WO2005096636A1 - 符号化信号分離装置、符号化信号合成装置および符号化信号分離合成システム - Google Patents
符号化信号分離装置、符号化信号合成装置および符号化信号分離合成システム Download PDFInfo
- Publication number
- WO2005096636A1 WO2005096636A1 PCT/JP2005/006171 JP2005006171W WO2005096636A1 WO 2005096636 A1 WO2005096636 A1 WO 2005096636A1 JP 2005006171 W JP2005006171 W JP 2005006171W WO 2005096636 A1 WO2005096636 A1 WO 2005096636A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- coefficient
- extended
- basic
- separation
- sequence
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/266—Channel or content management, e.g. generation and management of keys and entitlement messages in a conditional access system, merging a VOD unicast channel into a multicast channel
- H04N21/2662—Controlling the complexity of the video stream, e.g. by scaling the resolution or bitrate of the video stream based on the client capabilities
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/30—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using hierarchical techniques, e.g. scalability
Definitions
- Coded signal separation device coded signal synthesis device, and coded signal separation and synthesis system
- the present invention relates to a coded signal separation device, a coded signal synthesis device, and a coded signal separation / synthesis system that can perform scalable video transmission.
- a standard for encoding a digital video and an accompanying audio is referred to as ISOZIEC 13818 (also known as “MPEG-2” (Moving Picture Expert Group Phase 2).
- MPEG-2 bit stream The bit stream conforming to the MPEG-2 standard (hereinafter referred to as “MPEG-2 bit stream”) generated in this manner is used in a wide range of fields such as communication and television broadcasting!
- MPEG The two-bit stream has a hierarchical structure, and the GOP starts from the highest sequence layer.
- each screen is saved in a frame memory, and a difference between frames is taken to reduce redundancy in a time axis direction. Further, by performing orthogonal transform processing such as discrete cosine transform (hereinafter abbreviated as “DCT”) on a plurality of pixels constituting each frame to reduce the redundancy in the spatial axis direction, the efficiency is improved. Video compression encoding is realized.
- DCT discrete cosine transform
- the encoded signal is sent to a decoder, decoded and reproduced.
- the decoder reproduces the screen, stores it in the first frame memory, predicts the next screen to be continued based on the difference information, stores it in the second frame memory, and displays the screen inserted between the two frames. Furthermore, a series of screens are constructed and a moving image is reproduced by making predictions. Such a method is called bidirectional prediction.
- an I picture is an abbreviation of an intra-coded picture, and is a screen that is coded as a still picture independently of other pictures.
- the P picture is an abbreviation of a forward predictive coding picture, and is a picture that is predictively coded based on an I or P picture located in the past in time.
- a B picture is an abbreviation of a bidirectional predictive encoding picture, and is predictively encoded based on a forward, backward or bidirectional picture using I or P pictures located before and after in time. It is a screen. That is, after the I picture and the P picture are first encoded, the B picture inserted between them is encoded.
- the MPEG-2 bit stream encoded by the encoder is transmitted at a predetermined transfer rate to a transmission path, input to a decoder on the transmission path, decoded and reproduced.
- the amount of information generated by coding a moving image is not constant.
- the amount of information increases at a stretch.
- TM5 (April, 1993) (hereinafter abbreviated as "TM5") describes a rate control method.
- step 1 bits are first allocated to each picture type based on the code amount R allocated to uncoded pictures in a GOP.
- the virtual buffer occupancy calculated based on the bit allocation is also calculated for the quantization scale used when performing the encoding process on a macroblock basis.
- MPEG-2 bitstreams having different compression formats and different transmission speeds are used.
- a moving picture compression coding signal converting apparatus for converting the video signal is required.
- a device for realizing this is a so-called transcoder.
- the image compression coded signal transmitted from the encoder is converted into an appropriate signal by a transcoder, and the signal is supplied to each decoder.
- FIG. 25 shows a first example of a general conventional transcoder 50.
- a conventional transcoder 50 is connected to a first transmission line (not shown) having a first bit rate, and a first MPEG-2 bit A variable length decoding unit (VLD) 51 that inputs the stream bl, an inverse quantizer 53, a quantizer 55, and a second transmission path (not shown) having a second bit rate are connected to the second MPEG-2 A VLC 57 that outputs the bit stream b2 and a rate control unit 59 that controls the amount of code generated by the quantizer 55 are provided.
- the second bit rate is a lower transfer rate than the first bit rate.
- the VLD 51 and the inverse quantizer 53 decode the first MPEG-2 bit stream bl up to the DCT coefficient region in macroblock units, and encode the obtained DCT coefficient signal by the quantizer 55 and the VLC 57. Then, a second MPEG-2 bit stream b2 having a smaller code amount than the first MPEG-2 bit stream is generated.
- the coefficient obtained by the DCT transform is divided by a predetermined quantization step. As a result, the image signal is compressed.
- This quantization step is obtained by multiplying a plurality of quantization matrix values included in a predetermined quantization table by a quantization scale.
- the transcoder 50 almost reuses the coding information of the sequence layer, GOP layer, picture layer, slice layer, and macroblock layer in the first MPEG-2 bit stream bl. Basically, only the conversion of the block layer DCT coefficients and the conversion of the macroblock layer code that needs to be modified with the block layer conversion are performed
- the rate control unit 59 uses the MPEG-2 T
- FIG. 26 shows a flowchart of the rate control process of the conventional transcoder 50. As shown in the figure, the conventional rate control process has steps A1 to A14.
- a variable n is set to 1.
- the variable n indicates a number assigned to a plurality of pictures included in the input image signal, and the n-th picture is hereinafter referred to as pic (n).
- indices Xi, Xp and Xb indicating the complexity of the I, P and B pictures are calculated by the following equations (al), (a2) and (a3).
- Si, Sp, and Sb are generated code amounts of I, P, and B pictures, respectively
- Qi, Qp, and Qb are quantized scale codes of all macroblocks in I, P, and B pictures, respectively.
- the screen complexity index Xi, Xp, and Xb is large for an image in which a large amount of encoded information is generated, that is, for an image with a low compression rate, and conversely, for an image with a high compression rate. Will be smaller.
- target_Bitrate is a target bit rate of the transcoder 50.
- the code amounts Ti, Tp and Tb to be allocated to the I, P and B pictures in the GOP are calculated by the following equations (a7), (a8) and (a9).
- Np and Nb indicate the numbers of uncoded P and B pictures, respectively, in the GOP.
- step A4 it is determined whether or not the variable n is 1. That is, it is determined whether the picture to be coded is the first picture pic (l). If it is the first picture, go to step A5. If it is not the first picture, go to step A6. In step A5, the code amount R to be assigned to the uncoded picture in the GOP when the first picture pic (l) in the GOP is coded is calculated by the following equation (alO).
- R target—Bitrate X NZpicture— rate + R... Equation (alO)
- N is the total number of pictures in the GOP
- picture_rate is a value indicating the time resolution of the input image, and indicates the number of screens decoded and displayed in one second.
- step A6 the code amount R allocated to the uncoded picture in the GOP is changed to the generated code of the I, P, and B pictures when the (n-1) th picture pic (nl) is coded. Based on the quantity Si, Sp, or Sb, update with any of the following formulas (all), (al2) and (al3).
- R R-Si ... Formula (all)
- R R- Sb ... Formula (al3)
- Steps A5 and A6 both proceed to step A7, and set 1 to a variable j.
- the variable j indicates a number assigned to a plurality of macroblocks in one picture, and the j-th macroblock is hereinafter referred to as MB (j).
- step A8 the occupation amounts di (j), dp (j) and db (j) of the virtual buffer when encoding the j-th macroblock MB (j) in the I, P and B pictures. ) Is calculated by the following equations (al4), (al5) and (al6).
- NMB d b (j) d ( (0) + Bfj-1) Tb 1 -... "Equation (al6)
- B (j ⁇ 1) is the generated code amount of all macroblocks up to the (j ⁇ 1) th macroblock MB (j ⁇ 1).
- di (0), dp (0), and db (0) are initial values of the virtual buffer occupancy of the I, P, and B pictures, respectively. 18) and equation (a 19), respectively.
- r is called a reaction parameter, and is represented by the following equation (a20), and controls the response speed of the feedback loop.
- the virtual buffer occupancy at the end of I, P, and B picture encoding that is, the virtual buffer occupancy di (NMB), dp (NMB) when the NMB-th macroblock MB (NMB) is encoded.
- db (NMB) are used as initial values di (0), dp (0), and db (0) of the virtual buffer occupancy at the next encoding for each picture type.
- the quantization scale code QG for the j-th macroblock MB (j) is calculated for each picture by the following equation (a21). ).
- step A10 the j-th macroblock MB (j) is quantized using the quantization scale code Q (j) calculated in step A9.
- the variable j is incremented, and the process proceeds to step A12 to determine whether the variable j exceeds the total number of macroblocks NMB.
- NMB is the total number of macroblocks included in the n-th picture pic (n). If the variable j does not exceed the total number of macroblocks NMB, the process returns to step A8. If the variable j exceeds the total number of macroblocks NMB, the process proceeds to step A13.
- the variable j is also used as a loop counter for repeating the encoding process of steps A8 to A11.
- step Al 3 the variable n is incremented, and the flow advances to step A 14 to determine whether or not the variable n exceeds the total number NPIC of pictures to be coded. If the variable n does not exceed the total number of pictures NPIC, the process returns to step A2, and if the variable n exceeds the total number of pictures NPIC, the process ends.
- the first transcoder 50 since the first transcoder 50 cannot have information on the image structure such as the I and P picture periods, the first transcoder 50 uses an image such as the TM5 rate control shown in FIG. A method of allocating bits based on information such as the GOP structure cannot be performed unless the input image structure is assumed!
- the second conventional transcoder 60 includes, in addition to the configuration of the first conventional transcoder 50, a delay circuit 61, a bit rate ratio calculation unit 63, and an input code amount. It includes an integrating section 65, a difference code amount calculation section 67, a target output code amount update section 69, and a quantization scale code calculation section 71.
- FIG. 28 shows a processing flow of the transcoder 60 thus configured.
- the process of the transcoder 60 includes steps B1 to B13.
- Steps B6 to B13 are the same as steps A7 to A14 of the rate processing shown in the first conventional example.
- the virtual buffer occupancy is calculated based on the target output code amount Tout calculated by the target output code amount update unit 69.
- FIGS. 29 and 30 show a third example of a conventional transcoder.
- a third conventional transcoder 80 is connected to a first transmission line having a first bit rate, inputs a VLD 81 that inputs an input bit stream b3, and a first conventional transcoder 50.
- the code amount is described in advance in the bit stream b3 as information, and rate control is performed based on the information.
- the transcoder operates on the signal after the encoding process!
- the original signal before conversion cannot be known. Therefore, in the code amount control, attention is paid to the distortion newly generated by the requantization processing instead of the distortion of the image itself after the transcoding processing, and by suppressing this distortion, the deterioration of the image quality is suppressed.
- the amount of code must be reduced.
- the applicant of the present application considers the decoding quantization parameter and the quantization calculated in the preceding stage by considering the requantization rate distortion function that depends on the decoding quantization parameter and the requantization parameter first.
- a moving picture compression coded signal conversion method and apparatus for realizing the calculation of an optimal quantization parameter based on parameters, and a medium recording a conversion program (for example, Japanese Patent Application Laid-Open No. 2001-169283 (Europe corresponding to Europe). Patent Publication EP1067798).
- the quantization parameter switching unit for switching the quantization parameter, it is possible to minimize the error when converting the quantization coefficient area data into the requantization coefficient area data.
- the transcoder is a processor that realizes bit stream conversion into a form suitable for various use environments.
- bit rate scaling scheme As a technique for realizing scalability of generating and providing a video stream suitable for a usage environment in a network in which various bands are mixed, a bit rate scaling scheme and a hierarchical coding scheme exist.
- bit rate scaling transcoder installed in each router on a network
- stream conversion by bit rate reduction is realized according to a required bit rate.
- video scalability in response to network fluctuations is realized.
- bit rate scaling transcoder itself performs rate control and stream conversion, a high level of processing is required for the router, so that a large load S is applied.
- the stream that is preliminarily input by the server is converted into a stream having a hierarchical structure, and scalability is realized by selecting a layer in a router.
- rate control by a lance coder
- load on signal processing at each router on the network is also eliminated.
- each layer is given a priority, and there is a problem that it is necessary to perform control in consideration of the priority when selecting the layer.
- there is a problem in that a large amount of load is required to divide the stream, and the processing speed is reduced.
- the present invention has been made in order to solve such a conventional problem.
- a separator in a server an independent separated stream is generated with a simple configuration, and a router having a band selection function is provided.
- a coded signal separation device, a coded signal separation device and a coded signal separation / synthesis system capable of performing scalable video transmission with an improved processing speed by selecting a transmission path according to a network state Is provided. Disclosure of the invention
- the coded signal separation device comprises: a separator input means for inputting a primary encoded signal obtained by encoding a primary moving image composed of a plurality of pieces of image information; and the separator input means.
- a primary coded signal having a smaller code amount than the primary coded signal for restoring a secondary video which is a pseudo video of the primary video A tertiary moving image closer to the primary moving image than the secondary moving image restored by the basic encoding signal is separated into a plurality of extended coded signals to be restored together with the basic coded signal.
- a plurality of transmission coded signals are generated by arbitrarily combining and multiplexing the demultiplexer separation means, the basic coded signal separated by the demultiplexer separation means, and the plurality of extension coded signals.
- a primary coefficient conversion unit configured to convert the primary encoded signal into a primary quantized coefficient sequence, which is a continuous sequence of primary quantized coefficient values forming the primary video, from the primary encoded signal;
- the coefficient position index number when the reference order of the primary quantized coefficient sequence converted by the primary coefficient conversion means in the block is the order of zigzag scan is smaller than a predetermined basic extended separation coefficient number! / ⁇ is separated into a base layer coefficient column, which is a column of coefficient values, and an extended layer coefficient column, which is a column of coefficient values whose order is equal to or larger than the basic expansion separation coefficient number.
- the basic coded signal generation means generates a logarithm for each coefficient value of the base layer coefficient sequence.
- a basic quantized coefficient sequence which is a sequence of basic quantized coefficients obtained by rounding the coefficient values by transform, and a difference between each coefficient value of the basic hierarchical coefficient sequence and each coefficient value restored from the basic quantized coefficient map.
- a residual coefficient sequence that is a sequence, a basic quantization coefficient conversion unit that converts and separates the conversion into a basic quantization coefficient sequence, and a basic quantization coefficient encoding unit that encodes the basic quantization coefficient sequence into a basic quantization coefficient encoded signal.
- a remainder coefficient encoding unit that encodes the remainder coefficient sequence into a remainder coefficient encoded signal, the basic quantized coefficient encoded signal, and the residual coefficient encoded signal are multiplexed to generate the basic encoded signal
- Basic quantization coefficient multiplexing means A configuration in which wherein, Ru.
- the basic quantization coefficient coding means includes n kinds of eigenvalues and values other than these eigenvalues.
- An array pattern input means for inputting an array pattern code table that encodes an array pattern of a sequence of (n + 1) types of values, and an array pattern of coefficient values of the basic quantized coefficient sequence,
- a coefficient value array pattern coding means for applying and coding the array pattern, and a non-eigenvalue coefficient value code for coding values other than the n kinds of eigenvalues among the coefficient values of the basic quantized coefficient sequence.
- a converting means for converting an array pattern code table that encodes an array pattern of a sequence of (n + 1) types of values, and an array pattern of coefficient values of the basic quantized coefficient sequence.
- the array pattern input means of the basic quantization coefficient coding means includes: 0, 1, and 2; An array pattern code table that encodes an array pattern of a sequence of three values of the above values and the above values is input, and the coefficient value array pattern encoding means performs an array of coefficient values of the basic quantized coefficient sequence.
- the pattern is applied to the array pattern, and the eigenvalue extraneous value encoding means calculates two or more coefficient values among the respective coefficient values of the basic quantized coefficient sequence. It has a configuration characterized by signing.
- the coded signal separation device is the coded signal separation device according to claim 1, wherein the extended coded signal generation means separates the extended hierarchical coefficient sequence into a plurality.
- Extended quantized coefficient separating means for separating the extended hierarchical coefficient sequence into a plurality of extended quantized coefficient sequences in accordance with the extended hierarchical separation pattern information defining And an extended quantization coefficient encoding means for encoding.
- the extended layer separation pattern information is a pattern of a pattern that separates the coefficients of the extended layer coefficient sequence into a plurality.
- Extended layer separation minimum pattern information that defines the number of application coefficients and separation patterns, and a separation minimum pattern application unit that defines a layer in a coded signal to which the enhancement layer separation minimum pattern is applied
- Extended quantization coefficient separation means separates the extended hierarchy coefficient sequence into a plurality of the extended quantization coefficient sequences according to the extended hierarchy separation minimum pattern information for each layer specified in the minimum separation pattern application unit information. It has a configuration characterized by the following.
- the extended hierarchical separation pattern information is separated into the plurality of extended quantized coefficient sequences.
- the separation method is defined such that the number of coefficients in the hierarchical coefficient sequence is different for each of the extended quantized coefficient sequences, and the extended quantized coefficient separating means is configured so that the number of coefficients is different for each of the extended quantized coefficient sequences.
- the extended hierarchical coefficient sequence is separated into a plurality of extended quantized coefficient sequences according to the extended hierarchical separation pattern information in which the separation method is defined.
- the coded signal separation device is the coded signal separation device according to any one of claims 5 to 7, wherein the extended hierarchy separation pattern information is encoded by the extended coding.
- An extended layer separation pattern information transmitting unit that transmits only to a specific receiver that has been permitted to decode a moving image without including the signal in a signal, has a configuration characterized by comprising:
- the encoding apparatus for synthesizing signals includes a synthesizer input means for inputting a plurality of independent transmission coded signals obtained by coding a primary video composed of a plurality of pieces of image information.
- the code amount is smaller than that of the primary coded signal for restoring a secondary video, which is a pseudo video of the primary video, from the plurality of transmission coded signals input to the synthesizer input means.
- Combiner separating means for separating the coded signal into the coded signal, the basic coded signal separated by the combiner separating means, and a plurality of the extended coded signals to restore a tertiary video image
- Combining means for generating a tertiary coded signal, and combining means for outputting a tertiary coded signal combined with the combining means.
- Basic floor for restoring the secondary video from coded signal A basic coded signal converting means for converting into a basic hierarchical coefficient sequence of a plurality of extended coded signals, and extended encoding for converting the plurality of extended coded signals into an extended hierarchical coefficient sequence of an extended hierarchy used when restoring the tertiary video image
- the signal conversion means and the coefficient position index number when the reference order in the block is the zigzag scan order in the basic hierarchical coefficient sequence converted by the basic coded signal conversion means is smaller than a predetermined basic extension separation coefficient number.
- the base layer coefficient sequence and the extended layer coefficient sequence are combined as a coefficient sequence, and the extended layer coefficient sequence converted by the extended coded signal conversion means is a coefficient sequence whose order is equal to or larger than the basic extended separation coefficient number.
- the third-order coefficient converting means for converting the tertiary code I No. ⁇ , further comprising a has a configuration which is characterized, Ru.
- the coded signal synthesizing device converts the basic coded signal into a logarithmic transform of the basic hierarchical coefficient sequence.
- a basic quantization coefficient multiplexing / demultiplexing means for demultiplexing a basic quantization coefficient coded signal rounded by the following and a remainder coefficient coded signal rounded at the time of calculating the basic quantization coefficient;
- a basic quantization coefficient conversion means for performing coefficient conversion on the signal and converting the signal into a basic quantization coefficient sequence, a coefficient conversion means for performing a coefficient conversion on the residual coefficient coded signal and converting the signal into a residual coefficient sequence, An antilogarithmic transformation is performed on each coefficient value of the basic quantized coefficient sequence, and the result is added to each coefficient value of the remainder coefficient sequence.
- a basic quantization coefficient combining means for combining the hierarchical coefficient sequence.
- the basic quantization coefficient conversion means includes n kinds of eigenvalues and values other than these eigenvalues.
- a coefficient value array pattern decoding means for decoding the eigenvalue and a value other than the eigenvalue according to the array pattern code table; and decoding a value other than the eigenvalue from the basic quantized coefficient coded signal, And a coefficient value decoding means for decoding a numerical value outside the eigenvalue.
- the encoded signal synthesizing device is the encoded signal synthesizing device according to any one of claims 9 to 11, wherein the extended encoded signal converting means includes: Extended quantized coefficient decoding means for decoding each of the extended coded signals into an extended quantized coefficient sequence, and the plurality of the plurality of extended hierarchical coefficient sequences based on extended layer separation pattern information defining a separation method for separating the extended layer coefficient sequence into a plurality. Extended quantized coefficient synthesizing means for synthesizing an extended quantized coefficient sequence with the extended hierarchical coefficient sequence.
- the extended layer separation pattern information includes a plurality of coefficients of the extended layer coefficient sequence.
- An extended layer separation minimum pattern information that defines the number of application coefficients of a pattern to be separated and a separation pattern, and a separation minimum pattern application unit that defines a layer in a coded signal to which the extended layer separation minimum pattern is applied.
- the extended quantization coefficient synthesizing means for each of the layers defined in the separated minimum pattern application unit information, stores the plurality of extended quantized coefficient sequences based on the extended layer separated minimum pattern information. It has a configuration characterized by being combined into a column.
- a primary encoded signal obtained by encoding a primary moving image composed of a plurality of pieces of image information is a pseudo moving image of the primary moving image.
- a tertiary video closer to the primary video than the secondary video restored by the basic coded signal is separated into a plurality of extended coded signals to be recovered and used together with the basic coded signal.
- a separator for reconstructing and converting to a plurality of transmission coded signals to be transmitted on a network, and selecting the transmission coded signal to be input with the plurality of transmission coded signals and to be transferred,
- a transmission path selector for transmitting the selected transmission encoded signal, and a plurality of transmission encoded signals transmitted to the transmission path selector are input, and a tertiary encoded signal for restoring the tertiary moving image is synthesized.
- a synthesizer for reconstructing and converting to a plurality of transmission coded signals to be transmitted on a network, and selecting the transmission coded signal to be input with the plurality of transmission coded signals and to be transferred.
- the separator includes a separator input unit that inputs the primary coded signal, a primary coded signal input to the separator input unit, the base coded signal, and the plurality of extension coded signals.
- a plurality of extension coded signals arbitrarily combined with the base coded signal separated by the separator separation means, and the plurality of transmission signals.
- Separator multiplexing means for generating an encoded signal; andseparator output means for outputting a plurality of transmission encoded signals multiplexed to the separator multiplexing means,
- a combiner input means for inputting the plurality of transmission coded signals; a plurality of the transmission coded signals input to the combiner input means; a basic coded signal; A tertiary moving image by combining an extended coded signal, a combiner separating unit that separates the extended coded signal, the basic coded signal separated by the combiner separating unit, and a plurality of the extended coded signals.
- Synthesizer generating means for generating a tertiary encoded signal for restoring the signal, and synthesizer output means for outputting the tertiary encoded signal synthesized by the synthesizer synthesizing means.
- the separator separating means converts the primary coded signal into a primary quantized coefficient sequence which is a continuous sequence of primary quantized coefficient values constituting the primary moving picture.
- ! / ⁇ a base enhancement layer separation unit that separates into a base layer coefficient column that is a column of coefficient values, and an extension layer coefficient column that is a column of coefficient values whose order is equal to or greater than the basic expansion separation coefficient number;
- the basic coded signal is generated from a hierarchical coefficient sequence.
- Basic coded signal generating means for generating a plurality of the extended coded signals from the extended hierarchical coefficient sequence,
- Basic coded signal converting means for converting the basic coded signal into the basic layer coefficient sequence, and extended coding for converting the plurality of extended coded signals into the extended layer coefficient sequence.
- the sequence is a coefficient sequence whose order is equal to or greater than the basic extended separation coefficient number, and the base layer coefficient sequence and the extended layer coefficient sequence are combined to form a continuous sequence of tertiary quantization coefficient values.
- Basic enhancement layer synthesis means for generating a coefficient sequence
- tertiary coefficient conversion means for converting the tertiary quantization coefficient sequence synthesized by the basic enhancement layer synthesis means into the tertiary code conversion signal.
- the basic coded signal generation means of the separator separation means includes the basic layer coefficient.
- a basic quantized coefficient sequence which is a sequence of basic quantized coefficients obtained by rounding the coefficient values by logarithmic conversion for each coefficient value in the column, and each coefficient value of the basic hierarchical coefficient sequence and the basic quantized coefficient are restored.
- the basic coded signal converting means of the synthesizer combining means demultiplexes the basic coded signal into the basic quantized coefficient coded signal and the residual coefficient coded signal. Separating means, performing coefficient conversion on the basic quantized coefficient coded signal, and converting it into the basic quantized coefficient sequence, and performing coefficient conversion on the residual coefficient coded signal; A coefficient conversion means for converting into a coefficient sequence, and performing antilogarithmic conversion on each coefficient value of the basic quantized coefficient sequence! ⁇ , and each coefficient value of the remainder coefficient sequence And a basic quantized coefficient synthesizing means for synthesizing the basic hierarchical coefficient sequence by adding the above.
- the basic quantization coefficient coding means of the separator includes n kinds of eigenvalues and An array pattern input means for inputting an array pattern code table for encoding an array pattern of a sequence of (n + 1) values, using one value other than these eigenvalues, and the basic quantization coefficient sequence Coefficient value array pattern coding means for applying and coding the coefficient value array pattern to the array pattern, and, among the coefficient values of the basic quantization coefficient sequence, values other than the n kinds of eigenvalues.
- the basic quantized coefficient conversion means of the synthesizer inputs the array pattern code table, the array pattern input means, and the encoded value of the basic quantized coefficient coded signal according to the array pattern code table.
- the extended coded signal generation means of the separator separation means is characterized in that: Extended quantized coefficient separating means for separating the extended hierarchical coefficient sequence into a plurality of extended quantized coefficient sequences in accordance with extended hierarchical separation pattern information defining a separation method for separating the extended hierarchical coefficient sequence into a plurality; Extended quantization coefficient encoding means for encoding each of the extended quantization coefficient sequences into the extended encoded signal,
- the extended coded signal converting means of the combiner combining means decodes the plurality of extended coded signals into the extended quantized coefficient sequence, respectively, based on the extended quantized coefficient decoding means and the extended hierarchical separation pattern information.
- Extended quantized coefficient synthesizing means for synthesizing the plurality of extended quantized coefficient sequences into the extended hierarchical coefficient sequence.
- the coded signal separation / combination system of the present invention provides a coded signal separation / combination system according to claim 17.
- the extended layer separation pattern information includes extended layer separated minimum pattern information that defines the number of applied coefficients of a pattern that separates the coefficients of the extended layer coefficient sequence into a plurality of patterns and the pattern of separation.
- a separation minimum pattern application unit that defines a hierarchy in a coding signal to which the minimum separation pattern is applied, wherein the extended quantization coefficient separation means of the separator applies the extended hierarchy coefficient sequence to the separation minimum pattern application.
- the extended quantized coefficient sequence is separated into a plurality of extended quantized coefficient sequences according to the extended hierarchical separation minimum pattern information, Applying the coefficient sequence to the separation minimum pattern application unit for each layer specified in the unit information based on the expansion layer separation minimum pattern information Has a structure obtained by said synthesizing the layer coefficient sequence.
- the extended layer separation pattern information is included in the plurality of extended quantization coefficient sequences.
- the separation method is defined such that the number of coefficients of the extended hierarchical coefficient sequence to be separated is different for each of the extended quantization coefficient sequences
- the extended quantization coefficient separating means of the separator includes: Separating the extended hierarchical coefficient sequence into a plurality of extended quantized coefficient sequences in accordance with the extended hierarchical separation pattern information in which the separating method is defined such that the number of coefficients is different.
- Means The apparatus is characterized in that the plurality of extended quantized coefficient sequences are combined with the extended hierarchical coefficient sequence according to the extended hierarchical separation pattern information.
- the coded signal separation / combination system is configured such that the separator includes the extension unit.
- An extended layer separation pattern information transmitting means for transmitting the layer separation pattern information only to a specific receiver authorized to decode a moving image without including the layer separation pattern information in the extended coded signal; It has a configuration characterized by comprising extended layer separation pattern information receiving means for receiving layer separation pattern information.
- the encoded signal separation method and the encoded signal separation program of the present invention include a separator inputting step of inputting a primary encoded signal obtained by encoding a primary moving image composed of a plurality of pieces of image information.
- the primary encoding signal input in the separator input step A basic coded signal having a smaller code amount than the primary coded signal for restoring a secondary moving image, which is a pseudo moving image of the next moving image, and the first moving image recovered from the basic coded signal, A plurality of extended coded signals to be restored by using a tertiary moving image close to the next moving image together with the basic coding signal, and a separator separating step of separating into a plurality of extended coded signals; and A base coded signal and a plurality of the extended coded signals are multiplexed in any combination to generate a plurality of transmission coded signals, and multiplexed in the demultiplexer multiplexing step.
- a first-order coefficient conversion for converting the first-order coding signal into a first-order quantization coefficient sequence which is a continuous sequence of the first-order quantization coefficient values constituting the first-order moving image;
- the basic extended separation coefficient number having a predetermined index position index number when the step and the primary quantization coefficient sequence converted in the primary coefficient conversion step are arranged in a zigzag scan order in a block.
- the coded signal separation method and the coded signal separation program according to the present invention provide the coded signal separation method and the coded signal separation program according to claim 21 and claim 36.
- the step is a basic quantization coefficient sequence, which is a sequence of basic quantization coefficients obtained by rounding coefficient values by logarithmic conversion for each coefficient value of the basic layer coefficient sequence, and each coefficient value of the basic layer coefficient sequence and A basic quantized coefficient conversion step of converting and separating the residual quantized coefficient sequence, which is a sequence of differences from each coefficient value restored from the basic quantized coefficient, into a basic quantized coefficient code;
- Signal and a multiplexed, basic quantization coefficient multiplexing to generate the basic coded signal And a configuration characterized by comprising:
- the encoded signal separation method and the encoded signal separation program of the present invention are characterized in that, in the encoded signal separation method and the encoded signal separation program according to Claim 22 and Claim 37, Pattern encoding step, which encodes an array pattern code table that encodes an array pattern of a sequence of (n + 1) values, where n types of eigenvalues and one value other than these eigenvalues are one type
- Pattern encoding step which encodes an array pattern code table that encodes an array pattern of a sequence of (n + 1) values, where n types of eigenvalues and one value other than these eigenvalues are one type
- An input step a coefficient value array pattern encoding step of applying and encoding an array pattern of coefficient values of the basic quantization coefficient sequence to the array pattern, and among the respective coefficient values of the basic quantization coefficient sequence, A non-eigenvalue coefficient value encoding step of encoding values other than the n types of eigenvalues.
- the encoded signal separation program according to the present invention is the encoded signal separation program according to claim 38, wherein the array pattern input steps of the basic quantization coefficient encoding step are 0, 1 And an array pattern code table for encoding an array pattern of a sequence of three types of values, i.e., two or more values, wherein the coefficient value array pattern encoding step includes the step of encoding the basic quantized coefficient sequence.
- a numerical value array pattern is applied to the array pattern and encoded, and the non-eigenvalue coefficient value encoding step encodes two or more coefficient values among the coefficient values of the basic quantized coefficient sequence. It has a characteristic configuration.
- the coded signal separation method and the coded signal separation program according to the present invention provide the coded signal separation method and the coded signal separation program according to claim 21 and claim 36, wherein the extended code
- the enhanced signal generation step includes an extended quantization coefficient separation unit that separates the enhancement layer coefficient sequence into a plurality of extension quantization coefficient sequences in accordance with extension layer separation pattern information that defines a separation method for separating the enhancement layer coefficient sequence into a plurality.
- the coded signal separation method and the coded signal separation program according to the present invention in the coded signal separation method and the coded signal separation program according to claim 24 and claim 40, wherein the extended layer
- the separation pattern information specifies the number of applied coefficients of the pattern that separates the coefficients of the expanded hierarchy coefficient sequence into a plurality of coefficients and the expansion floor that defines the separation pattern.
- the coded signal separation method and the coded signal separation program of the present invention are characterized in that, in the coded signal separation method and the coded signal separation program according to claim 24, the extended layer The separation method is defined such that the number of coefficients of the extended hierarchical coefficient sequence in which separation pattern information is separated into the plurality of extended quantized coefficient sequences is different for each of the extended quantized coefficient sequences.
- the coefficient separating step converts the extended hierarchical coefficient sequence into a plurality of extended quantized coefficient sequences according to the extended hierarchical separation pattern information in which the separation method is defined such that the number of coefficients differs for each of the extended quantized coefficient sequences. It has a configuration characterized by separation.
- the coded signal separation method and the coded signal separation program of the present invention provide the coded signal separation method according to any one of claims 26 and 40 to 42. And coded signal separation program! And / or an enhanced layer separation pattern information transmitting step of transmitting the enhanced layer separation pattern information only to a specific receiver that has approved the decoding of a moving image without including the information in the enhancement coded signal. It has a configuration characterized by that.
- the coded signal synthesizing method and the coded signal synthesizing program of the present invention input a plurality of independent transmission coding signals obtained by coding a primary moving image composed of a plurality of pieces of image information.
- a basic encoding signal having a smaller code amount and a tertiary moving image closer to the primary moving image than the secondary moving image restored by the basic encoded signal are restored together with the basic encoded signal.
- the synthesizer synthesizing step includes: converting a basic coded signal from the basic coded signal to a base layer coefficient sequence of a base layer for restoring the secondary moving image; and An extended coding signal conversion step of converting an extended hierarchy coefficient sequence of an enhancement layer used when restoring the tertiary video, and a base layer coefficient sequence converted in the base coded signal conversion step,
- the reference order is a zigzag scan order
- the coefficient position index number is smaller than a predetermined basic extension separation coefficient number
- a coefficient sequence, and the extended hierarchical coefficient sequence converted in the extension code conversion signal conversion step is As a coefficient sequence whose order is equal to or greater than the basic extended separation coefficient number, the basic hierarchical coefficient sequence and the extended hierarchical coefficient sequence are combined to form a continuous sequence of third-order quantized coefficient values.
- the encoding program for synthesizing a signal according to the present invention is the program for synthesizing an encoding signal according to claim 44, wherein the basic coded signal conversion step converts the basic coded signal into the basic layer.
- a transforming step performing a reverse logarithmic transformation on each coefficient value of the basic quantized coefficient sequence, adding each coefficient value of the remainder coefficient sequence to the basic quantized coefficient sequence, and composing the basic hierarchical coefficient sequence.
- the basic quantization coefficient conversion step comprises: n kinds of eigenvalues; An array pattern inputting step of inputting an array pattern code table in which an array pattern of a sequence of (n + 1) types of values in which values other than eigenvalues are one type is input; A coefficient value array pattern decoding step of decoding an encoded value of the present quantized coefficient encoded signal into the eigenvalue and a value other than the eigenvalue according to the array pattern code table; and Decoding a coefficient value outside the eigenvalue for decoding each coefficient value by decoding from the basic quantized coefficient coded signal.
- the encoded signal combining program of the present invention is the encoded signal combining program according to any one of claims 44 to 46, wherein the extended encoded signal conversion step comprises: An extended quantization coefficient decoding step of decoding each of the plurality of extended coded signals into an extended quantization coefficient sequence, and the extended layer separation pattern information defining a separation method of separating the extended layer coefficient sequence into a plurality of segments. An extended quantized coefficient synthesizing step of synthesizing a plurality of extended quantized coefficient sequences into the extended hierarchical coefficient sequence.
- the code-shading signal synthesis program according to the present invention is the code-shading signal synthesis program according to claim 47, wherein the extended layer separation pattern information separates a coefficient of the extended layer coefficient sequence into a plurality of coefficients.
- Extended layer separation minimum pattern information that defines the number of application coefficients of the pattern to be applied and the separation pattern, and a separation minimum pattern application unit that defines the layer in the coding signal to which the extended layer separation minimum pattern is applied.
- the extended quantized coefficient synthesizing step includes, for each of the layers specified in the separated minimum pattern application unit information, the plurality of extended quantized coefficient sequences based on the extended hierarchical separated minimum pattern information. It has a configuration characterized by being combined into a column.
- the coded signal separation / combination method and the coded signal separation / combination program of the present invention provide a primary coded signal obtained by coding a primary moving image composed of a plurality of pieces of image information, A basic coded signal having a smaller code amount than the primary coded signal for restoring a secondary moving image that is a pseudo moving image of a moving image, and the primary coded signal from the secondary moving image restored by the basic coded signal.
- a plurality of extended coded signals which are restored by using a tertiary moving image close to a moving image together with the basic code signal, and are reconstructed and reconstructed, and a plurality of transmission coding signals transmitted over a network.
- a separator control step for controlling a separator to be converted into a signal, and inputting the plurality of transmission coded signals and selecting the transmission coded signal to be transferred. Selecting a transmission path selector for controlling the transmission path selector for transmitting the selected transmission coded signal, and inputting the plurality of transmission coded signals transmitted to the transmission path selector, A synthesizer control step of controlling a synthesizer for synthesizing a tertiary coded signal for restoring the next moving image.
- the separator control step is a separator input step of inputting the primary encoding signal, the primary encoding signal input in the separator input step, the basic encoded signal, the plurality of An extended coded signal, and a separator separating step of separating into the separated coded signal, the basic coded signal separated in the separator separating step, and a plurality of the extended coded signals.
- the combiner control step is a combiner input step of inputting the plurality of transmission coded signals, and from the plurality of transmission code input signals input in the combiner input step, the basic coded signal, Combining a plurality of extension coded signals, a combiner separation step for separating into the plurality of extension coded signals, the basic coded signal separated in the combiner separation step, and a plurality of the extension coded signals, A combiner combining step of generating a tertiary encoded signal for restoring a moving image; and a combiner output step of outputting the tertiary encoded signal combined in the combiner combining step.
- the separator separating step converts the primary coded signal into a primary quantized coefficient sequence which is a continuous sequence of primary quantized coefficient values constituting the primary moving image. And the step of converting the primary quantized coefficient sequence converted in the primary coefficient conversion step into a zigzag scan order in the reference order within the block, where the coefficient position index number is greater than a predetermined basic extension separation coefficient number.
- Serial combiner synthesis step from the basic coded signal, in the base layer coefficient sequence
- a coefficient sequence is smaller than the basic extended separation coefficient number! ⁇ A coefficient sequence, and the extended hierarchical coefficient sequence converted in the extended code conversion signal conversion step is a coefficient sequence whose order is equal to or larger than the basic extended separation coefficient number.
- a tertiary coefficient conversion step of converting the tertiary quantized coefficient sequence synthesized in the step into the tertiary code conversion signal.
- the coded signal separating / combining method of the present invention is the same as the coded signal separating / combining method of claim 29!
- the basic code generation signal generation step of the separator separation step is a basic quantization coefficient sequence obtained by rounding a coefficient value by logarithmic conversion for each coefficient value of the basic layer coefficient sequence.
- Basic quantization coefficient conversion for converting and separating into a coefficient sequence and a remainder coefficient sequence which is a sequence of a difference between each coefficient value of the base layer coefficient sequence and each coefficient value restored from the basic quantization coefficient.
- the basic quantizing signal conversion step of the synthesizer synthesizing step includes a basic quantizing coefficient for demultiplexing the basic coding signal into the basic quantized coefficient encoded signal and the residual coefficient encoded signal.
- Demultiplexing step performing a coefficient conversion on the basic quantized coefficient coded signal, a basic quantized coefficient conversion step of converting the basic quantized coefficient sequence into the basic quantized coefficient sequence, and performing a coefficient conversion on the residual coefficient coded signal,
- a basic quantization coefficient synthesizing step includes a basic quantizing coefficient for demultiplexing the basic coding signal into the basic quantized coefficient encoded signal and the residual coefficient encoded signal.
- the coded signal separation / combination method of the present invention is the same as the coded signal separation / combination method of claim 30, except that the basic quantization coefficient coding step iS of the separator control step is performed.
- the basic quantization coefficient conversion step of the synthesizer control step includes an arrangement pattern input step of inputting the arrangement pattern code table and an encoded value of the basic quantization coefficient encoded signal according to the arrangement pattern code table.
- a coefficient value array pattern decoding step for decoding the eigenvalue and a value other than the eigenvalue, and an eigenvalue outer relation for decoding a value other than the eigenvalue from the basic quantized coefficient coded signal to decode each coefficient value.
- a numerical value decoding step for decoding the eigenvalue and a value other than the eigenvalue, and an eigenvalue outer relation for decoding a value other than the eigenvalue from the basic quantized coefficient coded signal to decode each coefficient value.
- the coded signal separation / combination method of the present invention is the same as the coded signal separation / combination method of claim 29!
- the extended code generation signal generation step of the separator separation step includes a step of dividing the extended layer coefficient sequence into a plurality of extended quantums in accordance with extended layer separation pattern information defining a separation method for separating the extended layer coefficient sequence into a plurality.
- the extended coded signal conversion step of the combiner combining step includes an extended quantized coefficient decoding step of decoding the plurality of extended encoded signals into the extended quantized coefficient sequence, respectively, and the extended hierarchical separation pattern information.
- the extended layer separation pattern information separates a coefficient of the extended layer coefficient sequence into a plurality of coefficients. Stipulated the number of application coefficients and the pattern of separation Extended layer separation minimum pattern information, and a separation minimum pattern application unit that defines a layer in a coding signal to which the extension layer separation minimum pattern is applied, and an extended quantization coefficient separation step of the separator control step Separates the extended hierarchical coefficient sequence into a plurality of extended quantized coefficient sequences in accordance with the extended hierarchical separated minimum pattern information for each layer defined in the separated minimum pattern application unit information,
- the extended quantized coefficient synthesizing step includes the step of dividing the plurality of extended quantized coefficient sequences into the extended hierarchical coefficient sequence based on the extended hierarchical separated minimum pattern information for each layer defined in the separated minimum pattern application unit information. It has a configuration characterized by combining.
- the extended layer separation pattern information is separated into the plurality of extended quantization coefficient sequences.
- the separation method is defined such that the number of coefficients of the extended hierarchical coefficient sequence differs for each of the extended quantized coefficient sequences, and the extended quantization coefficient separating step of the separator control step includes the extended quantized coefficient sequence.
- the dani coefficient combining step Separating the extended hierarchical coefficient sequence into a plurality of extended quantized coefficient sequences in accordance with the extended hierarchical separation pattern information in which the number of coefficients is different for each of
- the dani coefficient combining step is characterized in that the plurality of extended quantized coefficient sequences are combined with the extended hierarchical coefficient sequence according to the extended hierarchical separation pattern information.
- the coded signal separation / combination method according to any one of claims 32 to 34 further includes the step of: An extended layer separation pattern information transmitting step of not including the layer separation pattern information in the extended coded signal but transmitting only to a specific receiver who has recognized the decoding of the moving image.
- An extended hierarchy separation pattern information receiving step for receiving hierarchy separation pattern information is provided.
- FIG. 1 is a block diagram of a coding / separation signal separation / synthesis system according to a first embodiment of the present invention.
- FIG. 2 is a block diagram showing a separator of the server of the present invention.
- FIG. 3 is a block diagram showing a separator in a method 1 (with a robustness transmission line).
- FIG. 4 is a block diagram showing a separator in a method 2 (without a robustness transmission line).
- FIG. 5 is a block diagram showing a transmission path selector of the router of the present invention.
- FIG. 6 is a diagram illustrating an example of transmission path selection by a transmission path selector of a router.
- FIG. 7 is a block diagram showing a combiner of the receiver of the present invention.
- FIG. 8 is a block diagram showing a combiner in scheme 1.
- FIG. 9 is a block diagram showing a combiner in scheme 2.
- FIG. 10 is a diagram showing a bit stream format of an extension stream.
- FIG. 11 is a diagram showing contents of user data defined in a slice layer.
- FIG. 12 is a diagram showing the contents of MB attribute information defined in the MB layer.
- Fig. 13 is a block diagram showing a separation unit of the separator.
- FIG. 14 is a diagram showing the principle of coefficient separation in a block.
- FIG. 15 is a diagram showing the principle of extended layer separation (rate control) in a block.
- FIG. 16 is a diagram showing the principle of extended hierarchy separation (scramble) in a block.
- FIG. 17 is a diagram showing an array code table (inter) when the number of coefficients is three.
- FIG. 18 is a diagram illustrating an array code table (intra) when the number of coefficients is three.
- FIG. 19 is a block diagram showing a synthesis unit of the synthesizer.
- FIG. 21 is a diagram illustrating the principle of intra-block extended hierarchical composition (rate control).
- FIG. 21 is a diagram illustrating the principle of intra-block extended hierarchical synthesis (scramble).
- FIG. 22 is a diagram illustrating the principle of intra-block coefficient synthesis.
- FIG. 23 is a block diagram showing a separation unit that multiplexes a basic quantization coefficient sequence C and a prediction error coefficient sequence D to generate a basic stream B.
- FIG. 24 is a block diagram showing a synthesis unit that inputs a basic stream B obtained by multiplexing a basic quantization coefficient sequence C and a prediction error coefficient sequence D.
- FIG. 25 is a schematic block diagram of a conventional transcoder.
- FIG. 26 is a flowchart showing a rate control process of TM5 of MPEG-2 in a conventional transcoder.
- FIG. 27 is a schematic block diagram of a conventional transcoder.
- FIG. 28 is a flowchart showing processing of a conventional transcoder.
- FIG. 29 is a schematic block diagram of a conventional transcoder.
- FIG. 30 is a flowchart showing processing of a conventional transcoder.
- the encoding / separation signal separation / synthesis system of the present invention provides a transcoder having a stream separation function in a server, which generates an independent separated stream, and a router having a band selection function, in which a transmission path corresponding to a network state is provided.
- a transcoder having a stream separation function in a server, which generates an independent separated stream
- a router having a band selection function, in which a transmission path corresponding to a network state is provided.
- FIG. 1 An encoded signal separation / combination system according to the first embodiment of the present invention is shown in FIG. 1 and described.
- the coded signal separation / combination system includes a server 1000 for transmitting a video, a router 2000 for selecting a transmission path on a network, a receiver 3000a for receiving and reproducing a video, 3000b- ⁇ ⁇ , 3000 ⁇ .
- Recino 3000a, 3000b ⁇ ⁇ ⁇ 3 OOOn can be any number.
- a plurality of routers 2000 may be provided.
- the stream input to the server 1000 is a stream generated by a standard encoder.
- the stream captured by the camera 600 is encoded by the encoder 700.
- a stream of content stored in the content storage 800 is encoded by the Sano 1000.
- the Sano 1000 includes a separator 1010, and the router 2000 includes a transmission path selector 2010.
- Recino 3000a, 3000b ⁇ ⁇ ⁇ , 3000 ⁇ are provided with combiners 3010a, 3010b ⁇ ⁇ ⁇ , 3010 ⁇ and decoders 3030a, 3030b ⁇ ⁇ ⁇ 3030 ⁇ respectively, and Recino 3000a, 3000b, ⁇ ⁇ ⁇ Decoders 3030a, 3030b, ⁇ , 3030 ⁇ that perform playback with 3000 ⁇ are used.
- Recino 3000a, 3000b ⁇ , 3000 ⁇ , combiners 3010a, 3010b ⁇ , 3010n and decoders 3030a, 3030b- ⁇ , 3030 ⁇ are the same, Let's describe one of them as receiver 3000, combiner 3010 and decoder 3030.
- the separator 1010 receives a stream already encoded and stored in an archive or a stream photographed and encoded by a camera as an input, separates the stream into independent streams, and outputs the separated streams. Is what you do.
- Separator 1010 includes a separating unit (separateUlOO) as a separator input unit and a separator separating unit, and a multiplexing unit (MUXU600) as a separator multiplexing unit and a separator output unit.
- a buffer or the like as separator input means may be provided before the separator separating means
- a buffer or the like as separator output means may be provided after the separator multiplexing means.
- the method of generating a separated stream by the separator 1010 includes the following two methods depending on the processing in the multiplexing unit 1600, and the processing diagrams of each method are shown in Figs. 3 and 4, respectively.
- Video is transmitted over a network having a robustness transmission path.
- Method (2) Video is transmitted over a network without a robustness transmission path.
- the separating unit (l lOO) converts the input coded stream into one elementary stream B.
- the multiplexing unit (MUX) 1600 converts the multiplexing of the basic stream B output from the demultiplexing unit 1100 and the extended stream E (m) (0 ⁇ m ⁇ M-1) into the characteristics of the transmission path. Perform according to.
- the method includes the following method (1) and method (2).
- St (B) and L streams St (l) (0 ⁇ l ⁇ L-1) are output as separated streams.
- the basic stream B is sent as a separated stream St (B) to the mouth bustness transmission line where no transmission error occurs, and the extended streams E (m) (0 ⁇ m ⁇ M-1) are separated streams St (l), respectively. (0 ⁇ 1 ⁇ L-1).
- St (l) is shown in equation (1).
- L streams St (l) (0 ⁇ l ⁇ L-1) are output.
- Elementary stream B is copied and multiplexed into all extension streams E (m).
- the separation stream St (l) can be represented as in Expression (3).
- Multiplex ['] means multiplexing.
- the router 2000 and the transmission path selector 2010 receive the L separated streams St (l) output from the server 1000 as input, and transmit by selecting a transmission path (on / off of the transmission path) according to the network condition. Control the stream rate to achieve video scalability.
- Rate [St (l)] Assuming that the target total rate of the output stream of the router 2000 is Rtarget, the transmission line switch of St1 where Rate [St out] ⁇ Rtarget is turned on.
- FIG. 6 shows a specific example of transmission path selection by the transmission path selector.
- the synthesizer 3010 synthesizes the input L ′ streams St (l) and reproduces an image. Video quality depends only on the total rate of the received stream, not on the type of received stream. The total rate of the stream to be received is determined by the transmission path selection in the router 2000.
- a block diagram of the synthesizer 3010 of the receiver 3000 is shown and described in FIG.
- the combiner 3010 is provided as combiner input means and combiner separation means.
- a merge controller 3020 is provided. Further, the composition controller 3020 may have a configuration included in the composition device 3010.
- FIGS. 8 and 9 show block diagrams of a combiner 3010 corresponding to the above methods (1) and (2), respectively.
- the demultiplexing unit (DEMUX) 3100 divides the input L, book stream St (l) into ⁇ basic streams ⁇ and ⁇ , book extended streams E (m).
- ⁇ indicates the total number of basic streams ⁇ multiplexed in all the separated streams St (l) input to the demultiplexing unit 3100, and corresponds to the above methods (1) and (2). The number is different, as shown in equation (5).
- the basic stream selector (B-Selector) 3200 under the control of the synthesis controller 3020, selects and outputs one stream with little bit error from the input ⁇ B streams. At this time, -1) the basic stream ⁇ is discarded.
- the combining unit (merge) 3300 receives control from the combining controller 3020, combines the input basic stream B and the extended stream E (m) (0 ⁇ m ⁇ M, 1 1) and combines them. Output stream. Therefore, not all of the extension streams E (m) input to the combining unit 3300 are combined.
- the merge controller 3020 receives the error detector power error correction information on the network, and the error information detected in the processing in the demultiplexer 3100 and the synthesizer 3300.
- the information is processed by the basic stream selector 3200, and the information on the extended stream E (m) is controlled by the synthesis unit 3300, and the stream for which error correction is impossible is discarded by the synthesis controller 3020. I do. [0144] (Code I Dani Syntax)
- the form of the bit stream in the present invention is such that an MPEG-2 bit stream (MPEG-2 before separation) conforming to the main profile is input and separated into a basic stream and M extension streams.
- MPEG-2 is explained as an example, but other DCT video coding schemes with motion compensation prediction (ITU-TH.261, ITU-TH.263, ISO / IEC 14496-2 (commonly known as The same configuration can be applied to the DCT coefficient coding part for “MPEG-4”), ITU-T II.264, etc.).
- the basic stream is output as an MPEG-2 bit stream with a reduced bit rate.
- the extension stream includes a prediction error stream composed of prediction error information generated in the requantization processing at the time of rate reduction and a difference bit stream composed of difference information before and after the rate reduction.
- the bit stream format of the extension stream is based on the bit stream format of the MPEG-2 syntax, and includes a sequence layer, a GOP layer, a picture layer, a slice layer, and a macro block (hereinafter MB). It has a hierarchical structure in which layers and blocks also have strength.
- the extension stream starts with the sequence header power and continues to the picture layer for the number of pictures.
- the picture layer data also comprises a picture header and picture data power.
- the picture data is composed of a plurality of slice layer data, and the slice layer data is composed of a slice header followed by the data of the MB layer.
- the bit fields of the sequence header, GOP header, picture header, and slice header are equal to MPEG-2.
- user data is defined between the slice header and the MB data, and the user data describes attributes unique to the present scheme in the slice layer shown in FIG.
- the MB layer data is composed of MB attribute information and coefficient information.
- Figure 12 shows the MB attribute information.
- the user data in the slice layer is used for separation and synthesis in the slice layer, and the information contained in the MB attribute information is stored in the MB layer. Separation in 'synthesis.
- the extended layer separation minimum pattern application unit and the extended layer separation minimum pattern will be described later.
- the sequence header, picture header, and slice header are used to synchronize with the MPEG-2 bit stream output in GOP units, picture units, and slice units, respectively.
- the slice unit which is the minimum unit of synchronization, synchronization is performed so that SSC (Slice Start Code) in the slice header becomes equal.
- SSC Selice Start Code
- the SSC in the slice header is a synchronization code indicating the start of the slice layer, and the last byte of the code indicates the vertical position of the slice.
- the block layer data is re-quantized, and the difference information of the change in the quantized coefficient value before and after the re-quantization is encoded.
- the coefficient information before requantization can be completely restored.
- the requantized output coefficients are coded as a base layer and become a base stream B.
- the extended layer is separated into N extended quantized coefficient sequences in order.
- the prediction error information generated by the requantization becomes a prediction error coefficient sequence, and the prediction error coefficient sequence and the extended quantization coefficient sequence are respectively encoded to generate M extended streams E (m). At this time, the rates of the M extended streams E (m) can be generated at almost the same rate.
- FIG. 13 shows a block diagram of a separating unit (separate) lOOa, which is one embodiment of the separating unit (separate) lOOa.
- Separating section 1100a includes a coefficient information separating section 1260, a variable length decoder (VLD) 1110, a run-level coefficient changing section 1120 as a primary coefficient converting section, and a basic extended layer separating section 1130 as a basic extended layer separating section.
- VLD variable length decoder
- Basic coded signal generation means, extended coded signal generation means, and basic quantization coefficient converter 1140 as basic quantization coefficient conversion means, extended coded signal generation means, extended quantization coefficient separation means, and extension Hierarchical separation pattern information
- Extended quantization coefficient separator 1150 as transmission means, basic rescanner 1160, prediction error Difference rescanner 1170, extended rescanner 1181, 1182, 1183, basic quantization coefficient encoding means, array pattern input means, and basic variable length encoder (VLC basic encoder) as coefficient value array pattern encoding means 1210)
- Prediction error as remainder coefficient coding means
- Coefficient information separation section 1260 of separation section 1100a separates the input coded stream (primary coded signal) into coefficient information and other than coefficient information, and outputs the separated stream.
- the coefficient information is a quantized coefficient code subjected to Huffman coding by two-dimensional run length
- the other than the coefficient information is MB layer data and intra MB DC coefficient code.
- variable length decoder (VLD) 1110 performs variable length decoding on the coefficient information separated by the coefficient information separating unit 1260, and converts the decoded information into run level information.
- the run level coefficient change 120 converts the run level information decoded by the VLD 1110 into a one-dimensional quantized coefficient sequence X.
- the base enhancement layer separator 1130 separates the quantized coefficient sequence X converted into the run-level coefficient modification 1120 into a coefficient sequence B of the base layer and a coefficient sequence E of the enhancement layer.
- i represents a coefficient position index number when the reference order in the block is the zigzag scan order.
- the base layer is a coefficient sequence whose index number is smaller than the basic extended separation coefficient number B among the quantized coefficients, and expands the coefficients having the basic extended separation coefficient number B or higher. It is a coefficient that constitutes the extension hierarchy.
- FIG. 14 shows an example of separation processing of the primary quantization coefficient sequence X in the basic enhancement layer separator 1130.
- the separation process is performed by setting ⁇ to “3”.
- the first-order quantization coefficient sequence ⁇ (8, 2, 0, 3, 1, —3, 0, 6, 5, —1, 0, 1, 1,-2, 0, 0, 0, 0, 4, 1, 0, 0, 2 ⁇ is separated into a base layer coefficient sequence ⁇ and an extended layer coefficient sequence ⁇ ,
- E ⁇ 3, 1,-3, 0, 6, 5, —1, 0, 1, 1, —2, 0, 0, 0, 0, —4, 1, 0, 0, 2 ⁇ It becomes.
- the basic quantization coefficient transformation 1140 inputs the base layer coefficient sequence ⁇ separated by the base enhancement layer separator 1130, and performs conversion for code amount reduction, generally quantization. Then, it is converted into a basic quantized coefficient sequence C, and a residual coefficient sequence (prediction error coefficient sequence) D which is difference information between the basic hierarchical coefficient sequence ⁇ and the basic quantized coefficient sequence C.
- the basic quantized coefficient sequence C is a sequence of basic quantized coefficients obtained by performing a conversion for code amount reduction on each coefficient value of the basic hierarchical coefficient sequence #.
- the remainder coefficient sequence D is a remainder coefficient value which is a difference between each coefficient value of the above-described basic hierarchy coefficient sequence ⁇ and each coefficient value restored from the basic quantization coefficient of the basic quantization coefficient sequence C. It is a column of (prediction error coefficient value).
- the generation of the basic quantized coefficient sequence C in the basic quantized coefficient converter 1140 is performed by taking the logarithm of base 2 for the absolute value of each coefficient value of the basic hierarchical coefficient sequence B.
- the integer value of the value obtained by the logarithmic conversion process as each coefficient value of the basic quantized coefficient sequence C, the expression range of the coefficient value can be reduced and the code amount can be suppressed.
- the base layer coefficient sequence B input to the base quantization coefficient converter 1140 is defined as the following equation (8). Justify.
- a basic quantization coefficient sequence C composed of coefficients obtained by taking logarithms with respect to the base layer coefficient sequence B, and a residual coefficient sequence D are defined as the following equation (9).
- foncl is a logarithm whose base is 2, and is expressed by the following equation (11).
- [ ⁇ ] is converted to an integer value.
- Equation (12) From Equation (9), the remainder coefficient sequence is obtained as in the following Equation (12).
- fonc2 is an inverse function of foncl, and is represented by the following equation (13).
- the extended quantized coefficient separator 1150 inputs the extended layer coefficient sequence E separated by the basic extended layer separator 1130, and also inputs the extended layer separation pattern information from the outside, and According to the hierarchical separation pattern information, the extended hierarchical coefficient sequence E is separated into a plurality of extended quantized coefficient sequences F, F 1 ′′ ′ F N — 1 .
- the extended layer separation pattern information is information that defines a separation method for separating the extended layer coefficient sequence E into a plurality.
- the extended layer separation pattern information is It has a minimum pattern application unit and extended layer separation minimum pattern information.
- the extension layer separation minimum pattern application unit and the extension layer separation minimum pattern information are stored in the user data in the slice layer and the MB attribute information in the MB layer as described in the bit stream format of the extension stream.
- the extension layer separation minimum pattern application unit defines a layer in a coded signal to which the enhancement layer separation minimum pattern is applied.
- the extended layer separation minimum pattern information defines the number of applied coefficients of the pattern for separating the coefficients of the extended layer coefficient sequence E into a plurality and the separation pattern.
- each coefficient value of the extended hierarchy coefficient sequence E a plurality of extended hierarchy coefficient sequence (F, F 1 ⁇ ⁇ ⁇ , F n ' ⁇ ⁇ , F N_ 1) How The allocation order is determined in accordance with the order of the separation layer number G of the expansion layer.
- the group is composed of J sets, and one group has the separation layer number G of the expansion layer of H (0 ⁇ j ⁇ J— 1).
- the state of the separation destination layer number G of the extended layer is shown below.
- the minimum information on the extended hierarchy separation minimum pattern is represented by ⁇ J of the set, separation destination layer number Gj, and the number of elements Hj in the set.
- the separation method may be such that the number of coefficients of the extended layer coefficient sequence E is different for each of the extended quantized coefficient sequences (F °, F 1 ′′ ′ F N — 1 ). By defining this, it is possible to make a difference in the code amount of the extended coding signal generated for each extended quantization coefficient sequence, and to perform rate control at the time of transmission.
- the enhanced layer separation pattern information is not included in the extension stream, and is separately transmitted by the server 1000 only to a specific receiver that has approved the decoding of the moving image, so that the scrambling key of the enhanced layer separation pattern information can be obtained. Can be used as
- the extended quantized coefficient sequence F n (0 ⁇ n ⁇ N— 1) Separation is performed in accordance with extended layer separation pattern information. Based on this extended layer separation pattern information, separation processing is performed according to the minimum expanded layer separation pattern for each slice or block.
- the extended hierarchical coefficient sequence E input to the extended quantized coefficient separator 1150 is defined as the following equation (15).
- F 3 ⁇ -1, 0, 1, 1, 1, 0, 0, 2 ⁇
- the number of extended quantized coefficient sequences N 4, the externally applied extended hierarchical separation minimum pattern application unit is a block, and the extended hierarchical coefficient sequence E Fig. 16 shows the state of the separation processing (extended hierarchy separation processing within a block).
- F 3 ⁇ -1, 0, 1, 0, 2 ⁇
- the basic quantized coefficient sequence C generated by the basic quantized coefficient converter 1140 is re-scanned and subjected to variable length coding to generate a basic quantized coefficient coded signal.
- variable length encoding process of the basic quantized coefficient sequence C in the basic variable length encoder (VLC basic) 1210 will be described in detail.
- variable length encoding of the basic quantized coefficient sequence C in the basic variable length encoder (VLC basic) 1210 a value representing the coefficient value of the basic quantized coefficient sequence C in a pattern is used.
- VLC basic basic variable length encoder
- the coefficient values of j8 basic quantization coefficients c are represented by three patterns of “0”, “1”, and “2 or more”.
- Figs. 17 and 18 show arrangement code tables in the case where the number of coefficients is three.
- the prediction error rescanner 1170 and the prediction error variable length encoder (VLC prediction error) 1220 rescan the remainder coefficient sequence D generated by the basic quantization coefficient converter 1140.
- a run-level coefficient value of the remainder coefficient is generated, the run-level coefficient value of the remainder coefficient is variable-length coded, a remainder coefficient coded signal is generated, and output as an extended code signal (E (0)). is there.
- the extended rescanning units 1181, 1182, and 1183 generate the extended quantized coefficient sequences generated by the extended quantized coefficient separator 1150, respectively.
- F N_1 is re-scanned to generate a run-level coefficient value of the extended quantization coefficient.
- Extended variable length encoder ( ⁇ 1 ⁇ ) 1231, ( VLC 'F 1) 1232, (VLC' F N_1) 1233 is extended quantization generated for each extended rescan unit 1181, 1182, 1183
- the run-level coefficient values of the coefficients are variable-length coded to generate extended coded signals E (1), E (2), and E (M-1).
- the coefficient information multiplexing unit 1270a receives the basic quantized coefficient coded signal generated by the VLC basic 1210 and data other than the coefficient information separated by the coefficient information separating unit 1260, and multiplexes the signals. , To generate the basic coded signal B.
- the coded stream (primary coded signal) is input, and the basic coded signal B, the extended coded signal ⁇ (0), ⁇ (1), ⁇ (2) ⁇ ⁇ ⁇ , E (M— 1
- synthesizing unit (merge) 3300a of the synthesizer 3010x will be described.
- FIG. 19 is a block diagram of the synthesizing unit (merge) 3300a.
- Combining section 3300a includes coefficient information separating section 3450a, basic variable length decoder (VLD basic) 3310 as coefficient value array pattern decoding means, prediction error variable length decoder (VLD prediction error) 33 20, extended quantization Extended variable length decoder as coefficient decoding means (VLD'E (0)) 3331, (VLD'E (1)) 3332, (VLD E (N, -1)) 3333, basic code conversion Means and basic coefficient conversion 3340 as conversion means for basic quantization coefficient, prediction error coefficient conversion 3350 as conversion means for residual coefficient, expansion coefficient conversion 33 as conversion means for extended coding signal 61, 3362, 3363, basic quantized coefficient synthesizer 3370 as basic quantized coefficient synthesizing means, extended quantized coefficient synthesizer 3380 as extended quantized coefficient synthesizing means, basic extended hierarchical synthesizing as basic expanded layer synthesizing means 3390, a run-level coefficient converter 3410, a variable length coder (VLC) 3420, and a coefficient information multiplexing unit
- the coefficient information separation unit 3450a of the synthesis unit 3300a separates the input basic coded signal B * into components other than the coefficient information and the coefficient information and outputs the result.
- the coefficient information is a quantized coefficient code that is Huffman-coded by two-dimensional run length
- the other than the coefficient information is the MB layer data and the intra MB DC coefficient code.
- VLD basic variable-length decoder
- basic coefficient translator 3340 perform variable-length decoding on the coefficient information separated by the coefficient information separation unit 3450a, and convert it to a basic quantized coefficient sequence C *. Things.
- VLD basic basic variable length decoder
- VLC basic basic coefficient conversion 3340 corresponding to the basic variable length encoder
- the prediction error variable length decoder (VLD prediction error) 3320 and the prediction error coefficient converter 3350 perform variable length decoding on the input extended coded signal E * (0), and calculate the remainder coefficient ( This is converted into run-level information of the prediction error coefficient) and further converted to a residual coefficient sequence (prediction error coefficient sequence).
- the extended coefficient variations ⁇ 3361, 3362, and 3363 are extended quantized coefficients converted to VLD'E (O) 3331, VLD-E (1) 3332, and VLD'E (N, -1) 3333, respectively. run level information, and extended quantization coefficient sequences F * °, and converts to F * ⁇ F * N '_1 .
- the basic quantized coefficient synthesizer 3370 includes a basic quantized coefficient sequence C * converted by the basic coefficient converter 3340 and a residual coefficient sequence (predicted error coefficient sequence) D converted by the prediction error coefficient converter 3350. And * to generate a base layer coefficient sequence B *.
- the above-mentioned basic quantization coefficient converter 1140 performs logarithmic conversion with the base being 2 for each coefficient value of the basic hierarchical coefficient sequence B, and converts the integer value of the value obtained by the logarithmic conversion process into a base number.
- Each coefficient value of the quantization coefficient sequence C is used.
- the difference between each coefficient value of the basic hierarchical coefficient sequence B and the value obtained by the inverse logarithmic conversion of each coefficient value of the basic quantization coefficient sequence C is used as a residual coefficient sequence (prediction error coefficient sequence) D as each coefficient value. Puru.
- the basic quantized coefficient sequence C * and the residual coefficient sequence (prediction error coefficient sequence) D * input to the basic quantized coefficient synthesizer 3370 are defined as the following equation (22).
- the base layer coefficient sequence B * generated by the base quantization coefficient synthesizer 3370 is defined as the following equation (23).
- the base layer coefficient sequence B * is generated by adding D * to the value obtained by performing lunc3 processing on C *. Is done.
- extended quantized coefficient synthesizer 3380 generates VLD'E (0) 3331, VLDE (1) 3332
- VLD'E (N'- 1) 3333 is converted into the extended quantization coefficient sequences F * °, inputs the F * 1 'F * N' _1, external mosquito ⁇ et extended hierarchy separating pattern information type, the extended quantization coefficient sequences F * in accordance with the extended hierarchy separating pattern information, F "''', F * N' is intended to be synthesized _1 enhancement layer coefficient sequence E *.
- the extended hierarchical coefficient sequence E is divided into a plurality of extended quantized coefficient sequences F °,? 1 ..., F N_1 . Therefore, in the extended quantization coefficient synthesizer 3380, a plurality of extended quantization coefficient sequences F * °, F * 1 - • ⁇ , F * a N '_1, extended hierarchy coefficient sequence E in accordance extended hierarchy separating pattern information Combine with *.
- the extended quantized coefficient synthesizer 3380 obtains the index number n of the extended quantized coefficient sequence from the extended stream number m specified by the user data in the slice layer, and uses the extended quantized coefficient sequence F * n using n. (0 ⁇ n ⁇ N'-1) is defined as the following equation (26).
- the extended quantized coefficient sequence F * n is synthesized according to the extended layer separation pattern information to generate an extended layer coefficient sequence E *.
- the extended hierarchical coefficient sequence E * is generated as in the following equation (27).
- the extension layer N 4 and the minimum separation pattern of the expansion layer shown in the above equation (16) is defined, and the minimum separation pattern application unit is a block unit.
- F * 3 ⁇ — 1, 0, 1, 1, 1, 0, 0, 2 ⁇
- FIG. 20 shows the state of the combining process of the extended hierarchical coefficient sequence E * (intra-block extended hierarchical combining process).
- the extension layer N 4 and the minimum separation pattern of the expansion layer shown in the above equation (19) is defined, and the minimum separation pattern application unit is a block unit.
- the extended quantized coefficient sequence F * ° By combining F * 2 and F * 3 , an extended hierarchical coefficient sequence E * can be generated as shown in the following equation (31).
- F * 3 ⁇ -1, 0, 1, 0, 2 ⁇
- FIG. 21 shows the state of the process of synthesizing the extended hierarchical coefficient sequence E * (intra-block extended hierarchical synthesizing process).
- the basic enhancement layer combiner 3390 includes the base layer coefficient sequence B * generated by the basic quantization coefficient combiner 3370 and the extension layer coefficient sequence E generated by the extension quantization coefficient combiner 3380. And * are combined to generate a combined quantized coefficient sequence (third-order quantized coefficient sequence) X *.
- the base layer coefficient sequence B * and the extended layer coefficient sequence E * that are input to the basic enhancement layer synthesizer 3390 are defined as the following equation (32).
- the synthetic quantized coefficient sequence X * generated by the basic enhancement layer synthesizer 3390 is defined as the following equation (33).
- the coefficients of the base layer coefficient sequence B * are inserted into the corresponding positions in the block according to the index numbers.
- the expanded hierarchical coefficient sequence E * is inserted into the corresponding positions in the block according to the index numbers.
- X * ⁇ 8, 2, 0, 3, 1, —3, 0, 6, 5, —1, 0, 1, 1, —2, 0, 0, 0, 0, —4, 1 , 0, 0, 2 ⁇
- FIG. 22 shows an example of the synthesis process (intra-block coefficient synthesis process) of the synthesized quantized coefficient sequence X *.
- the run-level coefficient transformation 3410 was generated by the basic extension layer synthesizer 3390.
- the coefficient conversion is performed on the composite quantized coefficient sequence (third-order quantized coefficient sequence) X * to convert it to run-level information.
- variable-length encoder (VLC) 3420 performs variable-length coding on the run-level information converted to the run-level coefficient change 3410, and generates a coding signal for the coefficient information.
- the coefficient information multiplexing unit 3460 receives the coded signal of the coefficient information generated by the VLC 3420 and the data other than the coefficient information separated by the coefficient information separating unit 3450a, and multiplexes the combined stream. (Tertiary coded signal).
- this combined stream (tertiary encoded signal) is a normal MPEG-2 stream, it can be decoded and reproduced by, for example, a decoder 3030a that is a general decoder.
- the present coded signal separation / combination system includes a base coded signal B, a plurality of extended coded signals E (m), and a coded stream obtained by the separator 1010 of the server 1000, depending on the coefficient position. , Multiplexed and transmitted arbitrarily in combination, and separated by the combiner 3010 of the Recino 3000 into the basic coded signal B and the plurality of extended coded signals E (m), and then combined. Scalable transmission of video can be realized, and the separation process between the base layer coefficient sequence and the extended layer coefficient sequence is a simple process that separates them in the order of the coefficients. The processing speed can be increased and the price of parts can be reduced.
- the remainder coefficient sequence (prediction error coefficient sequence) D is multiplexed with the basic quantization coefficient sequence C to generate the basic coded signal B, thereby performing the separation.
- a process for making the rates of the stream St (l) (0 ⁇ l ⁇ L-1) approximately equal will be described.
- FIG. 23 shows the separation unit that generates the coded signal B.
- demultiplexing section 1100b has the same configuration as demultiplexing section 1100a described in the above embodiment (see FIG. 13), and further has a function as a basic quantization coefficient multiplexing means.
- a basic quantized coefficient multiplexer 1240 is provided.
- the basic quantization coefficient multiplexer 1240 receives the basic quantization coefficient sequence C output from the VLC basic 1210 and the remainder coefficient sequence (prediction error coefficient sequence) D output from the VLC prediction error 1220. Then, the remainder coefficient sequence (prediction error coefficient sequence) D is multiplexed with the basic quantization coefficient sequence C to generate and output a basic coded signal B.
- the separation unit 1100b can multiplex the remainder coefficient sequence (prediction error coefficient sequence) D with the basic quantization coefficient sequence C to generate the basic coded signal B.
- combining section 3300b has the same configuration as combining section 3300a described in the above embodiment (see Fig. 19), and further has a function as a basic quantization coefficient demultiplexing unit.
- a basic quantization coefficient separator 3430 is provided.
- the basic quantized coefficient separator 3430 separates the input coefficient information of the basic coded signal B * into a basic quantized coefficient sequence C * and a residual coefficient sequence (prediction error coefficient sequence) D *. Then, they are output to the VLD basic 3310 and the VLD prediction error 3320, respectively.
- the combining section 3300b allows the basic quantized coefficient sequence C * and the residual coefficient sequence
- Prediction error coefficient sequence Even if a basic coded signal B * combined with D * is input, it is separated into a basic quantization coefficient sequence C * and a residual coefficient sequence (prediction error coefficient sequence) D *. , And a base layer coefficient sequence, and combined with an extended layer coefficient sequence E * synthesized from a plurality of extended coded signals E * (m) to generate a combined stream (tertiary coded signal).
- the coding / separation signal separation / synthesis system has an effect that the stream division processing can be simplified and the scalable transmission of video can be realized. It is useful as a server or transcoder that performs / v: / O l / J900sooifcl £ 9 £ 9960S00iAV
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Databases & Information Systems (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2004100369A JP2005286863A (ja) | 2004-03-30 | 2004-03-30 | 符号化信号分離装置、符号化信号合成装置および符号化信号分離合成システム |
JP2004-100369 | 2004-03-30 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2005096636A1 true WO2005096636A1 (ja) | 2005-10-13 |
Family
ID=35064158
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2005/006171 WO2005096636A1 (ja) | 2004-03-30 | 2005-03-30 | 符号化信号分離装置、符号化信号合成装置および符号化信号分離合成システム |
Country Status (2)
Country | Link |
---|---|
JP (1) | JP2005286863A (ja) |
WO (1) | WO2005096636A1 (ja) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101375663B1 (ko) * | 2007-12-06 | 2014-04-03 | 삼성전자주식회사 | 영상을 계층적으로 부호화/복호화하는 방법 및 장치 |
JP4844595B2 (ja) * | 2008-06-26 | 2011-12-28 | 日本ビクター株式会社 | 階層符号化装置、非階層符号化変換装置、階層符号化プログラム、および非階層符号化変換プログラム |
CN101964188B (zh) | 2010-04-09 | 2012-09-05 | 华为技术有限公司 | 语音信号编码、解码方法、装置及编解码系统 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0344187A (ja) * | 1989-07-11 | 1991-02-26 | Nec Corp | 画像信号の符号化装置 |
JPH05336505A (ja) * | 1992-06-02 | 1993-12-17 | Hitachi Ltd | 画像の階層符号化方式 |
JP2000253375A (ja) * | 1999-03-01 | 2000-09-14 | Sharp Corp | ディジタル画像暗号化方法、画像暗号化システム及び暗号化画像解読システム |
-
2004
- 2004-03-30 JP JP2004100369A patent/JP2005286863A/ja active Pending
-
2005
- 2005-03-30 WO PCT/JP2005/006171 patent/WO2005096636A1/ja active Application Filing
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0344187A (ja) * | 1989-07-11 | 1991-02-26 | Nec Corp | 画像信号の符号化装置 |
JPH05336505A (ja) * | 1992-06-02 | 1993-12-17 | Hitachi Ltd | 画像の階層符号化方式 |
JP2000253375A (ja) * | 1999-03-01 | 2000-09-14 | Sharp Corp | ディジタル画像暗号化方法、画像暗号化システム及び暗号化画像解読システム |
Non-Patent Citations (1)
Title |
---|
WAKUI . ET AL: "A Study on a Scalable Video Coding Scheme with Independent Layers by Transcoding.", IPSJ SIG TECHNICAL REPORT., vol. 2004, no. 25, 4 March 2004 (2004-03-04), pages 25 - 30, XP002990169 * |
Also Published As
Publication number | Publication date |
---|---|
JP2005286863A (ja) | 2005-10-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP3561485B2 (ja) | 符号化信号分離・合成装置、差分符号化信号生成装置、符号化信号分離・合成方法、差分符号化信号生成方法、符号化信号分離・合成プログラムを記録した媒体および差分符号化信号生成プログラムを記録した媒体 | |
JP2006237656A (ja) | 符号化信号分離・合成装置、差分符号化信号生成装置、差分符号化信号抽出装置、符号化信号分離・合成方法、符号化信号分離・合成プログラム | |
JPH0795090A (ja) | トランスコーディング装置 | |
JP2005507589A5 (ja) | ||
TW200400767A (en) | Method and apparatus for transcoding compressed video bitstreams | |
KR20040091686A (ko) | 더 높은 질의 참조 프레임을 사용하는 fgst 코딩 방법 | |
WO1994024822A1 (en) | Method and apparatus for coding image, and image recording medium | |
KR100952185B1 (ko) | 순방향 에러 정정 코드를 이용하여 비디오의 드리프트 없는 단편적인 다중 설명 채널 코딩을 위한 시스템 및 방법 | |
JP2007266749A (ja) | 符号化方法 | |
WO2005046244A1 (ja) | 符号化信号分離装置、符号化信号合成装置および符号化信号分離合成システムならびにそれらの方法 | |
JP2005295505A (ja) | 動画像符号化装置及び動画像復号装置及びそれらの制御方法、並びに、コンピュータプログラム及びコンピュータ可読記憶媒体 | |
JP2000511722A (ja) | 信号のビットレートの変更に関する改良 | |
WO2005096636A1 (ja) | 符号化信号分離装置、符号化信号合成装置および符号化信号分離合成システム | |
WO1999021367A1 (fr) | Codeur et decodeur d'image | |
US20070223573A1 (en) | Method and apparatus for encoding/decoding a first frame sequence layer based on a second frame sequence layer | |
US20070242747A1 (en) | Method and apparatus for encoding/decoding a first frame sequence layer based on a second frame sequence layer | |
US20070280354A1 (en) | Method and apparatus for encoding/decoding a first frame sequence layer based on a second frame sequence layer | |
JP2001145102A (ja) | 動画像圧縮符号化信号変換方法、装置および変換プログラムを記録した媒体 | |
JP3576660B2 (ja) | 画像符号化装置および画像復号化装置 | |
WO2005096635A1 (ja) | 符号化信号蓄積配信システム、符号化信号選択転送装置、符号化信号分離送信装置、符号化信号受信合成装置 | |
WO1998053613A1 (en) | Apparatus, method and computer readable medium for scalable coding of video information | |
JP2004056616A (ja) | 符号化信号復号装置、符号化信号復号方法および符号化信号復号プログラム | |
JP2001078193A (ja) | 動画像圧縮符号化信号変換方法および装置 | |
JP2001169283A (ja) | 動画像圧縮符号化信号変換方法、装置および変換プログラムを記録した媒体 | |
JP2004200760A (ja) | 画像処理方法および装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NA NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SM SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): BW GH GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LT LU MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWW | Wipo information: withdrawn in national office |
Country of ref document: DE |
|
122 | Ep: pct application non-entry in european phase |