Method and system for staggered parallelized video decoding

1. A method for processing video frames, comprising:

receiving a compressed video stream;

identifying even frames and odd frames from the video stream;

storing the even and odd frames in a hardware memory;

processing a first portion of a first even frame from among the stored even frames using a first decoder, the first portion of the first even frame associated with a number of lines of a reference area determined using a search range; and

after the first decoder completes processing the first portion of the first even frame, beginning processing a first odd frame from among the stored odd frames using a second decoder;

wherein the reference area is bounded by a rectangle starting at a location (x−the search range, y−the search range), and ending at a location (x+the search range, y+the search range); and wherein

x and y correspond to coordinates of an upper left corner of a macroblock of the first odd frame decodable by the second decoder.

2. The method for processing video frames of claim 1 , further comprising:

staggering the beginning processing times of the even and odd frames; and

decoding at least some of the even and odd frames simultaneously using the first and second decoders.

3. The method for processing video frames of claim 1 , wherein beginning processing the first odd frame further comprises beginning processing the first odd frame before the first decoder completes processing a remaining portion of the first even frame, the remaining portion being different from the first portion.

4. The method for processing video frames of claim 1 , wherein each of the first and second decoders comprises an H.264 compliant decoder.

5. A method for processing video frames, comprising:

receiving a compressed video stream;

identifying even frames and odd frames from the video stream;

storing the even and odd frames in a hardware memory;

processing a first portion of a first even frame from among the stored even frames using a first decoder;

after the first decoder completes processing the first portion of the first even frame, beginning processing a first odd frame from among the stored odd frames using a second decoder;

processing a first portion of a second even frame from among the stored even frames using the first decoder; and

after the first decoder completes processing the first portion of the second even frame, the second decoder beginning processing a second odd frame from among the stored odd frames;

wherein the first portions of the first and second even frames are associated with a number of lines of a reference area determined using a search range;

the reference area is bounded by a rectangle starting at a location (x−the search range, y−the search range), and ending at a location (x+the search range, y+the search range); and

x and y correspond to coordinates of an upper left corner of a macroblock of at least one of the first and second odd frames.

6. The method of claim 5 , wherein:

processing the first portion of the second even frame occurs after the first decoder processes the first portion of the first even frame; and

beginning processing the second odd frame occurs after beginning processing the first odd frame.

7. The method for processing video frames of claim 1 , wherein the decoding of any video frame is ahead of a next video frame by a processing time that is determined based on at least the number of lines of the reference area.

8. The method of claim 7 , further comprising the first decoder processing each of the even frames ahead of a next odd frame by the processing time.

9. The method of claim 7 , further comprising the second decoder processing each of the odd frames ahead of a next even frame by the processing time.

10. The method for processing video frames of claim 1 , further comprising:

outputting the processed video frames to one or more display devices.

11. A method for processing video frames, comprising:

receiving a compressed video stream;

identifying even frames and odd frames from the video stream;

storing the even and odd frames in a hardware memory;

processing a first portion of a first even frame from among the stored even frames using a first decoder;

after the first decoder completes processing the first portion of the first even frame, beginning processing a first odd frame from among the stored odd frames using a second decoder

determining a frame height H of video frames from the video stream;

dividing the frame height H by a number of lines L associated with a reference area determined using a search range to arrive at a number Z;

rounding down the number Z to a nearest whole number R; and

configuring R or fewer decoders to process at least some of video frames simultaneously.

12. The method of claim 11 , further comprising:

determining G groups of video frames and associating each group with a corresponding one of the R or fewer decoders.

13. The method for processing video frames of claim 11 , further comprising:

simultaneously decoding at least some of the video frames using the R or fewer decoders.

14. The method for processing video frames of claim 13 , wherein simultaneously decoding includes staggering a beginning processing time of each of the R or fewer decoders.

15. The method for processing video frames of claim 1 , wherein identifying the even frames and the odd frames from the video stream includes parsing start codes of at least some of the even frames and the odd frames.

16. A system, comprising:

a parser structured to receive a compressed video stream and to identify groups of video frames from the video stream;

a memory to store the groups of video frames; and

a plurality of video decoders, each decoder structured to process video frames from a corresponding one group of video frames, wherein the beginning processing times for each decoder is staggered;

wherein the number of video decoders in the plurality corresponds to a number no greater than R, wherein R corresponds to a frame height H of at least one of the video frames divided by a number of lines L associated with a reference area, and rounded down to a nearest whole number.

17. The system of claim 16 , wherein the video decoders are structured to simultaneously decode at least one video frame from each group.

18. The system of claim 16 , wherein the decoding of any video frame is ahead of a next video frame by a predefined processing time.

19. The system of claim 16 , wherein the plurality of video decoders corresponds to two decoders and the groups of video frames correspond to a first group having even video frames and a second group having odd video frames.

20. The method for processing video frames of claim 16 , wherein:

the reference area is bounded by a rectangle starting at a location (x−a search range, y−the search range), and ending at a location (x+the search range, y+the search range); and

x and y correspond to coordinates of a corner of a video macroblock.

21. The system of claim 16 , further comprising:

a buffer management unit to store information about decoded video frames from among the groups of video frames; and

one or more monitors to display at least one decoded video frame.

22. The system of claim 21 , wherein at least some of the video decoders are configured to read at least some of the information stored in the buffer management unit about video frames decoded by a different video decoder from among the plurality of video decoders.

23. The system of claim 16 , wherein the parser is configured to identify the groups of video frames based at least on start codes embedded in one or more of the video frames.