Transcoder for coded video

• US 7,151,799 B2
• Filed: 03/11/2003
• Issued: 12/19/2006
• Est. Priority Date: 03/14/2002
• Status: Expired due to Fees

• Current Assignee: KDDI Corporation
• Original Assignee: KDDI Corporation
• Inventors: Nakajima, Yasuyuki, Kato, Haruhisa
• Primary Examiner(s): PHILIPPE, GIMS S
• Application Number:
  US10/384,708
• Publication Number:
  US 20030174770A1
• Time in Patent Office: 1,379 Days
• Field of Search: 375/240.16, 375/240.02, 375/240.26, 375/240.1, 382/248, 382/236, 348/400.1, 348/397
• US Class Current: 375/240.16
• CPC Class Codes: H04N 19/132 Sampling, masking or truncation of coding units, e.g. adaptive resampling, frame skipping, frame interpolation or high-frequency transform coefficient masking
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/18 the unit being a set of transform coefficients
• CPC Class Codes: H04N 19/40 using video transcoding, i.e. partial or full decoding of a coded input stream followed by re-encoding of the decoded output stream
• CPC Class Codes: H04N 19/503 involving temporal prediction adaptive coding with adaptive selection between spatial and temporal predictive coding H04N19/107; adaptive coding with adaptive selection among a plurality of temporal predictive coding modes H04N19/109
• CPC Class Codes: H04N 19/59 involving spatial sub-sampling or interpolation, e.g. alteration of picture size or resolution

Abstract

Provided is a transcoder for coded video the encoding mode of that is changed according to the coded data, itself, that has been compressed by transcoding, or information prepared by decoding only a part thereof. A coded data extractor partially decodes DV coded data (a) that has been input, and sends DCT coefficients (b) to an approximate image generator and to a decoder, and outputs quantization parameters (c) to a first motion vector detector. The approximate image generator generates an approximate image according to a part of the DCT coefficients (b). A first motion vector detector roughly detects motion vector candidates (e). A second motion vector detector, using the motion vector candidates (e) as an initial position, hierarchically detects a motion vector (f). A re-encoder encodes the data into one the encoding mode of that is new, according to a luminance signal (h) that has been decoded in the decoder, the chrominance signal (j) the data format of that has been transcoded in a sampling format transcoder, and the motion vector (f) that has been detected in the second motion vector detector.

Claims

1. A transcoder for coded video, which performs transcoding an encoding mode, and outputting, comprising:

a coded data extractor that partially decodes coded information of compressed video data and extracts it;

an approximate image generator that generates an approximate image from the coded information that has been extracted;

a first motion vector detector that performs detection of motion vector candidates from the approximate image;

a decoder that decodes the coded information and generates a luminance signal and a chrominance signal;

a second motion vector detector that performs detection of a motion vector on the basis of the luminance signal that has been produced in the decoder and the motion vector candidates that have been detected in the first motion vector detector; and

a re-encoder that performs re-encoding of the data on the basis of the motion vector information that has been detected in the second motion vector detector, the luminance signal that has been decoded in the decoder, and the chrominance signal the data format of that has been transcoded.

3. A transcoder for coded video according to claim 1, wherein

input information to the coded data extractor is video data that has been compressed by only intra-frame encoding;

and

output information from the re-encoder is video data that has been compression-encoded in a hybrid mode of using intra-frame encoding and inter-frame encoding techniques.

5. A transcoder for coded video according to claim 3, wherein

the chrominance signal of the input information is one of a 4:

1;

1 signal that is obtained by horizontally sub-sampling to ¼

in comparison with the luminance signal, and a 4;

2;

0 signal that is obtained by horizontally and vertically sub-sampling to ½

in comparison with the luminance signal.

7. A transcoder for coded video according to claim 3, wherein

the chrominance signal of the output information is one of a 4:

2;

0 signal that is obtained by horizontally and vertically sub-sampling to ½

in comparison with the luminance signal, and a 4;

2;

2 signal that is obtained by horizontally sub-sampling to ½

in comparison with the luminance signal.

9. A transcoder for coded video according to claim 1, wherein

the coded data extractor extracts transcoding coefficients and quantization parameters from the compressed video data that has been input, outputs the transcoding coefficients to the approximate image generator and to the decoder, and outputs the quantization parameters to the first motion vector detector.

11. A transcoder for coded video according to claim 9, wherein

when transcoding data is constructed for each field the coded data extractor, outputs encoding coefficients of the sums of, and the differences between, the transcoding coefficients of a first field stored and the transcoding coefficients of a second field stored.

13. A transcoder for coded video according to claim 1, wherein

the approximate image generator generates an approximate image by decoding only low frequency components of the transcoding coded data signal.

15. A transcoder for coded video according to claim 13, wherein

when transcoding data is constructed for each field the approximate image generator, selects a direct-current component and three biased-toward-low-frequency alternating-current components from the sums of the coefficients of the first field and those of the second field stored.

17. A transcoder for coded video according to claim 1, wherein

the approximate image generator generates an approximate image by performing inverse transcoding for 2×

2 coefficients, with respect to the four selected coefficients.

19. A transcoder for coded video according to claim 17, wherein

the approximate image generator, in either case of the transcoding data constructed for each frame and that constructed for each field, performs inverse transcoding by inverse transcoding that uses the same transcoding base to thereby produce an approximate image.

21. A transcoder for coded video according to claim 1, wherein the first motion vector detector performs a motion vector detection on the basis of the motion vector detection result obtained from an approximate image of low resolution and the second motion vector detector performs motion vector searching with a high resolution so that the second motion vector detection is performed by hierarchically utilizing a plurality of images different in resolution.

23. A transcoder for coded video according to claim 21, wherein,

when the degree of fineness of the quantizer for a respective macro-block is high, the first motion vector detector has the use thereof omitted.

25. A transcoder for coded video according to claim 1, wherein

the first motion vector detector changes the number of the motion vector candidates on the basis of the quantization parameters sent from the coded data extractor.

27. A transcoder for coded video according to claim 25, wherein

the first motion vector detector increases the number of the motion vector candidates as the degree of fineness of the quantizer for a each macro-block becomes low.

29. A transcoder for coded video according to claim 1, wherein

the sampling format transcoder is constructed with combinations of horizontally and vertically enlargement and reduction processings.

31. A transcoder for coded video according to claim 29, wherein,

when transcoding a 4:

1;

1 signal to a 4;

2;

0 signal, the sampling format transcoder first vertically reduces and then horizontally enlarges the 4;

1;

1 signal;

when transcoding the 4;

1;

1 signal to a 4;

2;

2 signal, the sampling format transcoder horizontally enlarges the 4;

1;

1 signal; and

when transcoding the 4;

2;

0 signal to the 4;

2;

2 signal, the sampling format transcoder vertically enlarges the 4;

2;

0 signal.

33. A transcoder for coded video according to claim 29, wherein,

when the input and output of the sampling format transcoder are each a 4:

2;

0 signal, the sampling format transcoder does not execute any processing.

35. A transcoder for coded video according to claim 1, wherein

the second motion vector detector is inputted with the motion vector candidates from the first motion vector detector and the luminance signal from the decoder to detect a motion vector, and output the motion vector to the re-encoder.

37. A transcoder for coded video according to claim 35, wherein

the motion vector previously detected is added as the motion vector candidates.

39. A transcoder for coded video according to claim 35, wherein

when a plurality of the motion vector candidates exist, the second motion vector detector uses the point at which the sum of the absolute values of the differences becomes minimal as an initial value.

41. A transcoder for coded video according to claim 35, wherein

the second motion vector detector determines the motion vector candidate that has been detected by the first motion vector detector to be an initial value for performing searching operation and performs detection of the motion vector by using the initial value as a center.

43. A transcoder for coded video according to claim 41, wherein,

in a block matching that is performed in the second motion vector detector, in the process of calculation of the sum of the absolute values of the differences that is performed for each block, the moment when the current sum value that has been obtained has just exceeded the minimum value of the sum of the absolute values of the differences, up to the present, calculation of that block is interrupted and it is determined that that sum value is inadequate.

45. A transcoder for coded video according to claim 43, wherein

in the block matching, scanning of the search position is performed for each two pixels, and the unit of scanning is made stepwise smaller up to a lesser number of pixel components.

47. A transcoder for coded video according to claim 43, wherein

in the block matching, a position at which the sum of the absolute values of the differences has a minimum value is detected from the initial position and neighborhood 8 points.

49. A transcoder for coded video according to claim 43, wherein

in the block matching, a point at which the sum has a minimum value is used as a new initial position, block matching is repeatedly performed between the new initial position and its neighborhood 8 points, and, when a minimum value does not exist within a range of search and in the vicinity, new initial position is determined to be the motion vector.

2. A transcoder for coded video, which performs transcoding an encoding mode, and outputting, comprising:

a coded data extractor that partially decodes coded information of compressed video data and extracts it;

an approximate image generator that generates an approximate image from the coded information that has been extracted;

a first motion vector detector that performs detection of motion vector candidates from the approximate image;

a decoder that decodes the coded information and generates a luminance signal and a chrominance signal;

a sampling format transcoder that has input thereto the chrominance signal output from the decoder and performs transcoding of the data format of the chrominance signal and outputs the transcoded data signal to the re-encoder;

a second motion vector detector that performs detection of a motion vector on the basis of the luminance signal that has been produced in the decoder and the motion vector candidates that have been detected in the first motion vector detector; and

a re-encoder that performs re-encoding of the data on the basis of the motion vector information that has been detected in the second motion vector detector, the luminance signal that has been decoded in the decoder, and the chrominance signal the data format of that has been transcoded.

4. A transcoder for coded video according to claim 2, wherein

input information to the coded data extractor is video data that has been compressed by only intra-frame encoding;

and

output information from the re-encoder is video data that has been compression-encoded in a hybrid mode of using intra-frame encoding and inter-frame encoding techniques.

6. A transcoder for coded video according to claim 4, wherein

the chrominance signal of the input information is one of a 4:

1;

1 signal that is obtained by horizontally sub-sampling to ¼

in comparison with the luminance signal, and a 4;

2;

0 signal that is obtained by horizontally and vertically sub-sampling to ½

in comparison with the luminance signal.

8. A transcoder for coded video according to claim 4, wherein

the chrominance signal of the output information is one of a 4:

2;

0 signal that is obtained by horizontally and vertically sub-sampling to ½

in comparison with the luminance signal, and a 4;

2;

2 signal that is obtained by horizontally sub-sampling to ½

in comparison with the luminance signal.

10. A transcoder for coded video according to claim 2, wherein

the coded data extractor extracts transcoding coefficients and quantization parameters from the compressed video data that has been input, outputs the transcoding coefficients to the approximate image generator and to the decoder, and outputs the quantization parameters to the first motion vector detector.

12. A transcoder for coded video according to claim 10, wherein

when transcoding data is constructed for each field the coded data extractor, outputs encoding coefficients of the sums of, and the differences between, the transcoding coefficients of a first field stored and the transcoding coefficients of a second field stored.

14. A transcoder for coded video according to claim 2, wherein

the approximate image generator generates an approximate image by decoding only low frequency components of the transcoding coded data signal.

16. A transcoder for coded video according to claim 14, wherein

when transcoding data is constructed for each field the approximate image generator, selects a direct-current component and three biased-toward-low-frequency alternating-current components from the sums of the coefficients of the first field and those of the second field stored.

18. A transcoder for coded video according to claim 2, wherein

the approximate image generator generates an approximate image by performing inverse transcoding for 2×

2 coefficients, with respect to the four selected coefficients.

20. A transcoder for coded video according to claim 18, wherein

the approximate image generator, in either case of the transcoding data constructed for each frame and that constructed for each field, performs inverse transcoding by inverse transcoding that uses the same transcoding base to thereby produce an approximate image.

22. A transcoder for coded video according to claim 2, wherein the first motion vector detector performs a motion vector detection on the basis of the motion vector detection result obtained from an approximate image of low resolution and the second motion vector detector performs motion vector searching with a high resolution so that the second motion vector detection is performed by hierarchically utilizing a plurality of images different in resolution.

24. A transcoder for coded video according to claim 22, wherein,

when the degree of fineness of the quantizer for a respective macro-block is high, the first motion vector detector has the use thereof omitted.

26. A transcoder for coded video according to claim 2, wherein

the first motion vector detector changes the number of the motion vector candidates on the basis of the quantization parameters sent from the coded data extractor.

28. A transcoder for coded video according to claim 26, wherein

the first motion vector detector increases the number of the motion vector candidates as the degree of fineness of the quantizer for a each macro-block becomes low.

30. A transcoder for coded video according to claim 2, wherein

the sampling format transcoder is constructed with combinations of horizontally and vertically enlargement and reduction processings.

32. A transcoder for coded video according to claim 30, wherein,

when transcoding a 4:

1;

1 signal to a 4;

2;

0 signal, the sampling format transcoder first vertically reduces and then horizontally enlarges the 4;

1;

1 signal;

when transcoding the 4;

1;

1 signal to a 4;

2;

2 signal, the sampling format transcoder horizontally enlarges the 4;

1;

1 signal; and

when transcoding the 4;

2;

0 signal to the 4;

2;

2 signal, the sampling format transcoder vertically enlarges the 4;

2;

0 signal.

34. A transcoder for coded video according to claim 30, wherein,

when the input and output of the sampling format transcoder are each a 4:

2;

0 signal, the sampling format transcoder does not execute any processing.

36. A transcoder for coded video according to claim 2, wherein

the second motion vector detector is inputted with the motion vector candidates from the first motion vector detector and the luminance signal from the decoder to detect a motion vector, and output the motion vector to the re-encoder.

38. A transcoder for coded video according to claim 36, wherein

the motion vector previously detected is added as the motion vector candidates.

40. A transcoder for coded video according to claim 36, wherein

when a plurality of the motion vector candidates exist, the second motion vector detector uses the point at which the sum of the absolute values of the differences becomes minimal as an initial value.

42. A transcoder for coded video according to claim 36, wherein

the second motion vector detector determines the motion vector candidate that has been detected by the first motion vector detector to be an initial value for performing searching operation and performs detection of the motion vector by using the initial value as a center.

44. A transcoder for coded video according to claim 42, wherein,

in a block matching that is performed in the second motion vector detector, in the process of calculation of the sum of the absolute values of the differences that is performed for each block, the moment when the current sum value that has been obtained has just exceeded the minimum value of the sum of the absolute values of the differences, up to the present, calculation of that block is interrupted and it is determined that that sum value is inadequate.

46. A transcoder for coded video according to claim 44, wherein

in the block matching, scanning of the search position is performed for each two pixels, and the unit of scanning is made stepwise smaller up to a lesser number of pixel components.

48. A transcoder for coded video according to claim 44, wherein

in the block matching, a position at which the sum of the absolute values of the differences has a minimum value is detected from the initial position and neighborhood 8 points.

50. A transcoder for coded video according to claim 44, wherein

in the block matching, a point at which the sum has a minimum value is used as a new initial position, block matching is repeatedly performed between the new initial position and its neighborhood 8 points, and, when a minimum value does not exist within a range of search and in the vicinity, new initial position is determined to be the motion vector.