Processing

Please wait...

Settings

Settings

Goto Application

1. WO2021004887 - ENCODER, DECODER, METHODS AND COMPUTER PROGRAMS FOR AN IMPROVED LOSSLESS COMPRESSION

Note: Text based on automatic Optical Character Recognition processes. Please use the PDF version for legal matters

[ EN ]

Claims

1. Decoder for decoding a picture from a data stream, configured to

check (106) whether a plurality (98) of coding parameters, which are contained in the data stream (14), relate to a predetermined portion (100) of the picture (12) and control a prediction residual transform mode and a quantization accuracy with respect to the predetermined portion, are indicative of a coding parameter setting corresponding to a lossless prediction residual coding,

responsive to the plurality of coding parameters being indicative of the coding parameter setting corresponding to the lossless prediction residual coding, setting one or more predetermined coding options relating to one or more tools of the decoder for processing a prediction residual corrected predictive reconstruction with respect to the predetermined portion so that the one or more tools are disabled with respect to the predetermined portion.

2. Decoder of claim 1 , configured to

read from the data stream a lossless coding syntax element (102) which indicates whether a predetermined region (104) of the picture, which covers or contains the predetermined portion, is coded into the data stream using lossless coding or lossy coding, and

set the one or more predetermined coding options

so that the one or more tools are disabled with respect to the predetermined portion if the lossless coding syntax element indicates that the predetermined region of the picture is coded into the data stream using lossless coding, and if the plurality of coding parameters are indicative of the coding parameter setting corresponding to the lossless prediction residual coding and

to a predetermined tool state if the plurality of coding parameters not equal the coding parameter setting corresponding to the lossless prediction residual coding or the lossless coding syntax element indicates that the predetermined region of the picture is coded into the data stream using lossy coding.

3. Decoder of claim 2, configured to determine the predetermined tool state depending on one or more syntax elements (108) in the data stream.

4. Decoder of any of claims 1 to 3, configured to set one or more further coding options with respect to the predetermined portion to a default state responsive to the plurality of coding parameters being indicative of the coding parameter setting corresponding to the lossless prediction residual coding.

5. Decoder of claim 4, wherein the default state represents

a reduction of a filtering or a disabling of a filtering in terms of low-pass filtering for a derivation of a prediction signal (26) for the predetermined portion; and/or

a perfectly invertible transform to be performed on the prediction residual signal.

6. Decoder of claim 4 or 5, wherein the further coding options relate to

a binarization of prediction residual data into bin strings and context-adaptive binary entropy decoding of the bin strings;

a usage of a perfectly invertible transform on a prediction residual (24’", 24””) or a prediction residual corrected reconstruction (46) of the predetermined portion;

a disabling or reduction of a filtering for a derivation of a prediction signal (26) for the predetermined portion; and/or

a disabling of a processing of a prediction residual corrected predictive reconstruction with respect to the predetermined portion or a prediction residual re-quantization.

7. Decoder of any of claims 1 to 6, configured to

Decode a prediction residual from the data stream for the predetermined portion using binarization of prediction residual data into bin strings and context-adaptive binary entropy decoding of the bin strings in a first manner, if the plurality of coding parameters are indicative of the coding parameter setting corresponding to the lossless prediction residual coding, and in a second manner, if the plurality of coding parameters are not indicative of the coding parameter setting corresponding to the lossless prediction residual coding, wherein the first and second manners differ so that a computational complexity is reduced in the first manner compared to the second manner.

8. Decoder of any of claims 1 to 7, configured to

perform on a prediction residual (24’”, 24’”’) or a prediction residual corrected reconstruction (46) of the predetermined portion a perfectly invertible transform, if the plurality of coding parameters are indicative of the coding parameter setting corresponding to the lossless prediction residual coding, and a non-perfectly invertible transform, if the plurality of coding parameters are not indicative of the coding parameter setting corresponding to the lossless prediction residual coding.

9. Decoder of any of claims 1 to 8, configured to

derive a prediction signal (26) for the predetermined portion in a first manner, if the plurality of coding parameters are indicative of the coding parameter setting corresponding to the lossless prediction residual coding, and in a second manner, if the plurality of coding parameters are not indicative of the coding parameter setting corresponding to the lossless prediction residual coding, wherein the first and second manners differ so that the prediction signal is less filtered in the first manner than in the second manner or unfiltered in the first manner while being filtered in the second manner

10. Decoder of any of claims 1 to 9, configured to

infer that one or more predetermined coding options relating to one or more tools of the decoder for processing a prediction residual corrected predictive reconstruction with respect to the predetermined portion or for prediction residual re-quantization are to be set so that the one or more tools are disabled with respect to the predetermined portion, if the plurality of coding parameters are indicative of the coding parameter setting corresponding to the lossless prediction residual coding, and by deriving the one or more predetermined coding options from the plurality of coding parameters, if the plurality of

coding parameters are not indicative of the coding parameter setting corresponding to the lossless prediction residual coding.

11. Decoder of any of claims 1 to 10, configured to

set the one or more predetermined coding options with respect to the predetermined portion

so that the one or more tools are disabled with respect to the predetermined portion if the plurality of coding parameters are indicative of the coding parameter setting corresponding to the lossless prediction residual coding and

to a predetermined tool state if the plurality of coding parameters are not indicative of the coding parameter setting corresponding to the lossless prediction residual coding.

12. Decoder of claim 11 , configured to determine the predetermined tool state depending on one or more syntax elements in the data stream.

13. Decoder of claim 3 or 12, where at least one of the one or more syntax elements is absent from the data stream if the one or more predetermined coding options with respect to the predetermined portion are set so that the one or more tools are disabled with respect to the predetermined portion.

14. Decoder of any of claims 1 to 13, configured to

read the plurality of coding parameters from the data stream,

check whether the prediction residual transform mode indicated by the plurality of coding parameters corresponds to a transform skip mode and to a quantization step size finer than a predetermined quantization step size corresponding to no quantization, and, if yes, change the quantization step size to the predetermined quantization step size.

15. Decoder for decoding a picture from a data stream, configured to

determine for a predetermined portion of the picture, whether same is coded into the data stream using lossless coding or lossy coding, and

Decode a prediction residual from the data stream for the predetermined portion using binarization of prediction residual data into bin strings and context-adaptive binary entropy decoding of the bin strings in a first manner, if the predetermined portion of the picture is coded into the data stream using lossless coding, and in a second manner, if the predetermined portion of the picture is coded into the data stream using lossy coding, wherein the first and second manners differ so that a computational complexity is reduced in the first manner compared to the second manner.

16. Decoder of claim 15, configured so that the computational complexity is reduced in the first manner compared to the second manner by at least one of

a number of bins of the bin strings coded using equi-probability bypass mode is greater in the first manner than in the second manner, and

different binarization codes being used in the first and second manner.

17. Decoder of claim 16, configured so that

an amount of bins of the bin strings belonging to a unary code or truncated unary code of the bin strings is lower in the first manner than in the second manner.

18. Decoder for decoding a picture from a data stream, configured to

determine for a predetermined portion of the picture, whether same is coded into the data stream using lossless coding or lossy coding, and

perform on a prediction residual (24”’, 24””) or a prediction residual corrected reconstruction (46) of the predetermined portion a perfectly invertible transform, if the predetermined portion of the picture is coded into the data stream using lossless coding, and a non-perfectly invertible transform, if the predetermined portion of the picture is coded into the data stream using lossy coding.

19. Decoder of claim 18, wherein the perfectly invertible transform and the non-perfectly invertible transform are inter-color-component transforms or are spectrally decomposing intracolor-component transforms.

20. Decoder for decoding a picture from a data stream, configured to

determine, for a predetermined portion of the picture, whether same is coded into the data stream using lossless coding or lossy coding, and

derive a prediction signal (26) for the predetermined portion in a first manner, if the predetermined portion of the picture is coded into the data stream using lossless coding, and in a second manner, if the predetermined portion of the picture is coded into the data stream using lossy coding, wherein the first and second manners differ so that the prediction signal is less filtered in the first manner than in the second manner or unfiltered in the first manner while being filtered in the second manner.

21. Decoder of claim 20, wherein the prediction signal is less filtered in the first manner than in the second manner or unfiltered in the first manner while being filtered in the second manner in terms of low-pass filtering.

22. Decoder for decoding a picture from a data stream, configured to

determine, for a predetermined portion of the picture, whether same is coded into the data stream using lossless coding or lossy coding, and

infer that one or more predetermined coding options relating to one or more tools of the decoder for processing a prediction residual corrected predictive reconstruction with respect to the predetermined portion or for prediction residual re-quantization are to be set so that the one or more tools are disabled with respect to the predetermined portion, if the predetermined portion of the picture is coded into the data stream using lossless coding, and by deriving the one or more predetermined coding options from a plurality of coding parameters, if the predetermined portion of the picture is coded into the data stream using lossy coding.

23. Decoder of any of claims 15 to 22, configured to perform the determining based on the data stream.

24. Decoder of any of claims 15 to 23, configured to perform the determining by

Reading from the data stream a lossless coding syntax element which indicates whether the predetermined portion of the picture, or a predetermined region containing the predetermined portion, is coded into the data stream using lossless coding or lossy coding, and

Performing the determination depending on the lossless coding syntax element.

25. Decoder of any of claims 15 to 24 configured to perform the determining by

checking (106) whether a plurality (98) of coding parameters, which are contained in the data stream (14), relate to the predetermined portion (100) of the picture (12) and control a prediction residual transform mode and a quantization accuracy with respect to the predetermined portion, are indicative of a coding parameter setting corresponding to a lossless prediction residual coding,

determine whether the predetermined portion of the picture is coded into the data stream using lossless coding or lossy coding depending on whether the plurality (98) of coding parameters are indicative of the coding parameter setting corresponding to the lossless prediction residual coding.

26. Video decoder configured to

perform video decoding from a data stream at an internal bit-depth and video output at an input bit-depth or internal bit-depth,

read from the data stream a syntax element which indicates a difference between the internal bit-depth and the input bit-depth.

27. Video decoder of claim 26, configured to perform a bit-depth transition from an internally-reconstructed video version to the input bit-depth before the video output.

28. Video decoder of claim 26 or 27, configured to

deduce a minimum for a quantization step size scale parameter based on the difference, and

adhere to the minimum quantization step size scale parameter for portions of the video coded in a prediction residual transform skip mode.

29. Video decoder of any of claims 26 to 28 , configured to

in adhering to the minimum quantization step size scale parameter for video portions coded in a prediction residual transform skip mode, change a signaled quantization step size scale parameter signaled in the data stream for the video portions, to equal the minimum quantization step size scale parameter in case of the signaled quantization step size scale parameter falling below the minimum quantization step size scale parameter.

30. Video decoder of claim 28 or 29, configured to adhere to the minimum quantization step size scale parameter merely within video regions for which the data stream signals a lossless coding mode.

31. Video decoder configured to

perform video decoding from a data stream at an internal bit-depth and video output at an input bit-depth or interna! bit-depth,

read from the data stream a syntax element which indicates a minimum for a quantization step size scale parameter.

32 Video decoder of claim 31, configured to

adhere to the minimum quantization step size scale parameter for video portions coded in a prediction residual transform skip mode.

33. Video decoder of claim 31 or 32 , configured to

in adhering to the minimum quantization step size scale parameter for video portions coded in a prediction residual transform skip mode, change a signaled quantization step size scale parameter signaled in the data stream for the portions, to equal the minimum quantization step size scale parameter in case of the signaled quantization step size scale parameter falling below the minimum quantization step size scale parameter.

34. Video decoder of claim 32 or 33, configured to adhere to the minimum quantization step size scale parameter merely within video regions for which the data stream signals a lossless coding mode

35. Video decoder configured to

derive from a data stream video an indication of an internal bit-depth and an input bit-depth or a difference between same,

perform video decoding from the data stream at the internal bit-depth and video output at the input bit-depth,

check whether the internal bit-depth falls below the input bit-depth and change the internal bit-depth to correspond to the input bit-depth.

36. Video decoder of claim 35, configured to

derive from the data stream an indication of a lossless coded video portion, and

perform the checking and the changing within the lossless coded video portion and use the internal bit-depth as derived from the data stream for a lossy coded video portion.

37. Encoder for encoding a picture into a data stream, configured to

signal a plurality (98) of coding parameters in the data stream (14);

check (106) whether the plurality (98) of coding parameters, which relate to a predetermined portion (100) of the picture (12) and control a prediction residual transform mode and a quantization accuracy with respect to the predetermined portion, are indicative of a coding parameter setting corresponding to a lossless prediction residual coding,

responsive to the plurality of coding parameters being indicative of the coding parameter setting corresponding to the lossless prediction residual coding, setting one or more predetermined coding options relating to one or more tools of the encoder for processing a prediction residual corrected predictive reconstruction with respect to the predetermined portion so that the one or more tools are disabled with respect to the predetermined portion in a prediction-loop of the encoder.

38. Encoder of claim 37, configured to

encode into the data stream a lossless coding syntax element (102) which indicates whether a predetermined region (104) of the picture, which covers or contains the predetermined portion, is to be coded into the data stream using lossless coding or lossy coding, and

set the one or more predetermined coding options

so that the one or more tools are disabled with respect to the predetermined portion if the lossless coding syntax element indicates that the predetermined region of the picture is to be coded into the data stream using lossless coding, and if the plurality of coding parameters are indicative of the coding parameter setting corresponding to the lossless prediction residual coding and

to a predetermined tool state if the plurality of coding parameters not equal the coding parameter setting corresponding to the lossless prediction residual coding or the lossless coding syntax element indicates that the predetermined region of the picture is to be coded into the data stream using lossy coding.

39. Encoder of claim 38, configured to

determine and encode one or more syntax elements (108) in the data stream; and

determine the predetermined tool state depending on the one or more syntax elements (108).

40. Encoder of any of claims 37 to 39, configured to set one or more further coding options with respect to the predetermined portion to a default state responsive to the plurality of coding parameters being indicative of the coding parameter setting corresponding to the lossless prediction residual coding.

41. Encoder of claim 40, wherein the default state represents

a reduction of a filtering or a disabling of a filtering in terms of low-pass filtering for a derivation of a prediction signal (26) for the predetermined portion; and/or

a perfectly invertible transform to be performed on the prediction residual signal.

42. Encoder of claim 40 or 41 , wherein the further coding options relate to

a binarization of prediction residual data into bin strings and context-adaptive binary entropy encoding of the bin strings;

a usage of a perfectly invertible transform on a prediction residual (24’”, 24””) or a prediction residual corrected reconstruction (46) of the predetermined portion in a prediction-loop of the encoder;

a disabling or reduction of a filtering for a derivation of a prediction signal (26) for the predetermined portion; and/or

a disabling of a processing of a prediction residual corrected predictive reconstruction with respect to the predetermined portion or a prediction residual quantization.

43. Encoder of any of claims 37 to 42, configured to

encode a prediction residual into the data stream for the predetermined portion using binarization of prediction residual data into bin strings and context-adaptive binary entropy encoding of the bin strings in a first manner, if the plurality of coding parameters are indicative of the coding parameter setting corresponding to the lossless prediction residual coding, and in a second manner, if the plurality of coding parameters are not indicative of the coding parameter setting corresponding to the lossless prediction residual coding, wherein the first and second manners differ so that a computational complexity is reduced in the first manner compared to the second manner.

44. Encoder of any of claims 37 to 43, configured to

perform on a prediction residual (24’”, 24””) or a prediction residual corrected reconstruction (46) of the predetermined portion a perfectly invertible transform, if the plurality of coding parameters are indicative of the coding parameter setting corresponding to the lossless prediction residual coding, and a non-perfectly invertible transform, if the plurality of coding parameters are not indicative of the coding parameter setting corresponding to the lossless prediction residual coding.

45. Encoder of any of claims 37 to 44, configured to

encode a prediction signal (26) for the predetermined portion in a first manner, if the plurality of coding parameters are indicative of the coding parameter setting corresponding to the lossless prediction residual coding, and in a second manner, if the plurality of coding parameters are not indicative of the coding parameter setting corresponding to the lossless prediction residual coding, wherein the first and second manners differ so that the prediction signal is less filtered in the first manner than in the second manner or unfiltered in the first manner while being filtered in the second manner

46. Encoder of any of claims 37 to 45, configured such

infer that one or more predetermined coding options relating to one or more tools of the encoder for processing a prediction residual corrected predictive reconstruction with

respect to the predetermined portion or for prediction residual quantization are to be set so that the one or more tools are disabled with respect to the predetermined portion, if the plurality of coding parameters are indicative of the coding parameter setting corresponding to the lossless prediction residual coding, and by deriving the one or more predetermined coding options from the plurality of coding parameters, if the plurality of coding parameters are not indicative of the coding parameter setting corresponding to the lossless prediction residual coding.

47. Encoder of any of claims 37 to 46, configured to

set the one or more predetermined coding options with respect to the predetermined portion

so that the one or more tools are disabled with respect to the predetermined portion if the plurality of coding parameters are indicative of the coding parameter setting corresponding to the lossless prediction residual coding and

to a predetermined tool state if the plurality of coding parameters are not indicative of the coding parameter setting corresponding to the lossless prediction residual coding.

48. Encoder of claim 47, configured to

determine and encode one or more syntax elements (108) in the data stream; and

determine the predetermined tool state depending on the one or more syntax elements.

49 Encoder of claim 39 or 48, where at least one of the one or more syntax elements is not encoded into the data stream if the one or more predetermined coding options with respect to the predetermined portion are set so that the one or more tools are disabled with respect to the predetermined portion.

50. Encoder of any of claims 37 to 49, configured to

signal the plurality of coding parameters into the data stream

check whether the prediction residual transform mode indicated by the plurality of coding parameters corresponds to a transform skip mode and to a quantization step size finer than a predetermined quantization step size corresponding to no quantization, and, if yes, change the quantization step size to the predetermined quantization step size.

51. Encoder for encoding a picture into a data stream, configured to

determine for a predetermined portion of the picture, whether same is to be coded into the data stream using lossless coding or lossy coding, and

encode a prediction residual into the data stream for the predetermined portion using binarization of prediction residual data into bin strings and context-adaptive binary entropy encoding of the bin strings in a first manner, if the predetermined portion of the picture is to be coded into the data stream using lossless coding, and in a second manner, if the predetermined portion of the picture is to be coded into the data stream using lossy coding, wherein the first and second manners differ so that a computational complexity is reduced in the first manner compared to the second manner.

52. Encoder of claim 51 , configured so that the computational complexity is reduced in the first manner compared to the second manner by at least one of

a number of bins of the bin strings coded using equi-probability bypass mode is greater in the first manner than in the second manner, and

different binarization codes being used in the first and second manner

53. Encoder of claim 52, configured so that

an amount of bins of the bin strings belonging to a unary code or truncated unary code of the bin strings is lower in the first manner than in the second manner.

54. Encoder for encoding a picture into a data stream, configured to

determine for a predetermined portion of the picture, whether same is to be coded into the data stream using lossless coding or lossy coding, and

perform on a prediction residual (24'”, 24"") or a prediction residual corrected reconstruction (46) of the predetermined portion a perfectly invertible transform, if the predetermined portion of the picture is to be coded into the data stream using lossless coding, and a non-perfectly invertible transform, if the predetermined portion of the picture is to be coded into the data stream using lossy coding.

55. Encoder of claim 54, wherein the perfectly invertible transform and the non-perfectly invertible transform are inter-color-component transforms or are spectrally decomposing intra-color-component transforms.

56. Encoder for encoding a picture into a data stream, configured to

determine, for a predetermined portion of the picture, whether same is to be coded into the data stream using lossless coding or lossy coding, and

derive a prediction signal (26) for the predetermined portion in a first manner, if the predetermined portion of the picture is to be coded into the data stream using lossless coding, and in a second manner, if the predetermined portion of the picture is to be coded into the data stream using lossy coding, wherein the first and second manners differ so that the prediction signal is less filtered in the first manner than in the second manner or unfiltered in the first manner while being filtered in the second manner.

57. Encoder of claim 56, wherein the prediction signal is less filtered in the first manner than in the second manner or unfiltered in the first manner while being filtered in the second manner in terms of low-pass filtering.

58. Encoder for encoding a picture into a data stream, configured to

determine, for a predetermined portion of the picture, whether same is to be coded into the data stream using lossless coding or lossy coding, and

infer that one or more predetermined coding options relating to one or more tools of the encoder for processing a prediction residual corrected predictive reconstruction with respect to the predetermined portion or for prediction residual quantization are to be set so that the one or more tools are disabled with respect to the predetermined portion, if the predetermined portion of the picture is to be coded into the data stream using lossless coding, and by deriving the one or more predetermined coding options from a plurality of coding parameters, if the predetermined portion of the picture is to be coded into the data stream using lossy coding.

59. Encoder of any of claims 51 to 58, configured to signal the determining in the data stream.

60. Encoder of any of claims 51 to 59, configured to

encode into the data stream a lossless coding syntax element which indicates whether the predetermined portion of the picture, or a predetermined region containing the predetermined portion, is coded into the data stream using lossless coding or lossy coding, based on the determining.

61. Encoder of any of claims 51 to 60, configured to perform the determining by

checking (106) whether a plurality (98) of coding parameters relate to the predetermined portion (100) of the picture (12) and control a prediction residual transform mode and a quantization accuracy with respect to the predetermined portion, are indicative of a coding parameter setting corresponding to a lossless prediction residual coding,

signal the plurality (98) of coding parameters in the data stream (14);

determine whether the predetermined portion of the picture is coded into the data stream using lossless coding or lossy coding depending on whether the plurality (98) of coding parameters are indicative of the coding parameter setting corresponding to the lossless prediction residual coding.

62. Video encoder configured to

perform video encoding into a data stream at an internal bit-depth and receive video input at an input bit-depth or internal bit-depth,

encode into the data stream a syntax element which indicates a difference between the internal bit-depth and the input bit-depth.

63. Video encoder of claim 62, configured to perform a bit-depth transition from an internally-reconstructed video version (46) to the input bit-depth.

64. Video encoder of claim 62 or 63, configured to

deduce a minimum for a quantization step size scale parameter based on the difference, and

adhere to the minimum quantization step size scale parameter for portions of the video to be coded in a prediction residual transform skip mode,

65. Video encoder of any of claims 62 to 64, configured to

in adhering to the minimum quantization step size scale parameter for video portions to be coded in a prediction residual transform skip mode, change a quantization step size scale parameter for the video portions, to equal the minimum quantization step size scale parameter in case of the signaled quantization step size scale parameter falling below the minimum quantization step size scale parameter; and

signal the quantization step size scale parameter in the data stream.

66. Video encoder of claim 64 or 65, configured to adhere to the minimum quantization step size scale parameter merely within video regions for which a lossless coding mode is applied; and

signal the lossless coding mode in the data stream.

67. Video encoder configured to

perform video encoding into a data stream at an internal bit-depth and receive video input at an input bit-depth or internal bit-depth,

encode into the data stream a syntax element which indicates a minimum for a quantization step size scale parameter.

68. Video encoder of claim 67, configured to

adhere to the minimum quantization step size scale parameter for video portions to be coded in a prediction residua! transform skip mode.

69. Video encoder of claim 67 or 68, configured to

in adhering to the minimum quantization step size scale parameter for video portions to be coded in a prediction residual transform skip mode, change a signaled quantization step size scale parameter, to equal the minimum quantization step size scale parameter in case of the signaled quantization step size scale parameter falling below the minimum quantization step size scale parameter; and

signal the quantization step size scale parameter in the data stream.

70. Video encoder of claim 68 or 69, configured to adhere to the minimum quantization step size scale parameter merely within video regions for which a lossless coding mode is applied; and

signal the lossless coding mode in the data stream.

71. Video encoder configured to

encode into a data stream video an indication of an internal bit-depth and an input bit-depth or a difference between same,

perform video encoding into the data stream at the internal bit-depth and receive video input at the input bit-depth,

check whether the internal bit-depth falls below the input bit-depth and change the internal bit-depth to correspond to the input bit-depth.

72. Video encoder of claim 71 , configured to

encode into the data stream an indication of a lossless coded video portion, and

perform the checking and the changing within the lossless coded video portion and use the internal bit-depth for a lossy coded video portion; and

signal the internal bit-depth in the data stream.

73. Method for decoding a picture from a data stream, comprising:

checking (106) whether a plurality (98) of coding parameters, which are contained in the data stream (14), relate to a predetermined portion (100) of the picture (12) and control a prediction residual transform mode and a quantization accuracy with respect to the predetermined portion, are indicative of a coding parameter setting corresponding to a lossless prediction residual coding,

responsive to the plurality of coding parameters being indicative of the coding parameter setting corresponding to the lossless prediction residual coding, setting one or more predetermined coding options relating to one or more tools of the decoder for processing a prediction residual corrected predictive reconstruction with respect to the predetermined portion so that the one or more tools are disabled with respect to the predetermined portion.

74. Method for decoding a picture from a data stream, comprising:

determining for a predetermined portion of the picture, whether same is coded into the data stream using lossless coding or lossy coding, and

decoding a prediction residual from the data stream for the predetermined portion using binarization of prediction residual data into bin strings and context-adaptive binary entropy decoding of the bin strings in a first manner, if the predetermined portion of the picture is coded into the data stream using lossless coding, and in a second manner, if the predetermined portion of the picture is coded into the data stream using lossy coding, wherein the first and second manners differ so that a computational complexity is reduced in the first manner compared to the second manner.

75. Method for decoding a picture from a data stream, comprising:

determining for a predetermined portion of the picture, whether same is coded into the data stream using lossless coding or lossy coding, and

performing on a prediction residual (24'", 24””) or a prediction residua! corrected reconstruction (46) of the predetermined portion a perfectly invertible transform, if the predetermined portion of the picture is coded into the data stream using lossless coding, and a non-perfectly invertible transform, if the predetermined portion of the picture is coded into the data stream using lossy coding.

76. Method for decoding a picture from a data stream, comprising:

determining, for a predetermined portion of the picture, whether same is coded into the data stream using lossless coding or lossy coding, and

deriving a prediction signal (26) for the predetermined portion in a first manner, if the predetermined portion of the picture is coded into the data stream using lossless coding, and in a second manner, if the predetermined portion of the picture is coded into the data stream using lossy coding, wherein the first and second manners differ so that the prediction signal is less filtered in the first manner than in the second manner or unfiltered in the first manner while being filtered in the second manner.

77. Method for decoding a picture from a data stream, comprising:

determining, for a predetermined portion of the picture, whether same is coded into the data stream using lossless coding or lossy coding, and

infering that one or more predetermined coding options relating to one or more tools of the decoder for processing a prediction residual corrected predictive reconstruction with respect to the predetermined portion or for prediction residual re-quantization are to be set so that the one or more tools are disabled with respect to the predetermined portion, if the predetermined portion of the picture is coded into the data stream using lossless coding, and by deriving the one or more predetermined coding options from the plurality of coding parameters, if the predetermined portion of the picture is coded into the data stream using lossy coding.

78. Method, comprising:

performing video decoding from a data stream at an internal bit-depth and video outputing at an input bit-depth or internal bit-depth,

reading from the data stream a syntax element which indicates a difference between the internal bit-depth and the input bit-depth.

79. Method, comprising:

performing video decoding from a data stream at an internal bit-depth and video outputing at an input bit-depth or internal bit-depth,

reading from the data stream a syntax element which indicates a minimum for a quantization step size scale parameter.

80. Method, comprising:

deriving from a data stream video an indication of an internal bit-depth and an input bit-depth or a difference between same,

performing video decoding from the data stream at the internal bit-depth and video outputing at the input bit-depth,

checking whether the internal bit-depth falls below the input bit-depth and changing the internal bit-depth to correspond to the input bit-depth.

81 . Method for encoding a picture into a data stream, comprising:

signaling a plurality (98) of coding parameters in the data stream (14);

checking (106) whether the plurality (98) of coding parameters relate to a predetermined portion (100) of the picture (12) and control a prediction residual transform mode and a quantization accuracy with respect to the predetermined portion, are indicative of a coding parameter setting corresponding to a lossless prediction residual coding,

responsive to the plurality of coding parameters being indicative of the coding parameter setting corresponding to the lossless prediction residual coding, setting one or more predetermined coding options relating to one or more tools of the encoder for processing a prediction residual corrected predictive reconstruction with respect to the predetermined portion so that the one or more tools are disabled with respect to the predetermined portion in a prediction-loop of the encoder.

82. Method for encoding a picture into a data stream, comprising;

determining for a predetermined portion of the picture, whether same is to be coded into the data stream using lossless coding or lossy coding, and

encoding a prediction residual into the data stream for the predetermined portion using binarization of prediction residual data into bin strings and context-adaptive binary entropy encoding of the bin strings in a first manner, if the predetermined portion of the picture is to be coded into the data stream using lossless coding, and in a second manner, if the predetermined portion of the picture is to be coded into the data stream using lossy coding, wherein the first and second manners differ so that a computational complexity is reduced in the first manner compared to the second manner.

83. Method for encoding a picture into a data stream, comprising:

determining for a predetermined portion of the picture, whether same is to be coded into the data stream using lossless coding or lossy coding, and

performing on a prediction residual (24’”, 24””) or a prediction residual corrected reconstruction (46) of the predetermined portion a perfectly invertible transform, if the predetermined portion of the picture is to be coded into the data stream using lossless coding, and a non-perfectly invertible transform, if the predetermined portion of the picture is to be coded into the data stream using lossy coding.

84. Method for encoding a picture into a data stream, comprising:

determining, for a predetermined portion of the picture, whether same is to be coded into the data stream using lossless coding or lossy coding, and

deriving a prediction signal (26) for the predetermined portion in a first manner, if the predetermined portion of the picture is to be coded into the data stream using lossless coding, and in a second manner, if the predetermined portion of the picture is to be coded into the data stream using lossy coding, wherein the first and second manners differ so that the prediction signal is less filtered in the first manner than in the second manner or unfiltered in the first manner while being filtered in the second manner.

85. Method for encoding a picture into a data stream, comprising:

determining, for a predetermined portion of the picture, whether same is to be coded into the data stream using lossless coding or lossy coding, and

infering that one or more predetermined coding options relating to one or more tools of the encoder for processing a prediction residual corrected predictive reconstruction with respect to the predetermined portion or for prediction residual quantization are to be set so that the one or more tools are disabled with respect to the predetermined portion, if the predetermined portion of the picture is coded into the data stream using lossless coding, and by deriving the one or more predetermined coding options from the plurality of coding parameters, if the predetermined portion of the picture is coded into the data stream using lossy coding.

86. Method, comprising:

performing video encoding into a data stream at an internal bit-depth and receiving video input at an input bit-depth or internal bit depth,

encoding into the data stream a syntax element which indicates a difference between the internal bit-depth and the input bit-depth.

87. Method, comprising:

performing video encoding into a data stream at an internal bit-depth and receiving video input at an input bit-depth or internal bit-depth,

encoding into the data stream a syntax element which indicates a minimum for a quantization step size scale parameter.

88. Method, comprising:

encoding into a data stream video an indication of an internal bit-depth and an input bit-depth or a difference between same,

performing video encoding into the data stream at the internal bit-depth and receiving video input at the input bit-depth,

checking whether the internal bit-depth falls below the input bit-depth and changing the internal bit-depth to correspond to the input bit-depth.

89. A computer program for implementing the method of at least one of claims 100 to 115.

90. A data stream obtainable by the method of at least one of claims 108 to 115.