Back to results
US 10,771,822 B2Granted
Coding of a spatial sampling of a two-dimensional information signal using sub-division
AssigneeGe Video Compression, LLC
InventorsHeiner Kirchhoffer, Martin Winken, Philipp Helle, Detlev Marpe, Heiko Schwarz, Thomas Wiegand
Life of this patent
Estimating the term…
Abstract
Coding schemes for coding a spatially sampled information signal using sub-division and coding schemes for coding a sub-division or a multitree structure are described, wherein representative embodiments relate to picture and/or video coding applications.
Claims65
A decoder for decoding a data stream corresponding to a video comprising a plurality of arrays of information samples, comprising:
- an extractor configured for extracting, from the data stream, information related to first subdivision information, second subdivision information, and a maximum hierarchy level associated with each of the plurality of arrays of information samples, wherein the first subdivision information is associated with regions of the array of information samples in a spatial domain and includes a first maximum region size and multi-tree subdivision information, the second subdivision information is associated with regions of the array of information samples in a spectral domain and includes a second maximum region size and subordinate multi-tree subdivision information, and the maximum hierarchy level is associated with regions of the array of information samples in the spectral domain;
- a sub-divider configured for dividing each of the arrays of information samples in accordance with the first and second maximum region sizes, the multi-tree subdivision information and the subordinate multi-tree subdivision information, and the maximum hierarchy level for the array of information samples, by
- partitioning the array of information samples into a first set of regions in the spatial domain based on the first maximum region size,
- sub-partitioning at least a subset of the first set of regions into a first set of sub-regions in the spatial domain based on the multi-tree subdivision information associated therewith,
- responsive to a determination that a size of at least one of the first set of sub-regions exceeds the second maximum region size, partitioning the at least one of the first set of sub-regions into a second set of regions in the spectral domain, and
- responsive to a determination that at least one of the second set of regions of the second maximum size is to be sub-divided, sub-partitioning the at least one of the second set of regions into a second set of sub-regions in the spectral domain based on the subordinate multi-tree subdivision information and the maximum hierarchy level associated therewith;
- and
- a reconstructor configured for reconstructing each of the arrays of information samples based on the first and second sets of sub-regions generated by the sub-divider.
5 more elements
A method of decoding a data stream corresponding to a video comprising a plurality of arrays of information samples, comprising:
- extracting, from the data stream, information related to first subdivision information, second subdivision information, and a maximum hierarchy level associated with each of the plurality of arrays of information samples, wherein the first subdivision information is associated with regions of the array of information samples in a spatial domain and includes a first maximum region size and multi-tree subdivision information, the second subdivision information is associated with regions of the array of information samples in a spectral domain and includes a second maximum region size and subordinate multi-tree subdivision information, and the maximum hierarchy level is associated with regions of the array of information samples in the spectral domain;
- dividing each of the arrays of information samples in accordance with the first and second maximum region sizes, the multi-tree subdivision information, the subordinate multi-tree subdivision information, and the maximum hierarchy level for the array of information samples, by
- partitioning the array of information samples into a first set of regions in the spatial domain based on the first maximum region size,
- sub-partitioning at least a subset of the first set of regions into a first set of sub-regions in the spatial domain based on the multi-tree subdivision information associated therewith,
- responsive to a determination that a size of at least one of the first set of sub-regions exceeds the second maximum region size, partitioning the at least one of the first set of sub-regions into a second set of regions in the spectral domain, and
- responsive to a determination that at least one of the second set of regions of the second maximum size is to be sub-divided, sub-partitioning the at least one of the second set of regions into a second set of sub-regions in the spectral domain based on the subordinate multi-tree subdivision information and the maximum hierarchy level associated therewith;
- and
- reconstructing each of the arrays of information samples based on the first and second sets of sub-regions generated by said dividing.
5 more elements
A non-transitory machine readable digital storage medium having computer instructions stored thereon for decoding, wherein the computer instructions, when read by the machine, causing the machine to perform at least:
- extracting, from a data stream, information related to first subdivision information, second subdivision information, and a maximum hierarchy level associated with each of a plurality of arrays of information samples, wherein the first subdivision information is associated with regions of the array of information samples in a spatial domain and includes a first maximum region size and multi-tree subdivision information, the second subdivision information is associated with regions of the array of information samples in a spectral domain and includes a second maximum region size and subordinate multi-tree subdivision information, and the maximum hierarchy level is associated with regions of the array of information samples in the spectral domain;
- dividing each of the arrays of information samples in accordance with the first and second maximum region sizes, the multi-tree subdivision information and the subordinate multi-tree subdivision information, and the maximum hierarchy level for the array of information samples, by
- partitioning the array of information samples into a first set of regions in the spatial domain based on the first maximum region size,
- sub-partitioning at least a subset of the first set of regions into a first set of sub-regions in the spatial domain based on the multi-tree subdivision information associated therewith,
- responsive to a determination that a size of at least one of the first set of sub-regions exceeds the second maximum region size, partitioning the at least one of the first set of sub-regions into a second set of regions in the spectral domain, and
- responsive to a determination that at least one of the second set of regions of the second maximum size is to be sub-divided, sub-partitioning the at least one of the second set of regions into a second set of sub-regions in the spectral domain based on the subordinate multi-tree subdivision information and the maximum hierarchy level associated therewith;
- and
- reconstructing each of the arrays of information samples based on the first and second sets of sub-regions generated by said dividing.
5 more elements
An encoder for encoding a video comprising a plurality of arrays of information samples into a data stream, comprising:
- a sub-divider configured for dividing each of the plurality of arrays of information samples by
- determining, with respect to the array of information samples, a first maximum region size, multi-tree subdivision information, a second maximum region size, and subordinate multi-tree subdivision information for the array of information samples, wherein the first maximum region size and the multi-tree subdivision information are associated with regions of the array of information samples in a spatial domain, and the second maximum region size and the subordinate multi-tree subdivision information are associated with regions of the array of information samples in a spectral domain,
- determining a maximum hierarchy level for the array of samples, wherein the maximum hierarchy level is associated with regions of the array of information samples in the spectral domain,
- partitioning the array of information samples into a first set of regions in the spatial domain based on the first maximum region size,
- sub-partitioning at least a subset of the first set of regions into a first set of sub-regions in the spatial domain based on the multi-tree subdivision information associated therewith,
- responsive to a determination that a size of at least one of the first set of sub-regions exceeds the second maximum region size, partitioning the at least one of the first set of sub-regions into a second set of regions in the spectral domain, and
- responsive to a determination that at least one of the second set of regions of the second maximum size is to be sub-divided, sub-partitioning the at least one of the second set of regions into a second set of sub-regions in the spectral domain based on the subordinate multi-tree subdivision information and the maximum hierarchy level associated therewith;
- and
- a data stream generator configured for
- encoding each of the arrays of information samples by
- encoding, into the data stream, information related to first subdivision information, second subdivision information, and the maximum hierarchy level associated with each of the arrays of information samples, wherein the first subdivision information includes the first maximum region size and the multi-tree subdivision information, and the second subdivision information includes the second maximum region size and the subordinate multi-tree subdivision information.
8 more elements
A method of encoding a video comprising a plurality of arrays of information samples into a data stream, comprising:
- dividing each of the plurality of arrays of information samples by
- determining, with respect to the array of information samples, a first maximum region size, multi-tree subdivision information, a second maximum region size, and subordinate multi-tree subdivision information for the array of information samples, wherein the first maximum region size and the multi-tree subdivision information are associated with regions of the array of information samples in a spatial domain, and the second maximum region size and the subordinate multi-tree subdivision information are associated with regions of the array of information samples in a spectral domain,
- determining a maximum hierarchy level for the array of information samples, wherein the maximum hierarchy level is associated with regions of the array of information samples in the spectral domain,
- partitioning the array of information samples into a first set of regions in the spatial domain based on the first maximum region size,
- sub-partitioning at least a subset of the first set of regions into a first set of sub-regions in the spatial domain based on the multi-tree subdivision information associated therewith,
- responsive to a determination that a size of at least one of the first set of sub-regions exceeds the second maximum region size, partitioning the at least one of the first set of sub-regions into a second set of regions in the spectral domain, and
- responsive to a determination that at least one of the second set of regions of the second maximum size is to be sub-divided, sub-partitioning the at least one of the second set of regions into a second set of sub-regions in the spectral domain based on the subordinate multi-tree subdivision information and the maximum hierarchy level associated therewith;
- and
- encoding each of the arrays of information samples by encoding, into the data stream, information related to first subdivision information, second subdivision information, and the maximum hierarchy level associated with each of the arrays of information samples, wherein the first subdivision information includes the first maximum region size and the multi-tree subdivision information, and the second subdivision information includes the second maximum region size and the subordinate multi-tree subdivision information.
6 more elements
A non-transitory machine readable digital storage medium having computer instructions stored thereon for encoding, wherein the computer instructions, when read by the machine, causing the machine to at least perform:
- dividing each of the plurality of arrays of information samples by
- determining, with respect to the array of information samples, a first maximum region size, multi-tree subdivision information, a second maximum region size, and subordinate multi-tree subdivision information for the array of information samples, wherein the first maximum region size and the multi-tree subdivision information are associated with regions of the array of information samples in a spatial domain, and the second maximum region size and the subordinate multi-tree subdivision information are associated with regions of the array of information samples in a spectral domain,
- determining a maximum hierarchy level for the array of information samples, wherein the maximum hierarchy level is associated with regions of the array of information samples in the spectral domain,
- partitioning the array of information samples into a first set of regions in the spatial domain based on the first maximum region size,
- sub-partitioning at least a subset of the first set of regions into a first set of sub-regions in the spatial domain based on the multi-tree subdivision information associated therewith,
- responsive to a determination that a size of at least one of the first set of sub-regions exceeds the second maximum region size, partitioning the at least one of the first set of sub-regions into a second set of regions in the spectral domain, and
- responsive to a determination that at least one of the second set of regions of the second maximum size is to be sub-divided, sub-partitioning the at least one of the second set of regions into a second set of sub-regions in the spectral domain based on the subordinate multi-tree subdivision information and the maximum hierarchy level associated therewith;
- and
- encoding each of the arrays of information samples by
- encoding, into the data stream, information related to first subdivision information, second subdivision information, and the maximum hierarchy level associated with each of the arrays of information samples, wherein the first subdivision information includes the first maximum region size and the multi-tree subdivision information, and the second subdivision information includes the second maximum region size and the subordinate multi-tree subdivision information.
7 more elements
A data stream stored in a non-transitory storage medium, the data stream comprising an encoded plurality of arrays of information samples, wherein each of the plurality of arrays of information samples is divided by
- determining, with respect to the array of information samples, a first maximum region size, multi-tree subdivision information, a second maximum region size, and subordinate multi-tree subdivision information for the array of information sample, wherein the first maximum region size and the multi-tree subdivision information are associated with regions of the array of information samples in a spatial domain, and the second maximum region size and the subordinate multi-tree subdivision information are associated with regions of the array of information samples in a spectral domain,
- determining a maximum hierarchy level for the array of information samples, wherein the maximum hierarchy level is associated with regions of the array of information samples in the spectral domain,
- partitioning the array of information samples into a first set of regions in the spatial domain based on the first maximum region size,
- sub-partitioning at least a subset of the first set of regions into a first set of sub-regions in the spatial domain based on the multi-tree subdivision information associated therewith,
- responsive to a determination that a size of at least one of the first set of sub-regions exceeds the second maximum region size, partitioning the at least one of the first set of sub-regions into a second set of regions in the spectral domain, and
- responsive to a determination that at least one of the second set of regions of the second maximum size is to be sub-divided, sub-partitioning the at least one of the second set of regions into a second set of sub-regions in the spectral domain based on the subordinate multi-tree subdivision information and the maximum hierarchy level associated therewith,
- wherein each of the arrays of information samples is encoded by encoding, into the data stream, information related to first subdivision information, second subdivision information, and the maximum hierarchy level for the array of information samples, wherein the first subdivision information includes the first maximum region size and the multi-tree subdivision information and the second subdivision information includes the second maximum region size and the subordinate multi-tree subdivision information.
4 more elements
Dependencies are inferred from the claim text.