Entropy encoding and decoding scheme

An entropy encoding apparatus comprising:

• a subdivider configured to subdivide a sequence of source symbols into a first subsequence of source symbols and a second subsequence of source symbols;
• a VLC encoder configured to symbol-wisely convert the source symbols of the first subsequence into codewords forming a first bit stream;
• and
• a PIPE encoder configured to encode the second subsequence of source symbols, represented in form of a sequence of alphabet symbols, the PIPE encoder comprising:
• an assigner configured to assign a measure for an estimate of a probability distribution among the possible values the respective alphabet symbols may assume, to each alphabet symbol of the sequence of alphabet symbols based on information comprised within previous alphabet symbols of the sequence of alphabet symbols;
• a plurality of entropy encoders each of which is configured to convert the alphabet symbols forwarded to the respective entropy encoder into a respective second bit stream;
• and
• a selector configured to forward each alphabet symbol to a selected one of the plurality of entropy encoders, the selection depending on the measure assigned to the respective alphabet symbol;
• wherein
• the assigner is configured to, in determining the measure for the estimate of the probability distribution for each alphabet symbol, quantize the probability distribution estimate associated with the context assigned to the respective alphabet symbol to one of a plurality of probability distribution estimate representatives in order to acquire the measure for the estimate of the probability distribution and wherein the selector is configured such that a surjective association is defined from the plurality of probability distribution estimate representatives to the plurality of entropy encoders;
• the assigner is configured to, in determining the measure for the estimate of the probability distribution for each alphabet symbol, quantize the probability distribution estimate associated with the context assigned to the respective alphabet symbol to one of a plurality of probability distribution estimate representatives in order to acquire the measure for the estimate of the probability distribution and wherein the selector is configured such that a surjective association is defined from the plurality of probability distribution estimate representatives to the plurality of entropy encoders;
• the selector is configured to change a quantization mapping from a range of the probability distribution estimate to the plurality of probability distribution estimate representatives in a predetermined deterministic way depending on previous alphabet symbols of the sequence of alphabet symbols over time;
• and
• the selector is configured to change the quantization mapping such that bitrates of the second bit streams into which the entropy encoders convert the alphabet symbols, are rendered less dispersed.

An entropy encoding apparatus comprising:

• a subdivider configured to subdivide a sequence of source symbols into a first subsequence of source symbols and a second subsequence of source symbols;
• a VLC encoder configured to symbol-wisely convert the source symbols of the first subsequence into codewords forming a first bit stream;
• and
• a PIPE encoder configured to encode the second subsequence of source symbols, represented in form of a sequence of alphabet symbols, the PIPE encoder comprising:
• an assigner configured to assign a measure for an estimate of a probability distribution among the possible values the respective alphabet symbols may assume, to each alphabet symbol of the sequence of alphabet symbols based on information comprised within previous alphabet symbols of the sequence of alphabet symbols;
• a plurality of entropy encoders each of which is configured to convert the alphabet symbols forwarded to the respective entropy encoder into a respective second bit stream;
• and
• a selector configured to forward each alphabet symbol to a selected one of the plurality of entropy encoders, the selection depending on the measure assigned to the respective alphabet symbol;
• wherein
• the plurality of entropy encoders are variable length coders configured to map alphabet symbol sequences to codewords;
• the entropy encoding apparatus further includes an interleaver configured to reserve a sequence of codeword entries for the codewords within the first bitstream and the second bit streams in a sequential order depending on an order in which the alphabet symbols of the sequence of alphabet symbols forwarded by the selector to the plurality of entropy encoders result in a beginning of a new alphabet symbol sequence to be mapped to a codeword at the respective entropy encoder and a new source symbol of the first subsequence is mapped by the VLC encoder, respectively, and to remove codewords entered into the codeword entries in the sequential order to acquire a single stream of interleaved codewords from the first bit stream and second bit streams, respectively, and
• each entropy encoder is configured to sequentially enter its codewords into the codeword entries reserved for the respective entropy encoder, and the selector is configured to forward the alphabet symbols representing the source symbols of the second sub-stream in an order maintaining an order in which the source symbols of the first sub-stream and the second sub-stream were interspersed within the sequence of source symbols.

An entropy decoding apparatus comprising:

• a VLC decoder configured to codeword-wisely reconstruct source symbols of a first subsequence of source symbols from codewords of a first bitstream;
• and
• a PIPE decoder configured to reconstruct a second subsequence of source symbols, represented in form of a sequence of alphabet symbols, and responsive to alphabet symbol requests sequentially requesting for the alphabet symbols, the PIPE decoder comprising:
• a plurality of entropy decoders, each of which is configured to convert a respective second bitstream into alphabet symbols of the sequence of alphabet symbols;
• an assigner configured to assign to each request for an alphabet symbol of the sequence of alphabet symbols representing the second subsequence of source symbols to be reconstructed, a measure of an estimate of a probability distribution among the possible values the respective alphabet symbol may assume, based on information comprised within previously reconstructed alphabet symbols of the sequence of alphabet symbols;
• and
• a selector configured to retrieve, for each request for an alphabet symbol of the sequence of alphabet symbols representing the second subsequence of source symbols to be reconstructed, the respective alphabet symbol of the sequence of alphabet symbols from a selected one of the plurality of entropy decoders, the selection depending on the measure assigned to the respective request for the respective alphabet symbol, wherein
• the first subsequence of source symbols and the second subsequence of source symbols commonly form a sequence of source symbols;
• the entropy decoding apparatus further includes a composer configured to compose a sequence of syntax elements from the sequence of source symbols by individually composing each syntax element from a respective integer number of source symbols, wherein the integer number of at least one syntax element is greater than one;
• and
• the composer is configured such that one of the source symbols of the integer number of source symbols of the at least one syntax element belongs to the first subsequence of source symbols and another one of the source symbols of the integer number of source symbols of the at least one syntax element belongs to the second subsequence of source symbols.

An entropy decoding apparatus comprising:

• a VLC decoder configured to codeword-wisely reconstruct source symbols of a first subsequence of source symbols from codewords of a first bitstream;
• and
• a PIPE decoder configured to reconstruct a second subsequence of source symbols, represented in form of a sequence of alphabet symbols, and responsive to alphabet symbol requests sequentially requesting for the alphabet symbols, the PIPE decoder comprising:
• a plurality of entropy decoders, each of which is configured to convert a respective second bitstream into alphabet symbols of the sequence of alphabet symbols;
• an assigner configured to assign to each request for an alphabet symbol of the sequence of alphabet symbols representing the second subsequence of source symbols to be reconstructed, a measure of an estimate of a probability distribution among the possible values the respective alphabet symbol may assume, based on information comprised within previously reconstructed alphabet symbols of the sequence of alphabet symbols;
• and
• a selector configured to retrieve, for each request for an alphabet symbol of the sequence of alphabet symbols representing the second subsequence of source symbols to be reconstructed, the respective alphabet symbol of the sequence of alphabet symbols from a selected one of the plurality of entropy decoders, the selection depending on the measure assigned to the respective request for the respective alphabet symbol;
• wherein
• the first subsequence of source symbols and the second subsequence of source symbols commonly form a sequence of source symbols;
• the assigner is configured to internally assign a context to each request for an alphabet symbol of the subsequence of alphabet symbols, based on the information comprised within previously reconstructed alphabet symbols of the subsequence of alphabet symbols with each context comprising a respective probability distribution estimate associated therewith, and to adapt the probability distribution estimate for each context to an actual symbol statistic based on symbol values of previously reconstructed alphabet symbols to which the respective context is assigned, and to determine, for each request for an alphabet symbol of the subsequence of alphabet symbols, the measure for the estimate of the probability distribution for the respective alphabet symbol of the subsequence of alphabet symbols based on the probability distribution estimate associated with the context assigned to the respective alphabet symbol;
• the assigner is configured to, in determining the measure for the estimate of the probability distribution for each request for an alphabet symbol, quantize the probability distribution estimate associated with the context assigned with the respective alphabet symbol to one of a plurality of probability distribution estimate representatives in order to acquire the measure for the estimate of the probability distribution and wherein the selector is configured such that a surjective association is defined from the plurality of probability distribution estimate representatives to the plurality of entropy decoders;
• the selector is configured to change a quantization mapping from a range of the probability distribution estimates to the plurality of probability distribution estimate representatives in a predetermined deterministic way depending on previously reconstructed alphabet symbols of the sequence of alphabet symbols, over time;
• and
• the selector is configured to change the quantization mapping such that rates by which the alphabet symbols are retrieved from the plurality of entropy decoders, are made less dispersed.

An entropy decoding method comprising:

• codeword-wisely reconstructing source symbols of a first subsequence of source symbols from codewords of a first bitstream;
• and
• reconstructing a second subsequence of source symbols, represented in form of a sequence of alphabet symbols, responsive to alphabet symbol requests seuquentially requesting for the alphabet symbols, and using a plurality of entropy decoders, each of which is configured to convert a respective second bitstream into alphabet symbols of the sequence of alphabet symbols, comprising:
• assigning to each request for an alphabet symbol of the sequence of alphabet symbols representing the second subsequence of source symbols to be reconstructed, a measure of an estimate of a probability distribution among the possible values the respective alphabet symbols may assume, based on information comprised within previously reconstructed alphabet symbols of the sequence of alphabet symbols;
• and
• retrieving, for each request for an alphabet symbol of the sequence of alphabet symbols representing the second subsequence of source symbols to be reconstructed, the respective alphabet symbol of the sequence of alphabet symbols from a selected one of the plurality of entropy decoders, the selection depending on the measure assigned to the request for the respective alphabet symbols, wherein the first subsequence of source symbols and the second subsequence of source symbols commonly form a sequence of source symbols;
• and
• the entropy decoding method further includes composing a sequence of syntax elements from the sequence of source symbols by individually composing each syntax element from a respective integer number of source symbols, wherein the integer number of at least one syntax element is greater than one and the composer is configured such that one of the source symbols of the integer number of source symbols of the at least one syntax element belongs to the first subsequence of source symbols and another one of the source symbols of the integer number of source symbols of the at least one syntax element belongs to the second subsequence of source symbols.

An entropy decoding apparatus comprising:

• a VLC decoder configured to codeword-wisely reconstruct source symbols of a first subsequence of source symbols from codewords of a first bitstream;
• a PIPE decoder configured to reconstruct a second subsequence of source symbols, represented in form of a sequence of alphabet symbols, and responsive to alphabet symbol requests sequentially requesting for the alphabet symbols, the PIPE decoder comprising:
• a plurality of entropy decoders, each of which is configured to convert a respective second bitstream into alphabet symbols of the sequence of alphabet symbols;
• an assigner configured to assign to each request for an alphabet symbol of the sequence of alphabet symbols representing the second subsequence of source symbols to be reconstructed, a measure of an estimate of a probability distribution among the possible values the respective alphabet symbol may assume, based on information comprised within previously reconstructed alphabet symbols of the sequence of alphabet symbols;
• and
• a selector configured to retrieve, for each request for an alphabet symbol of the sequence of alphabet symbols representing the second subsequence of source symbols to be reconstructed, the respective alphabet symbol of the sequence of alphabet symbols from a selected one of the plurality of entropy decoders, the selection depending on the measure assigned to the respective request for the respective alphabet symbol;
• wherein
• the first subsequence of source symbols and the second subsequence of source symbols commonly form a sequence of source symbols;
• the entropy decoders are variable length decoders configured to map codewords to alphabet symbol sequences;
• the entropy decoding apparatus further includes a deinterleaver configured to generate the first and second bitstreams by deinterleaving codewords within a single bitstream in an order defined by the order of the source symbols of the first subsequence and the second subsequence within the sequence of source symbols;
• the deinterleaver is configured to:
• whenever a source symbol within the sequence of source symbols currently in line, belongs to the first subsequence, regard a current codeword within single bitstream as a codeword of the first bitstream;
• and
• whenever a source symbol within the sequence of source symbols currently in line, belongs to the second subsequence, regard a current codeword within single bitstream as a codeword of a respective one of the second bitstreams, if any of the alphabet symbols belonging to the current source symbol necessitates a new mapping of a codeword of the respective second bitstreams to a respective alphabet symbol sequence by the entropy decoder selected depending on the measure assigned to the request for the respective alphabet symbol, using, if several of such alphabet symbols exist which belong to the current source symbol, an order among such alphabet symbols, which depends on the order of the alphabet symbols within the sequence of alphabet symbols.

An entropy encoding method comprising:

• converting a sequence of syntax elements into the sequence of source symbols by individually decomposing at least a subgroup of the syntax elements into a respective integer number of source symbols such that a mathematical function common for the subgroup, applied upon the integer number of source symbols of the respective syntax element, yields the respective syntax element;
• subdividing the sequence of source symbols into a first subsequence of source symbols and a second subsequence of source symbols, such that one of the source symbols of the integer number of source symbols of the respective syntax elements belongs to the first subsequence of source symbols and another one of the source symbols of the integer number of source symbols of the respective syntax elements belongs to the second subsequence of source symbols;
• symbol-wisely converting the source symbols of the first subsequence into codewords forming a first bit stream;
• and
• encoding the second subsequence of source symbols, represented in form of a sequence of alphabet symbols, comprising:
• assigning a measure for an estimate of a probability distribution among the possible values the respective alphabet symbols may assume, to each alphabet symbol of the sequence of alphabet symbols based on information comprised within previous alphabet symbols of the sequence of alphabet symbols;
• and
• forwarding each alphabet symbol to a selected one of a plurality of entropy encoders each of which is configured to convert the alphabet symbols forwarded to the respective entropy encoder into a respective second bit stream, the selection depending on the measure assigned to the respective alphabet symbol.

An entropy decoding apparatus comprising:

• a VLC decoder configured to codeword-wisely reconstruct source symbols of a first subsequence of source symbols from codewords of a first bitstream:
• and
• a PIPE decoder configured to reconstruct a second subsequence of source symbols, represented in form of a sequence of alphabet symbols, and responsive to alphabet symbol requests sequentially requesting for the alphabet symbols, the PIPE decoder comprising:
• a plurality of entropy decoders, each of which is configured to convert a respective second bitstream into alphabet symbols of the sequence of alphabet symbols;
• an assigner configured to assign to each request for an alphabet symbol of the sequence of alphabet symbols representing the second subsequence of source symbols to be reconstructed, a measure of an estimate of a probability distribution among the possible values the respective alphabet symbol may assume, based on information comprised within previously reconstructed alphabet symbols of the sequence of alphabet symbols;
• and
• a selector configured to retrieve, for each request for an alphabet symbol of the sequence of alphabet symbols representing the second subsequence of source symbols to be reconstructed, the respective alphabet symbol of the sequence of alphabet symbols from a selected one of the plurality of entropy decoders, the selection depending on the measure assigned to the respective request for the respective alphabet symbol;
• wherein
• the first subsequence of source symbols and the second subsequence of source symbols commonly form a sequence of source symbols;
• at least one entropy decoder has a symbol output buffer associated therewith, wherein the selector is configured to retrieve the alphabet symbols from the at least one entropy decoder via the associated symbol output buffer;
• the entropy decoders are variable length decoders configured to map codewords to alphabet symbol sequences;
• the entropy decoding apparatus further comprises a deinterleaver configured to generate the first and second bitstreams by deinterleaving codewords within a single bitstream in an order defined by the order of the source symbols of the first subsequence and the second subsequence within the sequence of source symbols;
• and
• the deinterleaver is configured to:
• whenever a source symbol within the sequence of source symbols currently in line, belongs to the first subsequence, regard a current codeword within single bitstream as a codeword of the first bitstream,and
• whenever a source symbol within the sequence of source symbols currently in line, belongs to the second subsequence, regard a current codeword within single bitstream as a codeword of a respective one of the second bitstreams, if any of the alphabet symbols belonging to the current source symbol necessitates a new mapping of a codeword of the respective second bitstreams to a respective alphabet symbol sequence by the entropy decoder selected depending on the measure assigned to the request for the respective alphabet symbol, using, if several of such alphabet symbols exist which belong to the current source symbol, an order among such alphabet symbols, which depends on the order of the alphabet symbols within the sequence of alphabet symbols.