Video coding method and apparatus using a conversion skip flag

Abstract

To provide an image coding method and device using a transform skip flag in an image coding system.SOLUTION: An image decoding method includes the steps of: obtaining prediction mode information and residual related information from a bitstream; deriving prediction samples of a current block by performing prediction based on the prediction mode information; deriving residual samples of the current block based on the residual related information; and generating reconstruction samples of the current block based on the prediction samples and the residual samples. The residual related information includes a transform skip flag based on a size of the current block and a maximum transform skip size, the transform skip flag represents whether a transform skip is applied to the current block, and information about the maximum transform skip size is obtained from the bitstream.SELECTED DRAWING: Figure 15

Claims

1. In a method for decoding video performed by a decoding device, A step of obtaining prediction mode information and residual-related information from a bitstream, The steps include: performing a prediction based on the aforementioned prediction mode information to derive a prediction sample for the current block; The steps include: deriving the current block's residual sample based on the residual-related information; The step of generating a restored sample of the current block based on the predicted sample and the residual sample, The residual-related information includes (i) a transformation skip flag based on the size of the current block and the maximum transformation skip size, or (ii) an integrated transformation type index included in the transformation unit syntax that collectively indicates either the application of a Multiple Transform Selection with a specified horizontal and vertical transformation kernel pair or the application of a transformation skip to the current block. When the integrated conversion type index indicates a conversion skip, the conversion skip flag is not analyzed individually. When the aforementioned conversion skip flag exists, the conversion skip flag indicates whether the conversion skip was applied to the current block. The integrated transformation type index is decoded using CABAC (context-adaptive binary arithmetic coding) by deriving a context model based on at least one of (a) a comparison of the width and height of the current block, (b) whether the current block is intra-predicted or inter-predicted, and (c) whether a transformation skip is applied to an adjacent block, without signaling additional information to identify the context model for at least one bin of the binstring indicating the integrated transformation type index. The information regarding the maximum conversion skip size is obtained from the bitstream. The information regarding the maximum conversion skip size is included in one of the picture parameter set or sequence parameter set. One of the picture parameter set or the sequence parameter set includes a conversion skip enable flag related to whether or not the conversion skip is enabled, A method wherein the information for the maximum conversion skip size is included in one of the picture parameter set or the sequence parameter set, based on the value of the conversion skip enable flag being equal to 1.

2. The information for the maximum conversion skip size includes the log2_transform_skip_max_size_minus2 syntax element, The aforementioned maximum conversion skip size is given by the formula MaxTsSize=1<<(log2_transform_skip_max_size_minus2+2) Derived based on, The method according to claim 1, wherein MaxTsSize indicates the maximum conversion skip size, and log2_transform_skip_max_size_minus2 indicates the value of the log2_transform_skip_max_size_minus2 syntax element.

3. The method according to claim 2, wherein the value of the log2_transform_skip_max_size_minus2 syntax element is represented by one of 0 to 3 candidate values.

4. The method according to claim 1, wherein the maximum conversion skip size is derived as one of the candidate sizes including 4, 8, 16, or 32.

5. A non-temporary computer-readable medium for storing instructions, wherein, at execution time, the instructions are transmitted to a decoding device. A step of obtaining prediction mode information and residual-related information from a bitstream, The steps include: performing a prediction based on the aforementioned prediction mode information to derive a prediction sample for the current block; The steps include: deriving the current block's residual sample based on the residual-related information; The operation includes the step of generating a restored sample of the current block based on the predicted sample and the residual sample, The residual-related information includes (i) a transformation skip flag based on the size of the current block and the maximum transformation skip size, or (ii) an integrated transformation type index included in the transformation unit syntax that collectively indicates either the application of a Multiple Transform Selection with a specified horizontal and vertical transformation kernel pair or the application of a transformation skip to the current block. When the integrated conversion type index indicates a conversion skip, the conversion skip flag is not analyzed individually. When the aforementioned conversion skip flag exists, the conversion skip flag indicates whether the conversion skip was applied to the current block. The integrated transformation type index is decoded using CABAC (context-adaptive binary arithmetic coding) by deriving a context model based on at least one of (a) a comparison of the width and height of the current block, (b) whether the current block is intra-predicted or inter-predicted, and (c) whether a transformation skip is applied to an adjacent block, without signaling additional information to identify the context model for at least one bin of the binstring indicating the integrated transformation type index. The information regarding the maximum conversion skip size is included in one of the picture parameter set or sequence parameter set. One of the picture parameter set or the sequence parameter set includes a conversion skip enable flag related to whether or not the conversion skip is enabled, The information for the maximum conversion skip size is included in one of the picture parameter set or the sequence parameter set, based on the value of the conversion skip enable flag being equal to 1. Non-temporary computer-readable media.

6. The information for the maximum conversion skip size includes the log2_transform_skip_max_size_minus2 syntax element, The aforementioned maximum conversion skip size is given by the formula MaxTsSize=1<<(log2_transform_skip_max_size_minus2+2) Derived based on, The non-temporary computer-readable medium according to claim 5, wherein MaxTsSize indicates the maximum conversion skip size, and log2_transform_skip_max_size_minus2 indicates the value of the log2_transform_skip_max_size_minus2 syntax element.

7. The non-temporary computer-readable medium according to claim 6, wherein the value of the log2_transform_skip_max_size_minus2 syntax element is represented by one of 0 to 3 candidate values.

8. The non-temporary computer-readable medium according to claim 5, wherein the maximum conversion skip size is derived as one of the candidate sizes including 4, 8, 16, or 32.

9. A decoding device having memory for storing instructions, wherein, at execution time, the instructions are sent to one or more processors, A step of obtaining prediction mode information and residual-related information from a bitstream, The steps include: performing a prediction based on the aforementioned prediction mode information to derive a prediction sample for the current block; The steps include: deriving the current block's residual sample based on the residual-related information; The operation includes the step of generating a restored sample of the current block based on the predicted sample and the residual sample, The residual-related information includes (i) a transformation skip flag based on the size of the current block and the maximum transformation skip size, or (ii) an integrated transformation type index included in the transformation unit syntax that collectively indicates either the application of a Multiple Transform Selection with a specified horizontal and vertical transformation kernel pair or the application of a transformation skip to the current block. When the integrated conversion type index indicates a conversion skip, the conversion skip flag is not analyzed individually. When the aforementioned conversion skip flag exists, the conversion skip flag indicates whether the conversion skip was applied to the current block. The integrated transformation type index is decoded using CABAC (context-adaptive binary arithmetic coding) by deriving a context model based on at least one of (a) a comparison of the width and height of the current block, (b) whether the current block is intra-predicted or inter-predicted, and (c) whether a transformation skip is applied to an adjacent block, without signaling additional information to identify the context model for at least one bin of the binstring indicating the integrated transformation type index. The information regarding the maximum conversion skip size is included in one of the picture parameter set or sequence parameter set. One of the picture parameter set or the sequence parameter set includes a conversion skip enable flag related to whether or not the conversion skip is enabled, The information for the maximum conversion skip size is included in one of the picture parameter set or the sequence parameter set, based on the value of the conversion skip enable flag being equal to 1. Decoding device.

10. The information for the maximum conversion skip size includes the log2_transform_skip_max_size_minus2 syntax element, The aforementioned maximum conversion skip size is given by the formula MaxTsSize=1<<(log2_transform_skip_max_size_minus2+2) Derived based on, The decoding apparatus according to claim 9, wherein MaxTsSize indicates the maximum conversion skip size, and log2_transform_skip_max_size_minus2 indicates the value of the log2_transform_skip_max_size_minus2 syntax element.

11. The decoding apparatus according to claim 10, wherein the value of the log2_transform_skip_max_size_minus2 syntax element is represented by one of 0 to 3 candidate values.

12. The decoding apparatus according to claim 9, wherein the maximum conversion skip size is derived as one of the candidate sizes including 4, 8, 16, or 32.

13. The method according to claim 1, wherein the prediction mode information relates to whether interpretation or intrapretation is applied to the current block.

14. The non-temporary computer-readable medium according to claim 6, wherein the prediction mode information relates to whether interpretation or intrapretation is applied to the current block.

15. The decoding apparatus according to claim 9, wherein the prediction mode information relates to whether interpretation or intraprediction is applied to the current block.

16. The method according to claim 1, wherein when conversion skip is applied to the current block, residual decoding omits signaling of the last effective coefficient position and processes the coefficient subblocks from the rightmost subblock toward the DC subblock, the processing of which is accompanied by a coded_sub_block_flag signaled for all subblocks, and effective context modeling using only the right and down neighboring positions with an effective inference in the DC position at the DC subblock and a simplified context model for one or more abs_level_gt1_flags and par_level_flags.

Images