Transform coefficient coding
An idea used herein is to use the same function for the dependency of the context and the dependency of the symbolization parameter on previously coded/decoded transform coefficients. Using the same function—with varying function parameter—may even be used with respect to different transform block sizes and/or frequency portions of the transform blocks in case of the transform coefficients being spatially arranged in transform blocks. A further variant of this idea is to use the same function for the dependency of a symbolization parameter on previously coded/decoded transform coefficients for different sizes of the current transform coefficient's transform block, different information component types of the current transform coefficient's transform block and/or different frequency portions the current transform coefficient is located within the transform block.
1. A decoder for decoding a data stream comprising an encoded video including information related to a plurality of transform coefficients associated with the encoded video, the decoder comprising:
a context adaptive entropy decoder configured for, for a current transform coefficient in a transform coefficient block, entropy decoding, using a processor, a first symbol based on a context, wherein the context is determined based on an operation involving a first previously decoded transform coefficient;
an extractor configured for, using the processor, extracting a second symbol of the current transform coefficient from the data stream;
a desymbolizer configured for, for the current transform coefficient, mapping the second symbol to a transform coefficient level based on a symbolization parameter, wherein the symbolization parameter is determined based on the operation involving a second previously decoded transform coefficient and an information component type including a color component type related to the transform coefficient block; and
a picture decoder configured for applying, using a processor, predictive coding based on a prediction residual signal to reconstruct a block of a picture of the video, wherein the prediction residual signal is associated with the first symbol and the transform coefficient level,
wherein the decoder is configured for determining the first previously decoded transform coefficient based on a relative spatial arrangement of the first previously decoded transform coefficient relative to the current transform coefficient, wherein the spatial arrangement is based on a scan position of the current transform coefficient in the transform coefficient block.
2. The decoder according to claim 1 , wherein the decoder is further configured for extracting, from the data stream, information on a position of a last non-zero transform coefficient among transform coefficients of a transform coefficient block along a predetermined scan order, wherein the plurality of transform coefficients encompasses the transform coefficients of the transform coefficient block from the last non-zero transform coefficient along the scan order to a DC transform coefficient of the transform coefficient block.
3. The decoder according to claim 2 , wherein the desymbolizer is further configured for using a first symbolization scheme for mapping of the first symbol of the last non-zero transform coefficient to a level less than or equal to a maximum level, wherein a zero level does not apply for the last transform coefficient.
4. The decoder according to claim 2 , wherein the context adaptive entropy decoder is further configured for entropy decoding the first symbol for the last non-zero transform coefficient using a context that is different from a context used in entropy decoding the first symbol of a transform coefficient other than the last non-zero transform coefficient.
5. The decoder according to claim 2 , wherein the context adaptive entropy decoder traverses the transform coefficients of the transform coefficient block in a scan order leading from the last non-zero transform coefficient to the DC transform coefficient of the transform coefficient block.
6. The decoder according to claim 1 , wherein the decoder is configured for decoding the plurality of transform coefficients from the data stream in two scans, wherein the context adaptive entropy decoder is configured for entropy decoding the first symbol for the transform coefficients from the data stream in an order corresponding to a first scan of the transform coefficients, and the extractor is configured for subsequently extracting, from the data stream, the second symbols, for the transform coefficients for which the first symbols are mapped onto a maximum level, in an order corresponding to a second scan of the transform coefficients.
7. The decoder according to claim 1 , wherein the decoder is configured for decoding the plurality of transform coefficients from the data stream sequentially in one scan, wherein the second symbols are interspersed within the data stream between first symbols of the transform coefficients, and wherein the context adaptive entropy decoder and the extractor are configured for, for each transform coefficient in a scan order of the one scan, extracting, from the data stream, the second symbols of respective transform coefficients for which the first symbols are mapped onto a maximum level immediately subsequent to the context adaptive entropy decoder's entropy decoding of the first symbols of the respective transform coefficients for which the first symbols are mapped onto the maximum level.
8. The decoder according to claim 1 , wherein the extractor is configured for extracting the second symbol from the data stream directly or using entropy decoding using a fixed probability distribution.
9. The decoder according to claim 1 , wherein the desymbolizer is configured for mapping the first symbol to a transform coefficient level less than or equal to a maximum level associated with the first symbol based on a truncated unary binarization scheme.
10. The decoder according to claim 1 , wherein the desymbolizer is further configured for mapping the second symbol based on a Rice code.
11. An encoder for encoding into a data stream information related to a plurality of transform coefficients associated with a video, the encoder comprising:
a picture encoder configured for applying, using a processor, predictive coding on a block of a picture of the video to obtain a prediction residual signal related to the block;
a context adaptive entropy encoder configured for, for a current transform coefficient in a transform coefficient block related to the prediction residual signal, entropy encoding, using a processor, a first symbol based on a context, wherein the context is determined based on an operation involving a first previously encoded transform coefficient; and
a symbolizer configured for mapping, using a processor, a transform coefficient level of the current transform coefficient to a second symbol based on a symbolization parameter, wherein the symbolization parameter is determined based on the operation involving a second encoded transform coefficient and an information component type including a color component type related to the transform coefficient block,
wherein the encoder is configured for determining the first previously encoded transform coefficient based on a relative spatial arrangement of the first previously encoded transform coefficient relative to the current transform coefficient, wherein the spatial arrangement is based on a scan position of the current transform coefficient in the transform coefficient block.
12. The encoder according to claim 11 , wherein the encoder is further configured for inserting, into the data stream, information on a position of a last non-zero transform coefficient among transform coefficients of a transform coefficient block along a predetermined scan order, wherein the plurality of transform coefficients encompasses the transform coefficients of the transform coefficient block from the last non-zero transform coefficient along the scan order to a DC transform coefficient of the transform coefficient block.
13. The encoder according to claim 12 , wherein the context adaptive entropy encoder is configured for entropy encoding the first symbol for the last non-zero transform coefficient using a context that is different from a context used in entropy encoding the first symbol of a transform coefficient other than the last non-zero transform coefficient.
14. The encoder according to claim 12 , wherein the context adaptive entropy encoder traverses the transform coefficients of the transform coefficient block in a scan order leading from the last non-zero transform coefficient to the DC transform coefficient of the transform coefficient block.
15. The encoder according to claim 11 , wherein the encoder is configured to encoding the plurality of transform coefficients into the data stream in two scans, wherein the context adaptive entropy encoder is configured for entropy encoding the first symbol for the transform coefficients into the data stream in an order corresponding to a first scan of the transform coefficients,
the encoder comprises an inserter configured for subsequently inserting into the data stream the second symbols, for the transform coefficients for which the first symbols are mapped onto a maximum level, in an order corresponding to a second scan of the transform coefficients.
16. The encoder according to claim 11 , wherein the encoder is configured for encoding the plurality of transform coefficients into the data stream sequentially in one scan, wherein second symbols are interspersed within the data stream between first symbols of the transform coefficients, and
wherein the context adaptive entropy encoder and an inserter of the encoder are configured for, for each transform coefficient in a scan order of the one scan, inserting, into the data stream, the second symbols of respective transform coefficients for which the first symbols are mapped onto a maximum level, immediately subsequent to the context adaptive entropy encoder's entropy encoding of the first symbols of the respective transform coefficients for which the first symbols are mapped onto the maximum level.
17. The encoder according to claim 11 , further comprising an inserter configured for inserting the second symbol into the data stream directly or using entropy encoding using a fixed probability distribution.
18. The encoder according to claim 11 , wherein the symbolizer is configured for mapping the first symbol to a transform coefficient level less than or equal to the maximum level based on a truncated unary binarization scheme.
19. The encoder according to claim 11 , wherein the symbolizer is configured for mapping the second symbol based on a Rice code.
20. A method for decoding a data stream including information related to a plurality of transform coefficients associated with a video encoded into the data stream, comprising:
entropy decoding, for a current transform coefficient in a transform coefficient block, a first symbol based on a context, wherein the context is determined based on an operation involving a first previously decoded transform coefficient;
extracting a second symbol of the current transform coefficient from the data stream;
mapping, for the current transform coefficient, the second symbol to a transform coefficient level based on a symbolization parameter, wherein the symbolization parameter is determined based on the operation involving a second previously decoded transform coefficient and an information component type including a color component type related to the transform coefficient block; and
applying predictive coding based on a prediction residual signal to reconstruct a block of a picture of the video, wherein the prediction residual signal is associated with the first symbol and the transform coefficient level,
wherein the first previously decoded transform coefficient is determined based on a relative spatial arrangement of the first previously decoded transform coefficient relative to the current transform coefficient, wherein the spatial arrangement is based on a scan position of the current transform coefficient in the transform coefficient block.
21. The method according to claim 20 , further comprising:
extracting, from the data stream, information on a position of a last non-zero transform coefficient among transform coefficients of a transform coefficient block along a predetermined scan order,
wherein the plurality of transform coefficients encompasses the transform coefficients of the transform coefficient block from the last non-zero transform coefficient along the scan order to a DC transform coefficient of the transform coefficient block.
22. The method according to claim 20 , further comprising:
mapping the first symbol to a transform coefficient level less than or equal to a maximum level based on a truncated unary binarization scheme.
23. The method according to claim 22 , wherein the mapping of the second symbol is based on a Rice code.