Method, apparatus and system for encoding and decoding a subset of transform units of encoded video data

1. A decoding method for decoding a subset of a plurality of sub-sets of a transform unit of encoded video data using a Golomb-Rice method, the decoding method comprising:

determining, in a unit of the subset of the plurality of sub-sets of the transform unit, an initial value of a Rice parameter for decoding a current subset of the plurality of sub-sets of the transform unit, using statistics of previously decoded residual coefficient values for a previous sub-set of the plurality of sub-sets of the transform unit, each of the subsets comprising a plurality of residual coefficient values being arranged as an array of four residual coefficient values by four residual coefficient values, wherein the transform unit is a unit for transform between a frequency-domain and a spatial-domain, and the transform unit is scanned on a subset-by-subset basis;

performing initialization which initializes, in a unit of the subset of the plurality of sub-sets of the transform unit, the Rice parameter to the determined initial value; and

decoding the current subset of the plurality of sub-sets of the transform unit using the Golomb-Rice method, wherein the initialized Rice parameter is used as an initial Rice parameter for decoding the current subset, and wherein a modified Rice parameter, modified from the initialized Rice parameter by using a method different from the initialization, is used as a Rice parameter for decoding the current subset, wherein the method for modification uses a last coeff abs level remaining which is a remaining absolute value of a coefficient level that is coded with Golomb-Rice code.

2. The decoding method according to the claim 1 , wherein the initialized Rice parameter is determined from information decoded for the transform unit prior to decoding of the current sub-set.

3. The decoding method according to the claim 2 , wherein the initialized Rice parameter is offset from a zero setting when a quantization parameter is lower than a predetermined threshold.

4. The decoding method according to the claim 1 , wherein the initialized Rice parameter is determined from information decoded for the current subset of the transform unit.

5. The decoding method according to the claim 4 , wherein the initialized Rice parameter is offset from a zero setting when a number of significant residual coefficient values in the current subset is higher than a predetermined threshold.

6. An apparatus for decoding a subset of a plurality of sub-sets of a transform unit of encoded video data using a Golomb-Rice method, the apparatus comprising: a determining unit configured to determine an initial value of a Rice parameter for decoding a current subset of the plurality of sub-sets of the transform unit, using statistics of previously decoded residual coefficient values for a previous sub-set of the plurality of sub-sets of the transform unit, each of the subsets comprising a plurality of residual coefficient values being arranged as an array of four residual coefficient values by four residual coefficient values, wherein the transform unit is a unit for transform between a frequency-domain and a spatial-domain, and the transform unit is scanned on a subset-by-subset basis; an initialization unit configured to initialize the Rice parameter to the determined initial value; and a decoding unit configured to decode the current subset of the plurality of sub-sets of the transform unit using the Golomb-Rice method, wherein the initialized Rice parameter is used as an initial Rice parameter for decoding the current subset, and wherein a modified Rice parameter, modified from the initialized Rice parameter by using a method different from the initialization, is used as a Rice parameter for decoding the current subset, wherein the method for modification uses a last coeff_abs_level_remaining which is a remaining absolute value of a coefficient level that is coded with Golomb-Rice code.

7. A system for decoding a subset of a plurality of sub-sets of a transform unit of encoded video data using a Golomb-Rice method, the system comprising:

a memory for storing data and a computer program;

a processor coupled to said memory for executing said computer program, said computer program comprising code for:

determining, in a unit of the subset of the plurality of sub-sets of the transform unit, an initial value of a Rice parameter for decoding a current subset of the plurality of sub-sets of the transform unit, using statistics of previously decoded residual coefficient values for a previous sub-set of the plurality of sub-sets of the transform unit, each of the subsets comprising a plurality of residual coefficient values being arranged as an array of four residual coefficient values by four residual coefficient values, wherein the transform unit is a unit for transform between a frequency-domain and a spatial-domain, and the transform unit is scanned on a subset-by-subset basis; and

performing initialization which initializes, in a unit of the subset of the plurality of sub-sets of the transform unit, the Rice parameter to the determined initial value; and

decoding the current subset of the plurality of sub-sets of the transform unit using the Golomb-Rice method, wherein the initialized Rice parameter is used as an initial Rice parameter for decoding the current subset, and wherein a modified Rice parameter, modified from the initialized Rice parameter by using a method different from the initialization, is used as a Rice parameter for decoding the current subset, wherein the method for modification uses a last coeff abs level remaining which is a remaining absolute value of a coefficient level that is coded with Golomb-Rice code.

8. A non-transitory computer readable medium having a computer program recorded thereon for decoding a subset of a plurality of sub-sets of a transform unit of encoded video data using a Golomb-Rice method, the program comprising:

code for determining, in a unit of the subset of the plurality of sub-sets of the transform unit, an initial value of a Rice parameter for decoding a current subset of the plurality of sub-sets of the transform unit, using statistics of previously decoded residual coefficient values for a previous sub-set of the plurality of sub-sets of the transform unit, each of the subsets comprising a plurality of residual coefficient values being arranged as an array of four residual coefficient values by four residual coefficient values, wherein the transform unit is a unit for transform between a frequency-domain and a spatial-domain, and the transform unit is scanned on a subset-by-subset basis;

code for performing initialization which initializes, in a unit of the subset of the plurality of sub-sets of the transform unit, the Rice parameter to the determined initial value; and

code for decoding the current subset of the plurality of sub-sets of the transform unit using the Golomb-Rice method, wherein the initialized Rice parameter is used as an initial Rice parameter for decoding the current subset, and wherein a modified Rice parameter, modified from the initialized Rice parameter by using a method different from the initialization, is used as a Rice parameter for decoding the current subset, wherein the method for modification uses a last coeff abs level remaining which is a remaining absolute value of a coefficient level that is coded with Golomb-Rice code.

9. An encoding method for encoding a subset of a plurality of sub-sets of a transform unit of video data using a Golomb-Rice method, the encoding method comprising:

determining, in a unit of the subset of the plurality of sub-sets of the transform unit, an initial value of a Rice parameter for encoding a current subset of the plurality of sub-sets of the transform unit, using statistics of previously encoded residual coefficient values for a previous sub-set of the plurality of sub-sets of the transform unit, each of the subsets comprising a plurality of residual coefficient values being arranged as an array of four residual coefficient values by four residual coefficient values, wherein the transform unit is a unit for transform between a frequency-domain and a spatial-domain, and the transform unit is scanned on a subset-by-subset basis;

performing initialization which initializes, in a unit of the subset of the plurality of sub-sets of the transform unit, the Rice parameter to the determined initial value; and

encoding the current subset of the plurality of sub-sets of the transform unit using the Golomb-Rice method, wherein the initialized Rice parameter is used as an initial Rice parameter for encoding the current subset, and wherein a modified Rice parameter, modified from the initialized Rice parameter by using a method different from the initialization, is used as a Rice parameter for encoding the current subset, wherein the method for modification uses a last coeff abs level remaining which is a remaining absolute value of a coefficient level that is coded with Golomb-Rice code.

10. An apparatus for encoding a subset of a plurality of sub-sets of a transform unit of video data using a Golomb-Rice method, the apparatus comprising:

a determining unit configured to determine, in a unit of the subset of the plurality of sub-sets of the transform unit, an initial value of a Rice parameter for encoding a current subset of the plurality of sub-sets of the transform unit, using statistics of previously encoded residual coefficient values for a previous sub-set of the plurality of sub-sets of the transform unit, each of the subsets comprising a plurality of residual coefficient values being arranged as an array of four residual coefficient values by four residual coefficient values, wherein the transform unit is a unit for transform between a frequency-domain and a spatial-domain, and the transform unit is scanned on a subset-by-subset basis;

an initialization unit configured to perform initialization which initializes, in a unit of the subset of the plurality of sub-sets of the transform unit, the Rice parameter to the determined initial value; and

an encoding unit configured to encode the current subset of the plurality of sub-sets of the transform unit using the Golomb-Rice method, wherein the initialized Rice parameter is used as an initial Rice parameter for encoding the current subset, and wherein a modified Rice parameter, modified from the initialized Rice parameter by using a method different from the initialization, is used as a Rice parameter for encoding the current subset, wherein the method for modification uses a last coeff abs level remaining which is a remaining absolute value of a coefficient level that is coded with Golomb-Rice code.

11. A system for encoding a subset of a plurality of sub-sets of a transform unit of video data using a Golomb-Rice method, the system comprising:

a memory for storing data and a computer program;

a processor coupled to said memory for executing said computer program, said computer program comprising code for:

determining, in a unit of the subset of the plurality of sub-sets of the transform unit, an initial value of a Rice parameter for encoding a current subset of the plurality of sub-sets of the transform unit, using statistics of previously encoded residual coefficient values for a previous sub-set of the plurality of sub-sets of the transform unit, each of the subsets comprising a plurality of residual coefficient values being arranged as an array of four residual coefficient values by four residual coefficient values, wherein the transform unit is a unit for transform between a frequency-domain and a spatial-domain, and the transform unit is scanned on a subset-by-subset basis; and

performing initialization which initializes, in a unit of the subset of the plurality of sub-sets of the transform unit, the Rice parameter to the determined initial value; and

encoding the current subset of the plurality of sub-sets of the transform unit using the Golomb-Rice method, wherein the initialized Rice parameter is used as an initial Rice parameter for encoding the current subset, and wherein a modified Rice parameter, modified from the initialized Rice parameter by using a method different from the initialization, is used as a Rice parameter for encoding the current subset, wherein the method for modification uses a last coeff abs level remaining which is a remaining absolute value of a coefficient level that is coded with Golomb-Rice code.

12. A non-transitory computer readable medium having a computer program recorded thereon for encoding a subset of a plurality of sub-sets of a transform unit of video data using a Golomb-Rice method, the program comprising:

code for determining, in a unit of the subset of the plurality of sub-sets of the transform unit, an initial value of a Rice parameter for encoding a current subset of the plurality of sub-sets of the transform unit, using statistics of previously encoded residual coefficient values for a previous sub-set of the plurality of sub-sets of the transform unit, each of the subsets comprising a plurality of residual coefficient values being arranged as an array of four residual coefficient values by four residual coefficient values, wherein the transform unit is a unit for transform between a frequency-domain and a spatial-domain, and the transform unit is scanned on a subset-by-subset basis;

code for performing initialization which initializes, in a unit of the subset of the plurality of sub-sets of the transform unit, the Rice parameter to the determined initial value; and

code for encoding the current subset of the plurality of sub-sets of the transform unit using the Golomb-Rice method, wherein the initialized Rice parameter is used as an initial Rice parameter for encoding the current subset, and wherein a modified Rice parameter, modified from the initialized Rice parameter by using a method different from the initialization, is used as a Rice parameter for encoding the current subset, wherein the method for modification uses a last coeff abs level remaining which is a remaining absolute value of a coefficient level that is coded with Golomb-Rice code.