Method and arrangement for coding transform coefficients in picture and/or video coders and decoders and a corresponding computer program and a corresponding computer-readable storage medium

A video coder having at least one processor and/or chip formed such that a method for coding transform coefficients can be performed, wherein

• for blocks of (video) pictures containing transform coefficients being unequal to zero, a coding of transform coefficients takes place in such a way that, for each block,
• a significance map is coded, the significance map specifying the positions of transform coefficients being unequal to zero in the block in a scan order in a context-dependent way using contexts depending on the corresponding scan position of the transform coefficient considered, and subsequently,
• the values (levels) of the transform coefficients being unequal to zero are coded by, in a reverse scan order—starting with the last transform coefficient being unequal to zero within the block—binary arithmetically coding bins of a binarization of a magnitude of the transform coefficients being unequal to zero in a context-dependent way using contexts depending on how many transform coefficients having a magnitude of 1 have already been coded in the reverse scan order up to a respective currently coded transform coefficient, and how many transform coefficients having a magnitude of greater than 1 have already been coded in the reverse scan order up to the respective currently coded transform coefficient, respectively, with coding all bins of the binarization of the respective currently to be coded transform coefficient being unequal to zero before proceeding with coding the bins of the binarization of a —in the reverse scan order—succeeding transform coefficient being unequal to zero.

A computer program, stored on a computer readable storage medium and executable by a computer to perform a method for coding transform coefficients, wherein

• for blocks of (video) pictures containing transform coefficients being unequal to zero, a coding of transform coefficients takes place in such a way that, for each block,
• a significance map is coded, the significance map specifying the positions of transform coefficients being unequal to zero in the block in a scan order in a context-dependent way using contexts depending on the corresponding scan position of the transform coefficient considered, and subsequently,
• the values (levels) of the transform coefficients being unequal to zero are coded by, in a reverse scan order—starting with the last transform coefficient being unequal to zero within the block—binary arithmetically coding bins of a binarization of a magnitude of the transform coefficients being unequal to zero in a context-dependent way using contexts depending on how many transform coefficients having a magnitude of 1 have already been coded in the reverse scan order up to a respective currently coded transform coefficient, and how many transform coefficients having a magnitude of greater than 1 have already been coded in the reverse scan order up to the respective currently coded transform coefficient, respectively, with coding all bins of the binarization of the respective currently to be coded transform coefficient being unequal to zero before proceeding with coding the bins of the binarization of a —in the reverse scan order—succeeding transform coefficient being unequal to zero.

A computer-readable storage medium having a computer program stored thereon, said computer program executable by a computer to perform a method for coding transform coefficients, wherein

• for blocks of (video) pictures containing transform coefficients being unequal to zero, a coding of transform coefficients takes place in such a way that, for each block,
• a significance map is coded, the significance map specifying the positions of transform coefficients being unequal to zero in the block in a scan order in a context-dependent way using contexts depending on the corresponding scan position of the transform coefficient considered, and subsequently,
• the values (levels) of the transform coefficients being unequal to zero are coded by, in a reverse scan order—starting with the last transform coefficient being unequal to zero within the block—binary arithmetically coding bins of a binarization of a magnitude of the transform coefficients being unequal to zero in a context-dependent way using contexts depending on how many transform coefficients having a magnitude of 1 have already been coded in the reverse scan order up to a respective currently coded transform coefficient, and how many transform coefficients having a magnitude of greater than 1 have already been coded in the reverse scan order up to the respective currently coded transform coefficient, respectively, with coding all bins of the binarization of the respective currently to be coded transform coefficient being unequal to zero before proceeding with coding the bins of the binarization of a —in the reverse scan order —succeeding transform coefficient being unequal to zero.

A computer program stored in a computer memory and executable by a computer to perform a method for coding transform coefficients, wherein

• for blocks of (video) pictures containing transform coefficients being unequal to zero, a coding of transform coefficients takes place in such a way that, for each block,
• a significance map is coded, the significance map specifying the positions of transform coefficients being unequal to zero in the block in a scan order in a context-dependent way using contexts depending on the corresponding scan position of the transform coefficient considered, and subsequently,
• the values (levels) of the transform coefficients being unequal to zero are coded by, in a reverse scan order—starting with the last transform coefficient being unequal to zero within the block—binary arithmetically coding bins of a binarization of a magnitude of the transform coefficients being unequal to zero in a context-dependent way using contexts depending on how many transform coefficients having a magnitude of 1 have already been coded in the reverse scan order up to a respective currently coded transform coefficient, and how many transform coefficients having a magnitude of greater than 1 have already been coded in the reverse scan order up to the respective currently coded transform coefficient, respectively, with coding all bins of the binarization of the respective currently to be coded transform coefficient being unequal to zero before proceeding with coding the bins of the binarization of a —in the reverse scan order—succeeding transform coefficient being unequal to zero.

A method for decoding a coding of a significance map and a subsequent coding of values of transform coefficients being unequal to zero for blocks of (video) pictures containing transform coefficients being unequal to zero, the significance map specifying the positions of the transform coefficients being unequal to zero in a scan order, and the coding of values of transform coefficients being unequal to zero comprising coded values of the transform coefficients being unequal to zero in a reverse scan order—starting with the last transform coefficient being unequal to zero, comprising the steps of:

• decoding the significance mapping in a context-dependent way in the scan order using contexts depending on the corresponding scan position of the transform coefficient considered;
• and
• decoding the coded values of transform coefficients being unequal to zero, using a computer programmed to perform the decoding, by, in the reverse scan order, binary arithmetically decoding bins of a binarization of a magnitude of the transform coefficients being unequal to zero in a context-dependent way using contexts depending on how many transform coefficients having a magnitude of 1 have already been decoded in the reverse scan order up to a respective currently decoded transform coefficient, and how many transform coefficients having a magnitude of greater than 1 have already been decoded in the reverse scan order up to the respective currently decoded transform coefficient, respectively, with decoding all bins of the binarization of the respective currently to be decoded transform coefficient being unequal to zero before proceeding with decoding the bins of the binarization of a—in the reverse scan order—succeeding transform coefficient being unequal to zero.

A device for decoding a coding of a significance map and a subsequent coding of values of transform coefficients being unequal to zero for blocks of (video) pictures containing transform coefficients being unequal to zero, the significance map specifying the positions of the transform coefficients being unequal to zero in a scan order, and the coding of values of transform coefficients being unequal to zero comprising coded values of the transform coefficients being unequal to zero in a reverse scan order—starting with the last transform coefficient being unequal to zero, the device comprising a processor for

• decoding the significance mapping in the scan order in a context-dependent way using contexts depending on the corresponding scan position of the transform coefficient considered;
• and
• decoding the coded values of transform coefficients being unequal to zero by, in the reverse scan order, binary arithmetically decoding bins of a binarization of a magnitude of the transform coefficients being unequal to zero in a context-dependent way using contexts depending on how many transform coefficients having a magnitude of 1 have already been decoded in the reverse scan order up to a respective currently decoded transform coefficient, and how many transform coefficients having a magnitude of greater than 1 have already been decoded in the reverse scan order up to the respective currently decoded transform coefficient, respectively, with decoding all bins of the binarization of the respective currently to be decoded transform coefficient being unequal to zero before proceeding with decoding the bins of the binarization of a—in the reverse scan order—succeeding transform coefficient being unequal to zero.

A device for decoding a coding of a significance map and a subsequent coding of values of transform coefficients being unequal to zero for blocks of (video) pictures containing transform coefficients being unequal to zero, the significance map specifying the positions of the transform coefficients being unequal to zero in a scan order, and the coding of values of transform coefficients being unequal to zero comprising coded values of the transform coefficients being unequal to zero in a reverse scan order-starting with the last transform coefficient being unequal to zero, the device comprising:

• means for decoding the significance mapping in the scan order in a context-dependent way using contexts depending on the corresponding scan position of the transform coefficient considered;
• and
• means for decoding the coded values of transform coefficients being unequal to zero by, in the reverse scan order, binary arithmetically decoding bins of a binarization of a magnitude of the transform coefficients being unequal to zero in a context-dependent way using contexts depending on how many transform coefficients having a magnitude of 1 have already been decoded in the reverse scan order up to a respective currently decoded transform coefficient, and how many transform coefficients having a magnitude of greater than 1 have already been decoded in the reverse scan order up to the respective currently decoded transform coefficient, respectively, with decoding all bins of the binarization of the respective currently to be decoded transform coefficient being unequal to zero before proceeding with decoding the bins of the binarization of a—in the reverse scan order—succeeding transform coefficient being unequal to zero.

Video decoder for decoding a coding of a significance map and a subsequent coding of values of transform coefficients being unequal to zero for blocks of (video) pictures containing transform coefficients being unequal to zero, the significance map specifying the positions of the transform coefficients being unequal to zero in a scan order, and the coding of values of transform coefficients being unequal to zero comprising coded values of the transform coefficients being unequal to zero in a reverse scan order-starting with the last transform coefficient being unequal to zero, the video decoder having at least one processor and/or chip being configured to

• decode the significance mapping in the scan order in a context-dependent way using contexts depending on the corresponding scan position of the transform coefficient considered;
• and
• decode the coded values of transform coefficients being unequal to zero by, in the reverse scan order, binary arithmetically decoding bins of a binarization of a magnitude of the transform coefficients being unequal to zero in a context-dependent way using contexts depending on how many transform coefficients having a magnitude of 1 have already been decoded in the reverse scan order up to a respective currently decoded transform coefficient, and how many transform coefficients having a magnitude of greater than 1 have already been decoded in the reverse scan order up to the respective currently decoded transform coefficient, respectively, with decoding all bins of the binarization of the respective currently to be decoded transform coefficient being unequal to zero before proceeding with decoding the bins of the binarization of a—in the reverse scan order—succeeding transform coefficient being unequal to zero.