Method and apparatus for multiple description video coding

1. An encoder comprising:

a controller to determine whether macroblocks of a frame of video content are to be processed by intra-frame encoding or by predictive coding;

a switch coupled to the controller;

an intra-frame unit to receive the macroblocks via the switch when the controller determines to process the macroblocks by intra-frame encoding;

a predictive unit to receive the macroblocks via the switch when the controller determines to process the macroblocks by predictive encoding; and

a redundancy allocation unit coupled to the controller, wherein the controller determines whether to process the macroblocks by intra-frame encoding or predictive frame encoding based on information received from the redundancy allocation unit.

2. The encoder of claim 1 , further comprising a rate controller coupled to the controller, wherein the controller determines whether to process the macroblocks by intra-frame encoding or predictive frame encoding further based on information received from the rate controller.

3. The encoder of claim 1 , further comprising frame buffers to store versions of the frame decoded from a first output of the intra-frame unit or from a second output of the predictive unit.

4. The encoder of claim 1 , wherein the intra-frame unit encodes a particular macroblock of the macroblocks by coding color values of the particular macroblock to a first description and to a second description.

5. The encoder of claim 1 , wherein the predictive unit encodes a particular macroblock of the macroblocks by coding, based on a previously encoded frame, both a motion vector and an error into a first predictive description and into a second predictive description.

6. The encoder of claim 5 , wherein the error corresponds to a difference between the particular macroblock and another macroblock from the previously encoded frame.

7. The encoder of claim 5 , wherein the motion vector is descriptive of displacement between the particular macroblock and a best matching macroblock of the previously encoded frame.

8. The encoder of claim 1 , wherein the controller determines to process particular macroblocks corresponding to a particular frame by intra-frame coding after processing between 10 to 15 other macroblocks corresponding to other frames by predictive coding.

9. The encoder of claim 1 , wherein the predictive unit is configured to determine sets of best prediction macroblocks associated with particular macroblocks when the switch sends the macroblocks to the predictive unit, and wherein the predictive unit is configured to determine whether to process the particular macroblocks or to send the particular macroblocks to the intra-frame unit based on the sets of best prediction macroblocks.

10. The encoder of claim 1 , wherein the intra-frame unit outputs a first intra-frame description to a first channel, and wherein the predicative unit outputs a first predictive description to the first channel.

11. The encoder of claim 10 , wherein the intra-frame unit outputs a second intra-frame description to a second channel, and wherein the predicative unit outputs a second predictive description to the second channel.

12. A method comprising:

receiving, at a controller of an encoder, information from a redundancy allocation unit;

in response to a first determination by the controller to process macroblocks of a frame of video content by intra-frame coding, sending the macroblocks via a switch to an intra-frame unit, the first determination based on the information; and

in response to a second determination by the controller to process the macroblocks by predictive coding, sending the macroblocks via the switch to a predictive unit, the second determination based on the information.

13. The method of claim 12 , wherein the first determination and the second determination are further based on second information received by the controller from a rate controller.

14. The method of claim 12 , wherein the predictive unit receives macroblocks from a reconstructed frame from a set of frame buffers.

15. The method of claim 12 , wherein the predictive unit determines sets of best prediction macroblocks associated with particular macroblocks in response to receipt of the macroblocks from the switch.

16. The method of claim 15 , wherein the predictive unit determines whether to process the particular macroblocks or send the particular macroblocks to the intra-frame unit based on the sets of best prediction macroblocks.

17. A computer-readable storage device storing instructions that, when executed by a processor, cause the processor to perform operations comprising:

receiving information from a redundancy allocation unit;

in response to a first determination to process macroblocks of a frame of video content by intra-frame coding, sending the macroblocks via a switch to an intra-frame unit, the first determination based on the information; and

in response to a second determination to process the macroblocks by predictive coding, sending the macroblocks via the switch to a predictive unit, the second determination based on the information.

18. The computer-readable storage device of claim 17 , wherein the intra-frame unit codes color values of the macroblock into an intra-frame first description of a multiple description coding scheme and an intra-frame second description of the multiple description coding scheme.

19. The computer-readable storage device of claim 17 , wherein the predictive unit codes, based on a previously encoded frame, motion vectors and errors for the macroblock into a first predictive description and into a second predictive description.

20. The computer-readable storage device of claim 19 , wherein a motion vector of the motion vectors is descriptive of displacement between the macroblock and a best matching macroblock of the previously encoded frame.