Apparatus and method for decompressing high definition pictures

• US 6,104,751 A
• Filed: 10/26/1994
• Issued: 08/15/2000
• Est. Priority Date: 10/29/1993
• Status: Expired due to Term

• Current Assignee: SGS-Thomson Microelectronics (STMicroelectronics NV)
• Original Assignee: SGS-Thomson Microelectronics (STMicroelectronics NV)
• Inventors: Artieri, Alain
• Primary Examiner(s): Le, Vu
• Application Number:
  US08/329,945
• Time in Patent Office: 2,120 Days
• Field of Search: 348/390, 348/402, 348/403, 348/384, 348/421, 348/721, 348/408, 348/725, 348/699, 348/701, 348/416
• US Class Current: 375/240.14
• CPC Class Codes: H04N 19/42 characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation H04N19/635 takes precedence
• CPC Class Codes: H04N 19/423 characterised by memory arrangements H04N19/433 takes precedence
• CPC Class Codes: H04N 19/436 using parallelised computational arrangements
• CPC Class Codes: H04N 19/439 using cascaded computational arrangements for performing a single operation, e.g. filtering
• CPC Class Codes: H04N 19/61 in combination with predictive coding

Abstract

A system for processing compressed data corresponding to pictures includes a decoding mechanism, providing a picture memory with decoded picture data. The decoding mechanism requires, for decoding a current block of a previously decoded picture. A plurality of decoders are associated with respective picture memories, each storing a specific slice of corresponding blocks of a plurality of pictures, as well as at least one margin which is liable to be a predictor block serving to decode a block of the specific slice.

Claims

1. A system for processing compressed data corresponding to pictures, the compressed data including a plurality of slices of data, the system comprising:

a decoder system having an input that receives the compressed data, the decoder system generating decoded picture data based upon a current block of the compressed data and a predictor block of decoded picture data previously decoded by the decoder;

wherein the decoder system includes a plurality of decoders and a plurality of picture memories, each of the plurality of decoders being coupled to a respective one of the plurality of picture memories, each picture memory of the plurality of picture memories storing one of the plurality of slices of data and at least one boundary area of an adjacent slice of the plurality of slices of data that is decoded by at least one second decoder of the plurality of decoders.

2. The system of claim 1, wherein each decoder includes means for storing in a respective picture memory, the boundary area of the adjacent slice, and for providing to the at least one second decoder, as the boundary area of the at least one second decoder, a boundary area of the one of the plurality of slices stored in the respective picture memory.

3. The system of claim 1, wherein each decoder includes:

a first buffer memory receiving picture blocks included in the one of the plurality of slices of data;

at least one second buffer memory receiving picture blocks included in the boundary area of the adjacent slice;

a final processing circuit providing decoded picture data corresponding to the one of the plurality of slices of data to the first buffer memory of the each decoder and to a second buffer memory of at least one other decoder; and

a memory controller for reading the decoded picture data from the first buffer memory at addresses corresponding to the one of the plurality of slices of data, and for reading the decoded picture data from the at least one second decoder the second buffer memory and for writing the decoded picture data from the first buffer memory and the decoded picture data from another of the plurality of memories at addresses corresponding to the boundary area.

4. The system of claim 3, wherein each of the plurality of decoders further includes a barrier circuit, coupled to one of the at least one second buffer memory, for receiving and filtering the decoded picture data from another of the plurality of memories, and for storing in the second buffer memory only data corresponding to the boundary area of the adjacent slice.

5. The system of claim 1, wherein the pictures to be processed are high definition television pictures that are partitioned into four horizontal slices of equal height.

6. The apparatus of claim 1 wherein the boundary area of the adjacent slice contains the predictor block.

7. An apparatus for decompressing a compressed picture including a plurality of slices of compressed data, the apparatus comprising:

a plurality of decoders, each of the plurality of decoders having a first input that receives one of the plurality of slices of compressed data, a second input that receives reconstructed adjacent data from an adjacent one of the plurality of decoders, and an output that provides reconstructed data generated from the slice of compressed data and the reconstructed adjacent data; and

a plurality of memories, each of the plurality of memories being coupled to a respective one of the plurality of decoders and storing the reconstructed data and the reconstructed adjacent data.

8. The apparatus of claim 7 wherein the slice of compressed data of a first decoder of the plurality of decoders includes a first macroblock, and the reconstructed adjacent data of the first decoder includes a predictor macroblock, the first decoder decompressing the first macroblock with respect to the predictor macroblock.

9. The apparatus of claim 7, wherein each of the plurality of decoders includes an adder for reconstructing the reconstructed data from the slice of compressed data and one of the slice of compressed data and the reconstructed adjacent data.

10. The apparatus of claim 7, wherein each of the plurality of decoders includes a barrier circuit, the barrier circuit having an input coupled to an output of a second decoder and receiving the reconstructed adjacent data, the barrier circuit filtering the reconstructed adjacent data from the second decoder to provide a portion of the reconstructed adjacent data, the output of each of the plurality of decoders providing the reconstructed data generated from the slice of compressed data and the portion of the reconstructed adjacent data.

11. The apparatus of claim 10 wherein the reconstructed data represents a first part of the compressed picture, and the portion of the reconstructed adjacent data represents a second part of the compressed picture that is adjacent to the first part.

12. The apparatus of claim 7, wherein each of the plurality of slices of data is of equal height.

13. The apparatus of claim 7 wherein the plurality of decoders is four decoders.

14. The apparatus of claim 7 wherein the compressed picture is a high definition television picture having 1440×

960 pixels.

15. A method for decompressing a compressed picture, the compressed picture including a plurality of slices of compressed data, the method comprising the steps of:

reconstructing, by each of a plurality of decoders, one of the plurality of slices of compressed data from the one of the plurality of slices of compressed data and reconstructed adjacent data provided by a second decoder of the plurality of decoders, to provide reconstructed data; and

storing the reconstructed data and the reconstructed adjacent data in one of a plurality of memories.

16. The method of claim 15, wherein:

the one of the plurality of slices of compressed data includes a first macroblock;

the reconstructed adjacent data includes a predictor macroblock; and

the step of reconstructing includes reconstructing data based upon the macroblock and the predictor macroblock.

17. The method of claim 15 wherein the step of reconstructing includes adding a portion of the one of the plurality of slices of compressed data to a portion of the reconstructed adjacent data.

18. The method of claim 15, further comprising the steps, performed by each of the plurality of decoders, of:

receiving the reconstructed adjacent data from the second decoder; and

filtering the reconstructed adjacent data to provide a portion of the reconstructed adjacent data; and

wherein the step of reconstructing includes reconstructing the one of the plurality of slices of compressed data from the one of the plurality of slices of compressed data and the portion of the reconstructed adjacent data.

19. The method of claim 18, wherein the reconstructed data represents a first part of the compressed picture, and the portion of the reconstructed adjacent data represents a second part of the compressed picture that is adjacent to the first part.

20. The method of claim 15 wherein each of the plurality of slices of data is of equal height.

21. The method of claim 15, wherein the compressed picture is a high definition television picture having 1440×

960 pixels.

22. An apparatus for decompressing a compressed picture, the compressed picture including a plurality of slices of compressed data, the apparatus comprising:

a plurality of decoders, each of the plurality of decoders including means for reconstructing one of the plurality of slices of compressed data from the one of the plurality of slices of compressed data and reconstructed adjacent data provided by a second decoder of the plurality of decoders, to provide reconstructed data;

a plurality of memories, each of the plurality of memories corresponding to a respective one of the plurality of decoders; and

means for storing, in one of the plurality of memories, the reconstructed data and the reconstructed adjacent data.

23. The apparatus of claim 22, wherein:

the one of the plurality of slices of compressed data includes a first macroblock;

the reconstructed adjacent data includes a predictor macroblock; and

the means for reconstructing includes means for reconstructing data based upon the macroblock and the predictor macroblock.

24. The apparatus of claim 22 wherein the means for reconstructing includes means for adding a portion of the one of the plurality of slices of compressed data to a portion of the reconstructed adjacent data.

25. The apparatus of claim 22, wherein each of the plurality of decoders further includes:

means for receiving the reconstructed adjacent data from the second decoder; and

means for filtering the reconstructed adjacent data to provide a portion of the reconstructed adjacent data; and

wherein the means for reconstructing includes means for reconstructing the one of the plurality of slices of compressed data from the one of the plurality of slices of compressed data and the portion of the reconstructed adjacent data.

26. The apparatus of claim 25, wherein the reconstructed data represents a first part of the compressed picture, and the portion of the reconstructed adjacent data represents a second part of the compressed picture that is adjacent to the first part.

27. The apparatus of claim 22 wherein the plurality of decoders includes four decoders.

28. The apparatus of claim 22 wherein each of the plurality of slices of data is of equal height.

29. The apparatus of claim 22, wherein the compressed picture is a high definition television picture having 1440×

960 pixels.

30. A method for decompressing a compressed picture, the compressed picture including a plurality of slices of compressed data and a plurality of motion vectors, each of the plurality of motion vectors having a respective vertical amplitude of a plurality of amplitudes, the method comprising the steps of:

determining a maximum vertical amplitude of the plurality of amplitudes;

determining an amount of memory to be used for decoding each of the plurality of slices of compressed data, based upon the maximum vertical amplitude;

storing the plurality of slices of compressed data in a plurality of memories, each of the plurality of memories containing the amount of memory; and

decoding, in a plurality of decoders using the plurality of memories, the plurality of slices, to generate a decompressed picture from the compressed picture.

31. The method of claim 30, wherein the step of determining an amount of memory includes determining the amount of memory to be an amount sufficient to store one and one-half times an amount of data associated with one of the plurality of slices of compressed data.

32. The method of claim 30, wherein the step of determining an amount of memory includes determining the amount of memory to be an amount sufficient to store twice an amount of data associated with one of the plurality of slices of compressed data.