the dts of the current AU;

initiate a pointer to indicate a start of an AU from the list;

iteratively check the start of an AU from the list;

construct a header by calculating the timing information from the topmost AU of the list;

write the header in the TS; and

fill the remaining bytes of TS with data from a packetized elementary stream (PES), wherein the circuitry is further configured to:

read the elementary stream from the start while converting the elementary stream to the PES until a single AU is formed and noting start and end points of the single AU;

push the just formed single AU into the list;

check primary picture type coded in the AU pushed into the list;

calculate

dts ( n )= dts ( n− 1)+ tc , and

pts ( n )= dts ( n )

if the primary picture type coded is a B frame, and decode B frames and presenting at the same time without including any offset between pts and dts of the B frames;

calculate

dts ( n )= dts ( n− 1)+ tc , and

pts ( n )= dts ( n )+ pts _offset* tc

if the primary picture coded type is an I or P frame, and checking next access units for the number of B frames (pts_offset) that precede before the next I or P frames occurs; and

construct a next AU and pushing the next AU into the list for repeating the steps of calculating timing information.

6. The apparatus as in claim 5 wherein picture type (B, P, or I) information is used to create the PTS and DTS values of the Packetized Elementary Stream (PES).

7. The apparatus as in claim 5 wherein variable tc is frame rate or field rate and is derived as tc=num_units_in_tick/time_scale.

8. The apparatus as in claim 5 wherein PES packets are broken into fixed 188 byte TS packets.

9. The method as in claim 1 wherein the high definition elementary streams are H.264/AVC video elementary streams.

10. The method as in claim 1 wherein the TS is an MPEG-2 TS.

11. The method as in claim 1 wherein the multiplexing is deferred time multiplexing of H.264 video elementary streams that do not have either NAL/VCL-HRD parameters coded (as part of VUI information of the Sequence Parameter Set) or Supplemental Enhancement Information (SEI).

12. The apparatus in claim 5 wherein the high definition elementary streams are H.264/AVC video elementary streams.

13. The apparatus in claim 5 wherein the TS is an MPEG-2 TS.

14. The apparatus in claim 5 wherein the multiplexing is deferred time multiplexing of H.264 video elementary streams which do not have either NAL/VCL-HRD parameters coded (as part of VUI information of the Sequence Parameter Set) or Supplemental Enhancement Information (SEI).