Video decoder with functions to detect and process errors in encoded video data

• US 5,854,799 A
• Filed: 11/29/1995
• Issued: 12/29/1998
• Est. Priority Date: 11/29/1994
• Status: Expired due to Term

• Current Assignee: Godo Kaisha IP Bridge 1 (IP Bridge, Inc.)
• Original Assignee: Sanyo Electric Company Limited (Panasonic Holdings Corporation)
• Inventors: Okada, Shigeyuki, Tanahashi, Naoki, Nakashima, Hayato
• Primary Examiner(s): Elmore, Reba I.
• Assistant Examiner(s): MARC, MCDIEUNEL
• Application Number:
  US08/564,478
• Time in Patent Office: 1,126 Days
• Field of Search: 371/37.1, 378/98.5, 463/1, 463/43, 358/539, 348/239, 348/578, 469/474, 345/122, 395/173, 380/20
• US Class Current: 714/781
• CPC Class Codes: H04N 19/00 Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
• CPC Class Codes: H04N 19/577 Motion compensation with bidirectional frame interpolation, i.e. using B-pictures
• CPC Class Codes: H04N 19/61 in combination with predictive coding
• CPC Class Codes: H04N 19/70 characterised by syntax aspects related to video coding, e.g. related to compression standards
• CPC Class Codes: H04N 19/895 in combination with error concealment

Abstract

A video decoding apparatus for decoding encoded video data to continuously produce decoded pictures. The video data includes a series of pictures, each picture contains a series of slices, and each slice contains a series of macroblocks. A dequantizer performs dequantization of the video data based upon a quantization threshold value. A motion-vector restoring circuit restores data for each macroblock. A direct current error detector is provided to detect erroneous macroblocks based upon the dequantized data. A motion-area error detector is provided to detect erroneous macroblocks based upon the restored motion vector data. An erroneous macroblock is replaced by a corresponding macroblock from a preceding picture. Subsequent macroblocks in a slice may also be replaced by corresponding macroblocks from a preceding picture.

Claims

1. A video decoding apparatus for decoding encoded video data to continuously produce decoded pictures, said encoded video data including a series of pictures, with each picture including a plurality of slices, and each slice including a plurality of macroblocks, said video decoding apparatus comprising:

a dequantizer for performing dequantization on said encoded video data to produce dequantized data based on a quantization threshold value;

a first error detector for checking whether each of said slices of each of said pictures includes an erroneous macroblock, based on said dequantized data, said first error detector performs the checking macroblock by macroblock; and

an error processing circuit for replacing an erroneous macroblock with a preceding correct macroblock, wherein when a check target slice in a picture currently undergoing decoding includes an erroneous macroblock, said error processing circuit replaces said erroneous macroblock with a preceding correct macroblock, said preceding correct macroblock included in a preceding picture to be output from said video decoding apparatus prior to said picture currently undergoing decoding, and wherein a position of said preceding correct macroblock in said preceding picture corresponds to a position of said erroneous macroblock in said picture currently undergoing decoding.

2. The video decoding apparatus according to claim 1, wherein said error processing circuit replaces a first group of macroblocks including said erroneous macroblock and a plurality of macroblocks chronologically following said erroneous macroblock in said check target slice, with a second group of macroblocks including said preceding correct macroblock and a plurality of macroblocks chronologically following said preceding correct macroblock in a slice positionally corresponding to said check target slice.

3. The video decoding apparatus according to claim 1, wherein said dequantizer performs dequantization on said video data to produce discrete cosine transform coefficient data as said dequantized data, said dequantized data being produced macroblock by macroblock, and

wherein said first error detector determines, macroblock by macroblock, whether a direct current coefficient included in said discrete cosine transform coefficient data lies within a predetermined range, and determines a macroblock which is associated with a direct current coefficient lying outside of said predetermined range as an erroneous macroblock.

4. The video decoding apparatus according to claim 1, further comprising:

a variable length decoder for performing variable length decoding on said video data using Huffman codes and supplying variable-length decoded data to said dequantizer; and

a second error detector for checking whether each of said slices of said each picture contains an error, based on said variable-length decoded data,

wherein when said check target slice in said picture currently undergoing decoding is in error, said error processing circuit replaces said check target slice with a preceding correct slice, said preceding correct slice included in said preceding picture, a position of said preceding correct slice in said preceding picture corresponds to a position of said check target slice in said picture currently undergoing decoding.

5. The video decoding apparatus according to claim 4, further comprising a memory for storing a Huffman table indicating said Huffman codes, and

wherein said second error detector determines whether data of each of said slices corresponds to Huffman codes in said Huffman table and also determines whether current variable-length decoded data for a current check target slice is consistent with old variable-length decoded data and identifies a macroblock as erroneous if said data of each of said slices does not correspond to said Huffman codes or if said current variable-length decoded data is inconsistent with said old variable-length decoded data.

6. The video decoding apparatus according to claim 4, further comprising:

a motion-vector restoring circuit for restoring data of a motion vector for each of said macroblocks; and

a third error detector for checking whether each of said slices of said each picture includes an erroneous macroblock based on said restored motion vector data, said third error detector performs the checking macroblock by macroblock.

7. The video decoding apparatus according to claim 6, wherein said third error detector determines whether said restored motion vector data indicates a point inside or outside a picture, and identifies a macroblock corresponding to a motion vector indicating a point outside a picture as an erroneous macroblock.

8. The video decoding apparatus according to claim 1, further comprising:

a motion-vector restoring circuit for restoring data of a motion vector for each of said macroblocks; and

a second error detector for checking whether each of said slices of said each picture includes an erroneous macroblock based on said restored motion vector data, said second error detector performs the checking macroblock by macroblock.

9. The video decoding apparatus according to claim 8, wherein said second error detector determines whether said restored motion vector indicates a point inside or outside a picture, and identifies a macroblock corresponding to a motion vector indicating a point outside a picture as an erroneous macroblock.

10. A video decoding apparatus for decoding encoded video data to continuously produce decoded pictures, said encoded video data including a series of pictures, with each picture including a plurality of slices, and each slice including a plurality of macroblocks, said video decoding apparatus comprising:

a motion-vector restoring circuit for restoring data of a motion vector for each of said macroblocks;

an error detector for checking whether each of said slices of said each picture includes an erroneous macroblock based on said restored motion vector data, said error detector performs the checking macroblock by macroblock; and

an error processing circuit for replacing an erroneous macroblock with a preceding correct macroblock, wherein when a check target slice in a picture currently undergoing decoding includes an erroneous macroblock, said error processing circuit replaces said erroneous macroblock with a preceding correct macroblock, said preceding correct macroblock included in a preceding picture to be output from said video decoding apparatus prior to said picture currently undergoing decoding, and wherein a position of said preceding correct macroblock in said preceding picture corresponds to a position of said erroneous macroblock in said picture currently undergoing decoding.

11. The video decoding apparatus according to claim 10, wherein said error detector determines whether said restored motion vector data indicates a point inside or outside a picture, and identifies a macroblock corresponding to a motion vector indicating a point outside a picture as an erroneous macroblock.

12. The video decoding apparatus according to claim 10, wherein said error processing circuit replaces a first group of macroblocks including said erroneous macroblock and a plurality of macroblocks chronologically following said erroneous macroblock in said check target slice, with a second group of macroblocks including said preceding correct macroblock and a plurality of macroblocks chronologically following said preceding correct macroblock in a slice positionally corresponding to said check target slice.

13. A video decoding apparatus for decoding an encoded video bit stream to continuously produce decoded pictures, said encoded video bit stream including a series of pictures, with each picture including a plurality of slices, and each slice including a plurality of macroblocks, said video decoding apparatus comprising:

a slice header detector for detecting a slice header affixed at a head of each of said slices;

a variable length decoder for performing variable length decoding on said video bit stream based on Huffman codes, said variable length decoder supplying variable-length decoded data;

a dequantizer for performing dequantization on said variable-length decoded data based on a quantization threshold value to produce discrete cosine transform coefficient data;

an inverse discrete cosine transform circuit for performing an inverse discrete cosine transform on said discrete cosine transform coefficient data;

a motion compensated prediction circuit for performing motion compensated prediction based on a processing result from said inverse discrete cosine transform circuit to produce said decoded pictures;

a frame buffer for temporarily storing said produced decoded pictures, said motion compensated prediction circuit performing motion compensated prediction using said decoded pictures stored in said frame buffer;

a first error detector for checking whether each of said slices of each of said pictures includes an erroneous macroblock based on said discrete cosine transform coefficient data produced by said dequantizer, said first error detector performs the checking macroblock by macroblock; and

an error processing circuit for replacing an erroneous macroblock with a preceding correct macroblock using a slice header detected by said slice header detector, wherein when a check target slice in a picture currently undergoing decoding includes an erroneous macroblock, said error processing circuit replaces said erroneous macroblock with a preceding correct macroblock, said preceding correct macroblock included in a preceding decoded picture to be output from said frame buffer prior to said picture currently undergoing decoding, and wherein a position of said preceding correct macroblock in said preceding decoded picture corresponds to a position of said erroneous macroblock in said picture currently undergoing decoding.

14. The video decoding apparatus according to claim 13, wherein said error processing circuit replaces a first group of macroblocks including said erroneous macroblock and a plurality of macroblocks chronologically following said erroneous macroblock in said check target slice, with a second group of macroblocks including said preceding correct macroblock and a plurality of macroblocks chronologically following said preceding correct macroblock in a slice positionally corresponding to said check target slice.

15. The video decoding apparatus according to claim 14, wherein said error processing circuit inhibits operation of said variable length decoder, said dequantizer and said inverse discrete cosine transform circuit on said check target slice which includes said erroneous macroblock and invalidates processing results, and restarts operation of said variable length decoder, said dequantizer and said inverse discrete cosine transform circuit for a slice chronologically following said check target slice using a slice header detected by said slice header detector.

16. The video decoding apparatus according to claim 13, wherein said first error detector determines, macroblock by macroblock, whether a direct current coefficient included in said discrete cosine transform coefficient data lies within a predetermined range, and identifies a macroblock which produces a direct current coefficient lying outside said predetermined range as an erroneous macroblock.

17. The video decoding apparatus according to claim 13, further comprising a second error detector for checking whether each of said slices of said each picture contains an error, based on said variable-length decoded data,

wherein when said check target slice in said picture currently undergoing decoding is in error, said error processing circuit replaces said check target slice with a preceding correct slice, said preceding correct slice included in said preceding decoded picture, and

wherein a position of said preceding correct slice in said preceding decoded picture corresponds to a position of said check target slice in said picture currently undergoing decoding.

18. The video decoding apparatus according to claim 17, wherein said second error detector determines whether data of each of said slices corresponds to Huffman codes in a Huffman table and also determines whether variable-length decoded data for a current check target slice is consistent with old variable-length decoded data and identifies a macroblock as erroneous if said data of each of said slices does not correspond to said Huffman codes or if said variable-length decoded data is inconsistent with said old variable-length decoded data.

19. The video decoding apparatus according to claim 18, wherein said error processing circuit inhibits operation of said variable length decoder on said check target slice which includes said erroneous macroblock, invalidates said variable-length decoded data for said check target slice, and restarts operation of said variable length decoder for a slice chronologically following said check target slice using a slice header detected by said slice header detector.

20. The video decoding apparatus according to claim 17, wherein said motion compensated prediction circuit includes a motion-vector restoring circuit for restoring data of a motion vector for each of said macroblocks;

and

said video decoding apparatus further comprising a third error detector for checking whether each of said slices of each of said pictures includes an erroneous macroblock based on said restored motion vector data, said third error detector performs the checking macroblock by macroblock.

21. The video decoding apparatus according to claim 20, wherein said third error detector determines whether said restored motion vector indicates a point inside or outside a picture, and identifies a macroblock corresponding to a motion vector indicating a point outside a picture as an erroneous macroblock.

22. The video decoding apparatus according to claim 13, wherein said motion compensated prediction circuit includes a motion-vector restoring circuit for restoring data of a motion vector for each of said macroblocks;

and

said video decoding apparatus further comprising a second error detector for checking whether each of said slices of each of said pictures includes an erroneous macroblock based on said restored motion vector data, said second error detector performs the checking macroblock by macroblock.

23. The video decoding apparatus according to claim 22, wherein said second error detector determines whether said restored motion vector indicates a point inside or outside a picture, and identifies a macroblock corresponding to a motion vector indicating a point outside a picture as an erroneous macroblock.