Description:SAME AS UPLOADED , Claims:1. A decoder for block-based decoding of a picture from a data stream, the picture comprising a plurality of blocks to be decoded, the decoder comprising:
a processor; and
a memory storing instructions that, when executed by the processor, cause the processor to:
determine an intra-coding mode for a current block of the plurality of blocks;

wherein the current block is partitioned, along a partition dimension, into a number of partitions, wherein the number of partitions is greater than 2;

sequentially reconstruct each partition of the current block in a predetermined partition order along the partition dimension by, for a current partition by,

deriving a predictor for the current partition of the current block using the intra-coding mode of the current block based on one or more already reconstructed samples neighboring the current partition;

combining the predictor and a prediction residual for the current partition based on prediction residual signaled in the data steam,
wherein a coded partition flag for the current block is determined by:
if the current partition is the last partition in the current block, based on the predetermined partition order, and if each coded partition flag for each previously reconstructed partition in the current block is zero, inferring that the coded partition flag for the current partition is one,

otherwise, decoding the coded partition flag from the data stream; and

if the coded partition flag is zero, inferring the prediction residual for the current partition is zero, and

if the coded partition flag is one or inferred to be one, decoding a transform coefficient of the prediction residual of the current partition from the data stream,
wherein the decoding the transform coefficient of the prediction residual for the current partition from the data stream is performed in transform domain using a transform per partition, wherein the transform is of a type which is selected based on the current partition’s size.

2. The decoder as claimed in claim 1, wherein the partitions are one sample wide along to the partition dimension and the transform is a one-dimensional transform.

3. The decoder as claimed in claim 1, configured to:
decode a split mode flag for the current bock of the picture from the data stream;
if the split mode flag indicates a first split mode, perform the partitioning and the sequentially reconstructing;
if the split mode flag indicates a second split mode, reconstruct the current block by spatial intra prediction using the intra-coding mode of the current block.

4. The decoder as claimed in claim 3, configured to:
perform the spatial intra prediction in case of the split mode flag indicating the second split mode, by deriving a predictor for the current block by deriving the predictor for the current block depending on one or more already reconstructed samples neighboring the current block using the intra­ coding mode of the current block and then reconstructing the current block by combining the predictor and the prediction residual for the current block using the prediction residual signaled in the data steam, or
perform the spatial intra prediction in case of the split mode flag indicating the second split mode, by partitioning the current block into leaf blocks of an hierarchical multi-tree subdivision of the current block along horizontal and vertical subdivision leaf block borders and sequentially reconstructing the leaf blocks of the current block according to a predetermined leaf block order which sequentially traverses the leaf blocks.

5. The decoder of claim 3, configured to,
if the intra prediction mode is an angular mode, if a split mode flag indicates a first split mode, wherein deriving the predictor for the current partition depending on one or more already reconstructed samples neighboring the current partition in a manner depending on the intra-coding mode, subject the already reconstructed samples to a first interpolation filtering which depends on the angular mode;
if the split mode flag the coded partition flag indicates a second split mode, in reconstruct the predetermined block by spatial intra prediction, fill the predetermined block depending on the one or more already reconstructed samples neighboring the predetermined block in a manner depending on the intra-coding mode by subjecting the already reconstructed samples to a second interpolation filtering which depends on the angular mode, wherein the first interpolation filtering differs from the second interpolation filtering in a mean number of already reconstructed samples on which each sample of the predictor within the current block depends.

6. The decoder as claimed in claim 3, comprising one or more tools and configured to disable the tools for blocks of the picture for which the first split mode is indicated.

7. The decoder as claimed in claim 6, wherein the one or more tools relate to transformation and inverse transformation.

8. The decoder as claimed in claim 6, wherein the one or more tools comprise one or more of:
transform skipping,
explicit multiple core transform, and
mode dependent non-separable secondary transform.

9. The decoder as claimed in claim 3, configured to, in deriving the predictor for the current partition depending on one or more already reconstructed samples neighboring the current partition using the intra-coding mode of the current block, use a filter to derive the predictor, if the coded partition flag indicates the second split mode, and disable the filter, if the split mode flag is indicates the first split mode, or perform a filtering in deriving the predictor for the current partition in a manner so that a number of the already reconstructed samples which contribute to one sample in the current partition is reduced for a case that the split mode flag indicates the first split mode, compared to a case that the split mode flag indicates the second split mode.

10. An encoder for block-based encoding of a picture into a data stream, the picture comprising a plurality of blocks to be encoded, the encoder comprising:
a processor; and
a memory storing instructions that, when executed by the processor, cause the processor to:
determine an intra-coding mode for a current block of the picture;
partition, along a partition dimension, the current block into partitions, wherein the number of partitions is greater than 2;
sequentially process each partition of the current block in a predetermined partition order along the partition dimension by, for a current partition,
deriving a predictor for the current partition using the intra-coding mode of the current block based on one or more already encoded samples neighboring the current partition;
determining a prediction residual for use in correcting the predictor for reconstructing the current partition,
wherein the encoder is further configured to:
encode the prediction residual into the data stream in transform domain by encoding coded partition flags for the partitions into the data stream sequentially except the coded partition flag for a last partition in partition order, if all preceding coded partition flags are not set, which is then to be inferred to be set,
wherein, for each partition, wherein the coded partition flag being not set, indicates that the prediction residual for the respective partition is to be set to zero, and
the encoder is configured to, if the coded partition flag is set or inferred to be set, perform a transform of the prediction residual of the respective partition and encode the transformed prediction residual into the data stream,
wherein encoding the transformed prediction residual into the data stream is performed in transform domain using a transform per partition, wherein the transform is a of a type which is selected based on the respective partition’s size.

11. The encoder as claimed in claim 10, wherein the partitions are one sample wide along to the partition dimension and the transform is a one-dimensional transform.

12. The encoder as claimed in claim 10, configured to:
encode a split mode flag for the current block of the picture into the data stream;
if the split mode flag indicates a first split mode, perform the partitioning and the sequentially processing of the partitions;
if the split mode flag indicates a second split mode, subject the current block to spatial intra prediction using the intra-coding mode of the current block and encode a prediction residual of the current block.

13. The encoder as claimed in claim 12, configured to:
perform the spatial intra prediction in case of the split mode flag indicating the second split mode, by deriving a predictor for the current block by deriving the predictor for current block depending on one or more already encoded samples neighboring the current block using the intra-coding mode of the current block and determine the prediction residual of the current block for a reconstruction of the current block by combining the predictor and the prediction residual of the current block, or
perform the spatial intra prediction in case of the split mode flag indicating the second split mode, by partitioning the current block into leaf blocks of an hierarchical multi-tree subdivision of the current block along horizontal and vertical subdivision leaf block borders and determine the prediction residual of the current block for a sequential reconstruction of the leaf blocks of the current block according to a predetermined leaf block order which sequentially traverses the leaf blocks.

14. The encoder as claimed in claim 12, configured to, if the intra prediction mode is an angular mode,
if the split mode flag indicates a first split mode,

in deriving the predictor for the current partition depending on one or more already reconstructed samples neighboring the current partition using the intra-coding mode, subject the already encoded samples to a first interpolation filtering which depends on the angular mode;

if the split mode flag indicates a second split mode,

in reconstructing the current block by spatial intra prediction, derive the predictor depending on the one or more already reconstructed samples neighboring the current block using the intra-coding mode of the current block by subjecting the already encoded samples to a second interpolation filtering which depends on the angular mode,

wherein the first interpolation filtering differs from the second interpolation filtering in a mean number of already encoded samples on which each sample of the predictor within the current block depends.

15. The encoder as claimed in claim 12, comprising one or more tools and configured to disable the tools for blocks of the picture for which the first split mode is indicated.

16. The encoder as claimed in claim 15, wherein the one or more tools relate to transformation and inverse transformation.

17. The encoder as claimed in claim 15, wherein the one or more tools comprise one or more of:
transform skipping,
explicit multiple core transform, and
mode dependent non-separable secondary transform.

18. The encoder as claimed in claim 15,
configured to, in deriving the predictor for the current partition depending on one or more already reconstructed samples neighboring the current partition using the intra-coding mode of the current block, use a filter to derive the predictor for the current partition, if the split mode flag indicates the second split mode, and disable the filter, if the split mode flag indicates the first split mode, or perform a filtering in deriving the predictor for the current partition in a manner so that a number of the already reconstructed samples which contribute to one sample in the current partition is reduced for a case that the split mode flag indicates the first split mode, compared to a case that the split mode flag indicates the second split mode.

19. The encoder as claimed in claim 10, configured to
encode a split mode flag for the current block of the picture into the data stream, the split mode flag indicating, if being set, that the current block is coded using the partitioning and the sequentially processing of the prediction residual of the partitions, or, if not set, that the current block is coded using an alternative manner of spatial intra prediction using the intra prediction mode determined for the current block.

20. A method for block-based decoding of a picture from a data stream, the picture comprises a plurality of blocks to be decoded, the method comprising:
determining an intra-coding mode for a current block of the plurality of blocks;
wherein the current block is partitioned, along a partition dimension, into a number of partitions, wherein the number of partitions is greater than 2;
sequentially reconstructing each partition of the current block in a predetermined partition order along the partition dimension by, for a current partition by,
deriving a predictor for the current partition of the current block using the intra coding mode of the current block based on one or more already reconstructed samples neighboring the current partition;
combining the predictor and a prediction residual for the current partition based on prediction residual signaled in the data steam,
wherein a coded partition flag for the current block is determined by:
if the current partition is the last partition in the current block, based on the predetermined partition order, and if each coded partition flag for each previously reconstructed partition in the current block is zero, inferring that the coded partition flag for the current partition is one,
otherwise, decoding the coded partition flag from the data stream; and
if the coded partition flag is zero, inferring the prediction residual for the current partition is zero, and
if the coded partition flag is one or inferred to be one, decoding a transform coefficient of the prediction residual of the current partition from the data stream,
wherein the decoding the transform coefficient of the prediction residual for the current partition from the data stream is performed in transform domain using a transform per partition, wherein the transform is of a type which is selected based on the current partition’s size.

21. A method for block-based encoding of a picture into a data stream, to the picture comprising of blocks to be encoded, the method comprising:
determining an intra-coding mode for a current block of the picture;
partitioning, along a partition dimension, the current block into partitions, wherein the number of partitions is greater than 2;
sequentially processing each partition of the current block in a predetermined partition order along the partition dimension by, for a current partition,
deriving a predictor for the current partition using the intra-coding mode of the current block based on one or more already encoded samples neighboring the current partition;
determining a prediction residual for use in correcting the predictor for reconstructing the current partition,
wherein the method further comprises:
encoding the prediction residual into the data stream in transform domain by encoding coded partition flags for the partitions into the data stream sequentially except the coded partition flag for a last partition in partition order, if all preceding coded partition flags are not set, which is then to be inferred to be set, wherein, for each partition,
wherein the coded partition flag being not set, indicates that the prediction residual for the respective partition is to be set to zero, and
if the coded partition flag is set or inferred to be set, perform a transform of the prediction residual of the respective partition and encode the transformed prediction residual is encoded into the data stream,
wherein encoding the transformed prediction residual into the data stream is performed in transform domain using a transform per partition, wherein the transform is a of a type which is selected based on the respective partition’s size.