WO2012122798A1 - 编码方法以及装置、解码方法以及装置 - Google Patents
编码方法以及装置、解码方法以及装置 Download PDFInfo
- Publication number
- WO2012122798A1 WO2012122798A1 PCT/CN2011/081102 CN2011081102W WO2012122798A1 WO 2012122798 A1 WO2012122798 A1 WO 2012122798A1 CN 2011081102 W CN2011081102 W CN 2011081102W WO 2012122798 A1 WO2012122798 A1 WO 2012122798A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image block
- difference
- size
- lcu
- prediction value
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 99
- 238000013139 quantization Methods 0.000 claims abstract description 40
- 238000004364 calculation method Methods 0.000 claims description 25
- 238000004458 analytical method Methods 0.000 claims description 2
- 238000012795 verification Methods 0.000 claims 2
- 238000010586 diagram Methods 0.000 description 15
- 230000006835 compression Effects 0.000 description 8
- 238000007906 compression Methods 0.000 description 8
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
Classifications
-
- 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
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
-
- 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
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/124—Quantisation
-
- 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
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/146—Data rate or code amount at the encoder output
- H04N19/149—Data rate or code amount at the encoder output by estimating the code amount by means of a model, e.g. mathematical model or statistical model
-
- 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
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
- H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
-
- 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
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/189—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the adaptation method, adaptation tool or adaptation type used for the adaptive coding
- H04N19/196—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the adaptation method, adaptation tool or adaptation type used for the adaptive coding being specially adapted for the computation of encoding parameters, e.g. by averaging previously computed encoding parameters
-
- 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
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/115—Selection of the code volume for a coding unit prior to coding
-
- 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
- H04N19/46—Embedding additional information in the video signal during the compression process
-
- 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
- H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
- H04N19/96—Tree coding, e.g. quad-tree coding
Definitions
- the present invention relates to the field of data processing, and in particular, to an encoding method and apparatus, a decoding method, and an apparatus.
- the encoding end needs to spatially transform the original image block data or the predicted image block difference data, and quantize the transform coefficients, and then entropy encode the quantized coefficients.
- the decoding end performs inverse quantization operation on the coefficients obtained by entropy decoding, reconstructs the transform coefficients, and performs inverse transform operations to obtain original image block data or image block difference data.
- the decoding end In order to perform the inverse quantization operation on the decoding end, the decoding end needs to obtain the quantization step size (QStep) used in the quantization process of the encoding end. Therefore, the encoding side needs to write the adopted Qstep information into the code stream. Since different Qsteps can be represented by different quantization parameters (QP, Quantization Parameter), the encoder can encode the QP and transmit it to the decoder.
- QStep quantization step size
- QStep quantization step size
- the encoder can encode the QP and transmit it to the decoder.
- the image is divided into equal-sized Largest Coding Units (LCUs), and each LCU can include one or more CUs of different sizes.
- LCUs Largest Coding Units
- each LCU corresponds to a QP.
- slice header (SH, Slice Header) ⁇ slice_qp_delta
- Coding_tree ( ⁇ , yO, log2CUSize ) ⁇ if( !skip_flag[ xO ][ yO ] )
- the encoded data is sent to the decoding end, and the decoding end calculates the quantization parameter QPLCU of the current LCU for each LCU as follows:
- QPLCU lcu_qp_delta+ slice_qp_delta+ pic_init_qp_minus26+26.
- the decoder After the decoder obtains the QP CU of the current LCU, it can decode.
- each LCU corresponds to one QP, and when the code end is adjusted by QP for code rate control, the LCU can only be controlled with the finest granularity, because in actual applications, the LCU tends to be set relatively large, for example, 64 pixels * 64 pixels (described for the cartridge, the subsequent cartridge is called 64 * 64, other data is similar), so the prior art scheme affects the accuracy of the rate control.
- the qp_delta information of the CU is carried in each CU of each LCU, so that the accuracy of the rate control can be improved.
- the embodiment of the invention provides an encoding method and device, a decoding method and device, and a codec system, which can effectively improve the rate control precision and the compression efficiency.
- the encoding method provided by the embodiment of the present invention includes: the encoding end acquires data to be encoded, and the data to be encoded is at least one maximum coding unit LCU; the encoding end determines the quantization depth parameter of the LCU according to a preset code rate control algorithm, And writing the quantized depth parameter to the data to be encoded, where the quantized depth parameter is used to represent a minimum image block size with independent QP in the LCU; and the encoding end is in the LCU according to the minimum image block size.
- each CU included determines the QP of each CU; the encoding end calculates the QP difference of each CU according to the QP of each CU and the QP prediction value of each CU; for each CU that satisfies the preset condition, the encoding end is in the CU And carrying the QP difference of the CU; the encoding end encodes the quantized depth parameter, the QP difference of the CU that satisfies the preset condition, and each CU to obtain a code stream.
- the decoding method provided by the embodiment of the present invention includes: the decoding end parses the received code stream to obtain a quantization depth parameter, where the quantization depth parameter is used to represent a minimum image block size with an independent QP in the LCU; Calculating a QP prediction value of each CU by using a minimum image block size and a size of each CU included in the LCU; for each CU that satisfies a preset condition, the decoding end parses the QP difference from the code stream; the decoding end is according to the The QP prediction value of each CU and the QP difference obtained by the analysis calculate the QP of each CU; the decoding end decodes each CU according to the QP of each CU.
- the encoding apparatus includes: a data acquiring unit, configured to acquire data to be encoded, where the data to be encoded is at least one maximum coding unit LCU; and a depth parameter processing unit, configured to control a code rate according to a preset code rate Determining a quantized depth parameter of the LCU, and writing the quantized depth parameter to data to be encoded acquired by the data acquiring unit, where the quantized depth parameter is used to represent a minimum image block size with independent QP in the LCU And a quantization parameter determining unit, configured to calculate, according to a QP of each CU determined by the quantization parameter determining unit and a QP predicted value of each CU, a QP difference of each CU; a filling unit, configured to satisfy each of the CUs
- the CU of the preset condition carries the QP difference of the CU in the CU, and the coding unit is configured to perform coding on the quantized depth parameter, the QP difference of the CU that satisfie
- the decoding apparatus includes: a first parsing unit, configured to parse the received code stream to obtain a quantized depth parameter, where the quantized depth parameter is used to represent a minimum image block size having an independent QP in the LCU; a parameter prediction unit, configured to calculate a QP prediction value of each CU according to the minimum image block size and a size of each CU included in the LCU; and a second parsing unit, configured to Each CU that satisfies the preset condition parses the QP difference from the code stream; the parameter calculation unit is configured to: obtain a QP prediction value of each CU obtained by the parameter prediction unit, and a QP obtained by the second parsing unit The difference is calculated by the QP of each CU; and the decoding unit is configured to decode each CU according to the QP of each CU calculated by the parameter calculation unit.
- the embodiments of the present invention have the following advantages:
- the encoding end may write a quantization depth parameter for indicating a minimum image block size with an independent QP in the data to be encoded, and carry the same for each CU that satisfies the preset condition.
- the QP of the CU is different, so that each LCU not only corresponds to one QP, but the CU corresponding to the preset condition in the LCU corresponds to the QP, so the encoding end can perform rate control with the CU as the finest granularity, since an LCU often includes Multiple CUs can effectively improve the bit rate control accuracy of the LCU;
- the coder only carries the QP difference of the CU in the CU that satisfies the preset condition, and does not carry the QP difference in all CUs, so the coding overhead can be reduced, thereby effectively improving the overall compression efficiency.
- FIG. 1 is a schematic diagram of an embodiment of an encoding method according to the present invention.
- FIG. 2 is a schematic diagram of an embodiment of a decoding method according to the present invention.
- FIG. 3 is a schematic diagram of an embodiment of a codec method according to the present invention.
- FIG. 4 is a schematic diagram of a positional relationship between a CU and a neighboring CU according to the present invention
- FIG. 5 is a schematic structural view of an LCU of the present invention.
- FIG. 6 is a schematic diagram of a code stream structure after coding according to the present invention.
- FIG. 7 is a schematic diagram of another embodiment of a codec method according to the present invention.
- FIG. 8 is a schematic diagram of another code stream structure after encoding according to the present invention.
- FIG. 9 is a schematic diagram of an embodiment of an encoding apparatus according to the present invention.
- FIG. 10 is a schematic diagram of another embodiment of an encoding apparatus according to the present invention.
- FIG. 11 is a schematic diagram of another embodiment of an encoding apparatus according to the present invention.
- FIG. 12 is a schematic diagram of an embodiment of a decoding apparatus according to the present invention.
- FIG. 13 is a schematic diagram of another embodiment of a decoding apparatus according to the present invention.
- FIG. 14 is a schematic diagram of another embodiment of a decoding apparatus according to the present invention. detailed description
- the embodiment of the invention provides an encoding method and device, a decoding method and a device, which can effectively improve the rate control precision and the compression efficiency.
- an embodiment of an encoding method in an embodiment of the present invention includes:
- the encoder obtains data to be encoded.
- the encoding end can obtain data to be encoded, and the data to be encoded is at least one LCU, that is, the data to be encoded can be an LCU or data composed of multiple LCUs.
- the encoding end determines a quantized depth parameter of the LCU according to a preset code rate control algorithm, and writes the quantized depth parameter into the data to be encoded.
- the encoding end acquires a quantized depth parameter of the LCU according to a preset rate control algorithm, and the quantized depth parameter is used to represent a minimum image block size with an independent QP in the LCU.
- the size described in this embodiment and the subsequent embodiments refers to the side length. For example, if a CU is 32*32, the size is defined as 32.
- the coding end can obtain the quantized depth parameter of the LCU by using an alternative method, that is, continuously adjusting the value of the quantization depth parameter and performing analog coding, so that the coded code rate meets the expected requirement, and the specific process is not limited herein.
- the quantized depth parameter can be written into the data to be encoded to prepare for transmission to the decoding end.
- the encoding end determines a QP of each CU according to a minimum image block size and a size of each CU included in the LCU;
- the minimum image block size with independent QP corresponding to the quantized depth parameter may be known, and the QP of each CU is determined according to the minimum image block size and the size of each CU included in the LCU. .
- the coding end calculates a QP difference of each CU according to a QP of each CU and a QP prediction value of each CU.
- the encoding end may calculate the QP prediction value of each CU according to the QP of each CU obtained in step 103, and calculate the QP difference of each CU according to the QP prediction value and QP.
- the QP difference of a certain CU may be defined as the difference between the QP of the CU and the QP predicted value of the CU. 105. For each CU that meets the preset condition, the encoding end carries the QP difference of the CU in the CU.
- a CU that satisfies the preset condition can be selected from each CU, and the QP difference is carried in the CUs that satisfy the preset condition.
- the coding end obtains a code stream by using a quantization depth parameter, a QP difference of a CU that satisfies a preset condition, and a coding of each CU.
- the encoding end writes the quantized depth parameter into the data to be encoded, and after writing the QP difference to the CU that satisfies the preset condition, the entire data to be encoded may be encoded, that is, the quantization depth parameter is satisfied, and the preset is satisfied.
- the QP difference of the CU of the condition and the CU are encoded to obtain a code stream.
- the encoding end can be written in the data to be encoded for indicating independence when encoding.
- the quantized depth parameter of the minimum image block size of the QP and for each CU that satisfies the preset condition, carries the QP difference of the CU, so that each LCU not only corresponds to one QP, but the preset condition is satisfied by the LCU.
- the CU corresponds to the QP, so the encoding end can perform rate control with the CU as the finest granularity. Since one LCU often contains multiple CUs, the LCU's rate control accuracy can be effectively improved.
- the coder only carries the QP difference of the CU in the CU that satisfies the preset condition, and does not carry the QP difference in all CUs, so the coding overhead can be reduced, thereby effectively improving the overall compression efficiency.
- an embodiment of the decoding method of the present invention includes:
- the decoding end parses the received code stream to obtain a quantization depth parameter.
- the decoding end may receive the encoded code stream from the encoding end. Since the encoding end writes the quantization depth parameter in the data to be encoded when encoding, the decoding end may correspond to the code stream according to a predetermined convention. The location resolves the quantized depth parameter.
- This quantized depth parameter is used to represent the minimum image block size with independent QP in the LCU.
- the decoding end calculates a QP prediction value of each CU according to a minimum image block size and a size of each CU included in the LCU.
- the minimum image block size with an independent QP may be learned according to the quantized depth parameter, and according to the minimum image block size.
- the QP prediction value of each CU is calculated by the size of each CU included in the LCU.
- the decoding end parses the QP difference from the code stream.
- the decoding end may further parse the CU from the code stream for each CU that satisfies the preset condition.
- the QP of the CU is poor.
- the process of obtaining the QP difference between the decoding end and the process of calculating the QP prediction value of each CU by the decoding end are not limited in sequence, and the process of obtaining the QP difference may be performed first, or the QP prediction of each CU may be performed first.
- the process of the value, or two processes can also be performed at the same time, which is not limited herein.
- the decoding end calculates a QP of each CU according to the QP prediction value of each CU and the parsed QP difference.
- the decoding end After the decoding end knows the QP prediction value of each CU through step 202 and the QP difference is obtained by step 203, the QP of each CU can be calculated according to these parameters.
- the decoding end decodes each CU according to a QP of each CU.
- the CU may perform a dequantization process and a subsequent decoding process, which are not limited herein.
- the encoding end may write a quantization depth parameter for indicating a minimum image block size with independent QP in the data to be encoded during encoding, and carry the CU for each CU that satisfies the preset condition.
- the QP difference is such that each LCU not only corresponds to one QP, so decoding can be performed for each CU when decoding is performed at the decoding end, without decoding for an entire LCU, thereby improving decoding efficiency.
- an embodiment of the codec method of the present invention includes:
- the encoding end acquires a sequence, an image, or a stripe to be encoded.
- the data to be encoded acquired by the encoding end is a sequence or an image or a stripe. It should be noted that a sequence includes several images, and one image includes several strips, and one strip includes several strips. LCU.
- the data to be encoded acquired by the encoding end is a plurality of LCUs.
- the encoding end determines the LCU quantization depth parameter lcu_qp_depth according to a preset code rate control algorithm; The encoding end obtains the value of the corresponding lcu_qp_depth according to the preset code rate control algorithm. Specifically, the encoding end may obtain the value of lcu_qp_depth by using an alternative method, that is, continuously adjust the value of lcu_qp_depth and perform analog coding, so that the coded rate is encoded. Meet the expected requirements, the specific process is not limited here.
- lcu_qp_depth in this embodiment can be from 0 to all integers in MaxSymbol, where MaxSymbol can be calculated as follows:
- MaxSymbol log 2 ( max_coding_block_size ) -log 2 ( min_coding_block_size ); where max_coding_block_size represents the size of the largest CU, ie the size of the LCU, and min_coding_block_size represents the size of the smallest CU.
- the size described in this embodiment and the subsequent embodiments refers to the side length. For example, if a CU is 32*32, the size is defined as 32.
- the lcu_qp_depth is used to represent the minimum image block size with independent QP in the LCU.
- lcu_qp_depth is 0, the smallest image block with independent QP is the LCU.
- lcu_qp_depth is 1, the LCU is divided into Four 32*32 CUs with a minimum image block size of 32, and so on.
- the lcu_qp_depth in this embodiment refers to the minimum image block size of each LCU in the sequence, image or stripe with independent QP, that is, all LCUs in the sequence, image or stripe use the same lcu_qp_depth.
- the encoding end writes lcu_qp_depth into the sequence parameter set, the image parameter set, or the strip header information.
- the encoding end may write the lcu_qp_depth into the data to be encoded:
- the encoding end can write lcu_qp_depth into the sequence parameter set (SPS, Sequence Parameter Set) of the sequence, which can be as follows:
- the encoding end can write lcu_qp_depth to the PPS of the image, as shown below: Pic_parameter_set_rbsp( ) ⁇ lcu_qp_depth
- the encoding end can write lcu_qp_depth into the strip header information of the stripe, as shown below:
- the write position of the lcu_qp_depth is only described by a few examples in the embodiment. In actual applications, the lcu_qp_depth can also be written in other positions of the sequence, the image or the stripe, which is not limited herein.
- the encoding end calculates a minimum image block size including an independent QP. QPBlkSize;
- the size of the smallest image block QPBlk can be calculated according to the leu_qp_depth.
- QPBlkSize the coder obtains leu_qp_depth.
- QPBlkSize can be calculated as follows:
- QPBlkSize max_coding_block_size/ ( 2 lcu - qp - depth ).
- the encoding end acquires a QP of each CU according to a QPBlkSize, a size of the CU.
- the encoding end can obtain the QP of each CU in the following situations:
- the encoding end calculates the QP of the CU according to a preset rate control algorithm
- the encoding end For a CU whose size is smaller than QPBlkSize, the encoding end first determines the smallest image block to which the CU belongs, and then calculates the QP of the minimum image block according to a preset rate control algorithm, and calculates the minimum image.
- the QP of the block acts as the QP for all CUs in the smallest image block.
- the process of calculating the QP by the encoding end according to the preset code rate control algorithm is common knowledge of the person skilled in the art, and is not limited herein.
- the encoder obtains a QP prediction value of each CU.
- the QP prediction value of each CU can be obtained by using the QP of the coded CU adjacent to each CU according to the coding order (the decoding order is the same as the coding order).
- the neighboring coded CU includes a left CU, an upper CU and an upper left CU.
- the encoding end uses each CU whose size is greater than or equal to QPBlkSize as a reference CU, and for a CU whose size is smaller than QPBlkSize, the encoding end determines the smallest image block to which the CU belongs, and uses the upper left CU of the minimum image block as a reference cu.
- the encoding end can calculate the QP prediction value of each reference CU according to the QP of the neighboring CU of each reference CU.
- the specific calculation manner can be various. The following describes several examples:
- the CU may be a CU whose size is greater than or equal to QPBlkSize, or may be a CU located in the upper left corner of the smallest image block.
- CU C is the CUu CU CU
- CU L C is a left CU CU
- CUUL the upper left CU CU C in this embodiment, CU C adjacent to CU CU left, upper left and upper CU CU, wherein:
- CU C The pixel coordinates of the upper left corner of CU C are (x c , y c ), then CU L is the pixel containing (x c -l , y c )
- CU is a CU containing a pixel (x c , y c -l )
- CU UL is a CU containing a pixel (x c -l , y c -l ).
- the QPs of the CUs may be obtained according to the result calculated in step 305, where QP of CU C is QP C , QP of CU D is QPu, QP of CU L is QPL, CUUL QP is QPUL.
- each CU can acquire its left CU, upper CU, and upper left CU.
- the CU that is located at the leftmost side of the image or strip will not have the left CU. Therefore, the encoding end can follow
- the predicted values of each CU are determined by the following specific conditions:
- QPp slice_qp_delta+pic_init_qp_minu s26+26, where slice_qp_delta is information included in the slice header information, and pic_init_qp_minus26 is information included in the image parameter set.
- IQP L - QP UL l IQPu - QP UL I
- CU L and CU UL are likely to belong to the same object in the image, and the edge of the object appears at the CUu and CU UL edges, so QPu can be used as QP P ;
- CU L and CUu can be used as a reference CU C.
- the neighboring CU, CU C may be a CU whose size is greater than or equal to QPBlkSize, or may be a CU located in the upper left corner of the smallest image block.
- the pixel coordinates of the upper left corner of CU C are (x c , y c ), then CU L is a CU containing pixel points (x c -l , y c ), and CUu is a pixel containing (x c , y c -l )
- the QP of CU C is QP C
- the QP of CUu is QPu
- the QP of CU L is QP L .
- the encoding end can be as follows. The condition determines the predicted value of each CU:
- QPL is taken as QPp
- QPp slice_qp_delta+pic_init_qp_minu s26+26, where slice_qp_delta is information included in the slice header information, and pic_init_qp_minus26 is information included in the image parameter set.
- the encoding end writes the QP difference cu_qp_delta of the CU that satisfies the preset condition into the CU.
- the encoding end calculates the QP of each CU by using step 305, and calculates the QP prediction value of each reference CU through step 306, and the encoding end.
- the QP difference cu_qp_delta of each CU can be further calculated according to these parameters, and the specific calculation manner can be:
- the encoding side refers to the reference CU.
- the QP difference is the QP difference of other CUs in the smallest image block.
- the encoding end can use the difference between the QP and QP prediction values of these CUs as cu_qp_delta of these CUs;
- a minimum image block includes a plurality of CUs
- the CU of the upper left corner of the minimum image block is a reference CU
- the QP prediction value of the reference CU can be obtained through step 306, and then combined.
- the CU_qp_delta of the reference CU of the smallest image block to which the CU belongs may be carried in the CU, for example, if it is determined that a certain CU in the minimum image block satisfies the preset condition.
- the CDR after acquiring the cu_qp_delta of each CU, the CDR can carry the cu_qp_delta of the CU in the CU that meets the preset condition, and the specific CU that satisfies the preset condition may have non-zero quantization in the compressed code stream.
- CU of the post-transform coefficient after acquiring the cu_qp_delta of each CU, the CDR can carry the cu_qp_delta of the CU in the CU that meets the preset condition, and the specific CU that satisfies the preset condition may have non-zero quantization in the compressed code stream.
- a CU adopts the skip coding mode, there is no non-zero quantized transform coefficient in the compressed code stream, so the CU does not satisfy the preset condition. If the skip coding mode is not used, the CU may be in its compressed code stream. It will contain non-zero quantized transform coefficients, and these CUs will satisfy the preset conditions.
- the encoding end may write the cu_qp_delta of the CU to the following location of the CU:
- Coding_unit ( ⁇ , y0, log2CUSize ) ⁇
- sig_coeff_num 0 is a decision condition of whether cu_qp_delta exists in the code stream corresponding to the current CU. If the decision condition is true, it indicates that cu_qp_delta exists in the code stream corresponding to the current CU, and if the decision condition is false, the current CU is There is no cu_qp_delta in the corresponding code stream.
- the encoding end sequentially determines, according to the coding order, whether the CU in the smallest image block to which the CU belongs satisfies the preset condition. If a CU does not satisfy the preset condition, it continues to determine the subsequent Whether the CU satisfies the preset condition, if a subsequent CU satisfies the preset condition, the cu_qp_delta of the CU is written into the CU, and the judgment is stopped, that is, the cu_qp_delta is no longer written for other CUs in the minimum image block. , regardless of whether these CUs meet the preset conditions.
- the encoding end can re-carry the QP of the CU that does not satisfy the preset condition, in addition to the cu_qp_delta of the CU in the CU that satisfies the preset condition.
- the specific way can be:
- the encoding end For each CU that does not meet the preset condition, the encoding end sets the QP of the CU by: For a CU whose size is greater than or equal to QPBlkSize, the encoding end uses the QP prediction value of the CU as the QP of the CU;
- the encoding end uses the QP prediction value of the reference CU in the smallest image block to which the CU belongs as the QP of the CU. .
- the encoding end encodes the lcu_qp_depth, cu_qp_delta, and each CU to obtain a code stream; the encoding end can encode each CU after acquiring lcu_qp_depth, cu_qp_delta, in this embodiment,
- the process of encoding the lcu_qp_depth by the encoding end may include:
- the encoding end binarizes lcu_qp_depth by using a fixed length code or a variable length code to obtain a corresponding binary codeword;
- the encoding side writes the obtained binary codeword into the code stream corresponding to the position determined in step 302 in accordance with the method of fixed length coding, variable length coding or arithmetic entropy coding.
- variable length code used by the coding end may be an unsigned exponential Columbus code, or other similar variable length code, which is not limited herein.
- the process of encoding cu_qp_delta by the encoding end may include:
- the encoding end uses a variable length code to binarize the cu_qp_delta of the CU that satisfies the preset condition to obtain a corresponding binary codeword;
- the encoding side writes the obtained binary codeword into the code stream according to a method of variable length coding or arithmetic entropy coding.
- the encoding side writes the obtained binary codeword into the code stream corresponding to the position determined in step 307 in accordance with the method of fixed length coding, variable length coding or arithmetic entropy coding.
- variable length code used by the coding end may be a signed exponential Columbus code, or other similar variable length code, which is not limited herein.
- the encoding process of the CU by the encoding end is not limited herein.
- the code stream is obtained, and the encoding end may send the code stream to the decoding end.
- the decoding end decodes the received code stream to obtain lcu_qp_depth.
- the decoding end After receiving the code stream, the decoding end can know that the position of lcu_qp_depth in the code stream is SPS, or PPS, or SH according to the agreement with the encoding end in advance, thereby using fixed length decoding, variable length decoding or arithmetic for the content of the part.
- Entropy decoding method for obtaining a binary codeword SPS, or PPS, or SH according to the agreement with the encoding end in advance, thereby using fixed length decoding, variable length decoding or arithmetic for the content of the part.
- the decoding end uses a fixed length code or a variable length code to inversely binarize the binary codeword to obtain lcu_qp_depthippo
- variable length code in this embodiment may be an unsigned exponential Columbus code, or other similar variable length code, which is not limited herein.
- the lcu_qp_depth in this embodiment is used to represent each LCU in a sequence, or an image, or a stripe.
- the decoder calculates a minimum image block size including an independent QP. QPBlkSize;
- the QPBlkSize can be calculated according to the lcu_qp_depth.
- the specific calculation method can be the same as the method of calculating the QPBlkSize in the foregoing step 304, which is not described here.
- the decoding end parses the cu_qp_delta of the CU from the CU that meets the preset condition. In this embodiment, for each CU that satisfies the preset condition, the decoding end may use the variable position from the corresponding position in the code stream of the CU. Decoding or arithmetic entropy decoding method for obtaining a binary codeword;
- the decoding end uses a variable length code to inversely binarize the binary codeword to obtain cu_qp_delta of the CU.
- the variable length code in this embodiment may be a signed exponential Columbus code, or other similar variable length code, which is not limited herein.
- sig_coeff_num > 0 is a decision condition of whether cu_qp_delta exists in the code stream corresponding to the current CU, so the decoding end can check whether the CU satisfies the preset condition by checking the decision condition.
- the CU satisfying the preset condition may also be a CU having a non-zero quantized transform coefficient in its compressed code stream.
- the decoding end can determine cu_qp_delta of all CUs as follows:
- the decoding end determines whether the CU satisfies a preset condition, and if so, parses the cu_qp_delta of the CU from the CU, and if not, sets the cu_qp_delta of the CU to 0;
- the decoding end determines the smallest image block to which the CU belongs, and sequentially determines whether each CU in the minimum image block satisfies a preset condition according to the decoding order of the CU in the minimum image block, and if yes, The judgment is stopped, and cu_qp_delta of the CU is taken as cu_qp_delta of all CUs in the minimum image block. If all the CUs in the minimum image block do not satisfy the preset condition, cu_qp_delta of all CUs in the minimum image block is set to 0.
- the decoding end acquires the QP prediction value of each CU according to the QPBlkSize, the size of the CU. After the decoding end obtains the QPBlkSize, the QP prediction value of each CU may be obtained according to the size of each CU and the QPBlkSize. the same. As described in 306, the neighboring area is an encoded or decoded area, and since the encoding order is the same as the decoding order, the encoded area and the decoded area are unified areas. Since the QP of the CU in the adjacent area is available, the QP prediction value of each CU can be obtained in the same manner as in 306.
- the decoding end calculates the QP of each CU according to the QP prediction value of each CU and cu_qp_delta. After the decoding end calculates the QP prediction value of each CU, the QP of each CU may be calculated according to the QP prediction value of each CU and cu_qp_delta. Can be:
- the decoding end uses the QP prediction value of the CU as the QP of the CU;
- the decoding end uses the sum of the QP predicted value of the CU and the cu_qp_delta of the CU as the QP of the CU;
- the decoding end acquires the QP prediction value of the upper left CU of the smallest image block to which the CU belongs, and uses the sum of the QP prediction value and the cu_qp_delta as the QP of the CU. And using the QP of the CU as the QP of all CUs in the minimum image block;
- the decoding end acquires the QP prediction value of the upper left CU of the smallest image block to which the CU belongs, and the predicted value of the QP is obtained. As the QP of all CUs in the smallest image block.
- the decoding end decodes each CU according to the QP of each CU.
- the decoding end may perform the inverse quantization process and the subsequent decoding process for each CU by using the QP of each CU in step 313, which is not limited herein.
- FIG. 5 is a schematic structural diagram of an LCU according to the present invention, wherein max_coding_block_size is 64 and QPBlkSize is 16.
- the LCU includes 37 CUs, wherein the CU of the fork is a CU that uses skip coding, that is, a CU that does not satisfy the preset condition, and in the remaining CUs, it is assumed that CU3 and CU23 are compressed in it.
- the quantized transform coefficients in the code stream are all 0, and CU3 and CU23 are also CUs that do not satisfy the preset condition, and the remaining CUs are CUs that satisfy the preset conditions.
- the code stream structure obtained by encoding the LCU shown in FIG. 5 is as shown in FIG. 6, and Ml, M2, M37, etc. are header information, CI, C4. ..C37 and so on are transform coefficients, DQP1, DQP4...DQP37, etc. are cu_qp_delta.
- the encoding end may write a quantization depth parameter for indicating a minimum image block size with independent QP in the data to be encoded during encoding, and carry the CU for each CU that satisfies the preset condition.
- the QP difference is such that each LCU not only corresponds to one QP, but the CU that satisfies the preset condition in the LCU corresponds to the QP, so the encoding end can perform rate control with the CU as the finest granularity, since an LCU often includes Multiple CUs can effectively improve the bit rate control accuracy of the LCU;
- the encoder only carries the QP difference of the CU in the CU that satisfies the preset condition, and does not carry the QP difference in all the CUs, so the coding overhead can be reduced, thereby effectively improving the overall compression efficiency;
- a plurality of prediction modes may be adopted, so that the calculation process is more flexible, and thus can adapt to a plurality of different environments.
- Another embodiment of the codec method of the present invention includes:
- the encoder obtains an LCU to be encoded.
- the data to be encoded acquired by the encoding end is an LCU.
- the encoding end determines the LCU quantization depth parameter lcu_qp_depth according to a preset code rate control algorithm
- the encoding end obtains the value of the corresponding lcu_qp_depth according to the preset code rate control algorithm. Specifically, the encoding end may obtain the value of lcu_qp_depth by using an alternative method, that is, continuously adjust the value of lcu_qp_depth and perform analog coding, so that the coded rate is encoded. Meet the expected requirements, the specific process is not limited here.
- lcu_qp_depth in this embodiment can be from 0 to all integers in MaxSymbol, where MaxSymbol can be calculated as follows:
- MaxSymbol log 2 ( max_coding_block_size ) -log 2 ( min_coding_block_size ); where max_coding_block_size represents the size of the largest CU, ie the size of the LCU, and min_coding_block_size represents the size of the smallest CU. It should be noted that the size described in this embodiment and the subsequent embodiments refers to the side length. For example, if a certain CU is 32*32, the size is defined as 32.
- the lcu_qp_depth is used to represent the minimum image block size with independent QP in the LCU.
- lcu_qp_depth is 0, the smallest image block with independent QP is the LCU.
- lcu_qp_depth is 1, the LCU is divided into Four 32*32 CUs with a minimum image block size of 32, and so on.
- the lcu_qp_depth in this embodiment refers to the minimum image block size of the current LCU with independent QP, that is, each LCU in the sequence, image or stripe uses an independent lcu_qp_depth, which is used by each LCU.
- Lcu_qp_depth may be different.
- the encoding end writes lcu_qp_depth into the CU in the current LCU in the coding order, where the first one has a non-zero quantized transform coefficient in the compressed code stream;
- the encoding end may write the lcu_qp_depth into the current LCU in the CU that has the first non-zero quantized transform coefficient in its compressed code stream in the coding order:
- CurrTbAddr first_tb_in_slice
- xCU HorLumaLocation( CurrTbAddr )
- Coding_tree ( xCU, yCU, Log2TbSize ) ⁇ coding_unit( xO, yO, log2CUSize ) ⁇
- the lcu_qp_depthFlag variable is introduced to achieve the purpose of transmitting lcu_qp_depth only in the first CU in which there is a non-zero quantized transform coefficient in its compressed code stream.
- other methods can be used to implement the same code stream structure.
- sig_coeff_num>0 is a decision condition of whether or not lcu_qp_depth exists in the current CU code stream.
- Lcu_qp_depth can also be placed in other locations in the current CU code stream when the decision condition is met, which is not limited herein.
- the encoding end checks, in the current LCU, whether each CU has a condition of carrying lcu_qp_depth according to the coding order of the CU, that is, whether the CU is The first CU that has non-zero quantized transform coefficients in its compressed code stream.
- the encoder continues to check the subsequent CUs of the current LCU. If it is found in a CU that it has the condition of carrying lcu_qp_depth, then lcu_qp_depth is written to the CU at the specified location.
- the decoding end decodes the received code stream to obtain lcu_qp_depth.
- the decoder After the decoder receives the code stream, it can know the lcu_qp_depth according to the agreement with the encoding end in advance. Positioning in the code stream, thereby acquiring binary code words for the content of the part using fixed length decoding, variable length decoding or arithmetic entropy decoding;
- the decoding end uses a fixed length code or a variable length code to inversely binarize the binary codeword to obtain lcu_qp_depthippo
- variable length code in this embodiment may be an unsigned exponential Columbus code, or other similar variable length code, which is not limited herein.
- the lcu_qp_depth in this embodiment is used to indicate the minimum image block size with an independent QP in the current LCU.
- FIG. 5 is a schematic structural diagram of an LCU according to the present invention, where max_coding_block_size is 64 and QPBlkSize is 16.
- the LCU includes 37 CUs, wherein the CU of the fork is a CU that uses skip coding, that is, a CU that does not satisfy the preset condition, and in the remaining CUs, it is assumed that CU3 and CU23 are compressed in it.
- the quantized transform coefficients in the code stream are all 0, and CU3 and CU23 are also CUs that do not satisfy the preset condition, and the remaining CUs are CUs that satisfy the preset conditions.
- the code stream structure obtained after the LCU shown in FIG. 5 is encoded is as shown in FIG. 8, and the depth is lcu_qp_depth, Ml, M2, M37, etc. are the heads.
- Information, CI, C4...C37, etc. are transform coefficients, DQP1, DQP4...DQP37, etc. are cu_qp_delta.
- the encoding end may write a quantization depth parameter for indicating a minimum image block size with independent QP in the data to be encoded during encoding, and carry the CU for each CU that satisfies the preset condition.
- the QP difference is such that each LCU not only corresponds to one QP, but the CU corresponding to the preset condition in the LCU corresponds to the QP, so the encoding end can perform rate control with the CU as the finest granularity, since an LCU often contains many CU, therefore, can effectively improve the bit rate control accuracy of the LCU;
- the coder only carries the QP difference of the CU in the CU that satisfies the preset condition, and does not carry the QP difference in all CUs, so the coding overhead can be reduced, thereby effectively improving the overall Compression efficiency
- a plurality of prediction modes may be adopted, so that the calculation process is more flexible, and thus can adapt to multiple different environments;
- lcu_qp_depth is used to indicate the minimum image block size with independent QP in the current LCU, so each LCU carries a lcu_qp_depth, so that different LCUs can use different lcu_qp_depth, thereby further improving the LCU. Rate control accuracy.
- an embodiment of the encoding apparatus of the present invention includes:
- the data obtaining unit 901 is configured to obtain data to be encoded, where the data to be encoded is at least one maximum coding unit LCU;
- the depth parameter processing unit 902 is configured to determine a quantized depth parameter of the LCU according to a preset code rate control algorithm, and write the quantized depth parameter to the data to be encoded acquired by the data acquiring unit 901, where the quantized depth parameter is used.
- a minimum image block size having an independent QP in the LCU a quantization parameter determining unit 903, configured to determine a QP of each CU according to the minimum image block size and a size of each CU included in the LCU;
- the calculating unit 904 is configured to calculate a QP difference of each CU according to the QP of each CU determined by the quantization parameter determining unit 903 and the QP predicted value of each CU;
- a filling unit 905, configured to carry a QP difference of the CU in the CU for each CU that satisfies the preset condition
- the coding unit 906 is configured to encode the quantized depth parameter, the QP difference of the CU that satisfies the preset condition, and each CU to obtain a code stream.
- FIG. 10 another embodiment of the coding device of the present invention includes:
- the data obtaining unit 1001 is configured to obtain data to be encoded, where the data to be encoded is at least one largest coding unit LCU;
- the depth parameter processing unit 1002 is configured to determine a quantized depth parameter of the LCU according to a preset code rate control algorithm, and write the quantized depth parameter to the data to be encoded acquired by the data acquiring unit 1001, where the quantized depth parameter is used.
- the calculating unit 1005 is configured to calculate a QP difference of each CU according to the QP of each CU determined by the quantization parameter determining unit 1003 and the QP predicted value of each CU;
- a filling unit 1006 configured to carry the CU in the CU for each CU that meets the preset condition
- the coding unit 1007 is configured to encode the quantized depth parameter, the QP difference of the CU that satisfies the preset condition, and each CU to obtain a code stream.
- a prediction unit 1004 configured to calculate, according to a QP of a neighboring CU of each reference CU, each reference CU
- each CU serves as a reference CU, and for a CU whose size is smaller than the minimum image block size, the upper left of the smallest image block to which the CU belongs
- the prediction unit 1004 in this embodiment includes:
- a first check module 10041 configured to determine, for each reference CU, whether a neighboring CU of the reference CU exists, where a neighboring CU of the reference CU includes a left CU, an upper CU, and an upper left CU of the reference CU;
- a first prediction module 10042 configured to determine, according to a strip and an image to which the reference CU belongs, a QP prediction value of the reference CU when none of the neighboring CUs of the reference CU are present;
- a second prediction module 10043 configured to: when the reference CU only has a left CU, use the QP of the left CU as a QP prediction value of the reference CU;
- a third prediction module 10044 configured to: when the reference CU only has an upper CU, use the QP of the upper CU as a QP prediction value of the reference CU;
- the fourth prediction module 10045 is configured to acquire a first difference between a QP of a left CU and a QP of an upper left CU when a neighboring CU of the reference CU exists, and a first between a QP of the upper CU and a QP of the upper left CU If the first difference is smaller than the second difference, the QP of the upper CU is used as the QP prediction value of the reference CU, and if the first difference is greater than or equal to the second difference, the QP of the left CU is taken as the Refer to the QP prediction value of the CU.
- another embodiment of the encoding apparatus of the present invention includes:
- the data obtaining unit 1101 is configured to obtain data to be encoded, where the data to be encoded is at least one largest coding unit LCU;
- the depth parameter processing unit 1102 is configured to determine a quantized depth parameter of the LCU according to a preset code rate control algorithm, and write the quantized depth parameter to the data to be encoded acquired by the data acquiring unit 1101, where the quantized depth parameter is used. Representing a minimum image block size having an independent QP in the LCU; a quantization parameter determining unit 1103, configured to determine a QP of each CU according to the minimum image block size and a size of each CU included in the LCU;
- the calculating unit 1105 is configured to calculate a QP difference of each CU according to the QP of each CU determined by the quantization parameter determining unit 1103 and the QP predicted value of each CU;
- a padding unit 1106, configured to carry a QP difference of the CU in the CU for each CU that satisfies the preset condition
- the coding unit 1107 is configured to encode the quantized depth parameter, the QP difference of the CU that satisfies the preset condition, and each CU to obtain a code stream.
- a prediction unit 1104 configured to calculate a QP prediction value of each reference CU according to a QP of a neighboring CU of each reference CU;
- each CU serves as a reference CU, and for a CU whose size is smaller than the minimum image block size, the upper left CU of the smallest image block to which the CU belongs is used as a reference CU.
- the prediction unit 1104 in this embodiment includes:
- a second check module 11041 configured to determine, for each reference CU, whether a left CU of the reference CU exists
- a fifth prediction module 11042 configured to: when the left CU of the reference CU exists, use the QP of the left CU as a QP prediction value of the reference CU;
- the sixth prediction module 11043 is configured to: when the left CU of the reference CU does not exist, and the upper CU of the reference CU or the previous coded CU of the reference CU exists, the QP of the upper CU or the previous coded CU QP as the QP prediction value of the reference CU;
- the seventh prediction module 11044 is configured to: when the left CU of the reference CU, the upper CU, and the previous one are edited When the code CU does not exist, the QP prediction value of the reference CU is determined according to the strip and the image to which the reference CU belongs.
- the encoding end may write a quantization depth parameter for indicating a minimum image block size with independent QP in the data to be encoded during encoding, and carry the CU for each CU that satisfies the preset condition.
- the QP difference is such that each LCU not only corresponds to one QP, but the CU corresponding to the preset condition in the LCU corresponds to the QP, so the encoding end can perform rate control with the CU as the finest granularity, since an LCU often contains many CU, therefore, can effectively improve the bit rate control accuracy of the LCU;
- the encoder only carries the QP difference of the CU in the CU that satisfies the preset condition, and does not carry the QP difference in all the CUs, so the coding overhead can be reduced, thereby effectively improving the overall compression efficiency.
- an embodiment of the decoding apparatus of the present invention includes:
- a first parsing unit 1201 configured to parse the received code stream to obtain a quantized depth parameter, where the quantized depth parameter is used to represent a minimum image block size with an independent QP in the LCU;
- the parameter prediction unit 1202 is configured to calculate a QP prediction value of each CU according to the minimum image block size and a size of each CU included in the LCU;
- the second parsing unit 1203 is configured to parse the QP difference from the code stream for each CU that satisfies the preset condition
- the parameter calculation unit 1204 is configured to calculate a QP of each CU according to the QP prediction value of each CU obtained by the parameter prediction unit 1202 and the QP difference parsed by the second parsing unit 1203;
- the decoding unit 1205 is configured to decode each CU according to the QP of each CU calculated by the parameter calculation unit 1204.
- Another embodiment of the decoding device of the present invention includes:
- a first parsing unit 1301, configured to parse the received code stream to obtain a quantized depth parameter, where the quantized depth parameter is used to represent a minimum image block size with an independent QP in the LCU;
- the parameter prediction unit 1302 is configured to calculate a QP prediction value of each CU according to the minimum image block size and a size of each CU included in the LCU; a second parsing unit 1303, configured to parse the QP difference from the code stream for each CU that satisfies the preset condition;
- the parameter calculation unit 1304 is configured to calculate a QP of each CU according to the QP prediction value of each CU obtained by the parameter prediction unit 1302 and the QP difference obtained by the second parsing unit 1303;
- Decoding unit 1305, configured to perform QP pairs of each CU calculated according to the parameter calculation unit
- the CU performs decoding.
- the parameter prediction unit 1302 may further include:
- the first parameter check module 13021 is configured to determine, for each reference CU, whether a neighboring CU of the reference CU exists, where the neighboring CU of the reference CU includes a left CU, an upper CU, and an upper left CU of the reference CU, where the size is greater than or a CU equal to the minimum image block size, each CU as a reference CU, and for a CU having a size smaller than the minimum image block size, the upper left CU of the smallest image block to which the CU belongs is used as a reference CU;
- the first parameter prediction module 13022 is configured to determine, according to the strip and the image to which the reference CU belongs, the QP prediction value of the reference CU when none of the neighboring CUs of the reference CU are present;
- the second parameter prediction module 13023 is configured to: when the reference CU only has a left CU, use the QP of the left CU as a QP prediction value of the reference CU;
- the third parameter prediction module 13024 is configured to: when the reference CU only has the upper CU, use the QP of the upper CU as the QP prediction value of the reference CU;
- the fourth parameter prediction module 13025 is configured to acquire, when the neighboring CUs of the reference CU are both, a first difference between a QP of a left CU and a QP of an upper left CU, and a QP between the upper CU and a QP of the upper left CU. a second difference, if the first difference is smaller than the second difference, the QP of the upper CU is used as the QP prediction value of the reference CU, and if the first difference is greater than or equal to the second difference, the QP of the left CU is taken as The reference CU's QP prediction value.
- the parameter calculation unit 1304 in this embodiment may further include:
- the first calculating module 13041 is configured to: for a CU whose size is greater than or equal to the minimum image block size and does not satisfy a preset condition, use a QP predicted value of the CU as a QP of the CU;
- the second calculation module 13042 is configured to use, for a CU whose size is greater than or equal to the minimum image block size, and meets a preset condition, a sum of a QP predicted value of the CU and a QP difference of the CU as a QP of the CU; a third calculation module 13043, configured to: for a CU whose size is smaller than the minimum image block unit size, if the CU meets a preset condition, acquire a QP prediction value of an upper left CU of the smallest image block to which the CU belongs, and predict the QP The sum of the value and the QP difference of the CU is taken as the QP of the CU, and the QP of the CU is used as the QP of all the CUs in the minimum image block;
- the fourth calculation module 13044 is configured to: for a CU whose size is smaller than the minimum image block unit size, if all CUs in the smallest image block to which the CU belongs do not satisfy the preset condition, acquire an upper left CU of the smallest image block to which the CU belongs The QP prediction value, and the QP prediction value is taken as the QP of all CUs in the minimum image block.
- another embodiment of the decoding apparatus of the present invention includes:
- a first parsing unit 1401, configured to parse the received code stream to obtain a quantized depth parameter, where the quantized depth parameter is used to represent a minimum image block size with an independent QP in the LCU;
- a parameter prediction unit 1402 configured to calculate a QP prediction value of each CU according to the minimum image block size and a size of each CU included in the LCU;
- a second parsing unit 1403, configured to parse the QP difference from the code stream for each CU that satisfies the preset condition
- the parameter calculation unit 1404 is configured to calculate a QP of each CU according to the QP prediction value of each CU obtained by the parameter prediction unit 1402 and the QP difference obtained by the second parsing unit 1403;
- the decoding unit 1405 is configured to decode each CU according to the QP of each CU calculated by the parameter calculation unit.
- the parameter prediction unit 1402 may further include:
- the second parameter check module 14021 is configured to determine, for each reference CU, whether a left CU of the reference CU exists. For a CU whose size is greater than or equal to the minimum image block size, each CU is used as a reference CU, and the size is smaller. The CU of the minimum image block size, the upper left CU of the smallest image block to which the CU belongs is used as a reference CU;
- the fifth parameter prediction module 14022 is configured to: when the left CU of the reference CU exists, use the QP of the left CU as the QP prediction value of the reference CU;
- the sixth parameter prediction module 14023 is configured to: when the left CU of the reference CU does not exist, and the upper CU of the reference CU or the previous coded CU of the reference CU exists, the QP of the upper CU or the previous one is encoded.
- the QP of the CU is used as the QP prediction value of the reference CU;
- the seventh parameter prediction module 14024 is configured to: when the left CU of the reference CU, the upper CU, and the previous coded CU are not present, determine a QP prediction value of the reference CU according to the strip and the image to which the reference CU belongs.
- the parameter calculation unit 1404 in this embodiment may further include:
- a first calculation module 14041 configured to use a CU of the CU as a QP of the CU for a CU whose size is greater than or equal to the minimum image block size and does not satisfy the preset condition;
- a second calculation module 14042 configured to use, for a CU whose size is greater than or equal to the minimum image block size and meets a preset condition, a sum of a QP predicted value of the CU and a QP difference of the CU as a QP of the CU;
- a third calculating module 14043 configured to: for a CU whose size is smaller than the minimum image block unit size, if the CU satisfies a preset condition, acquire a QP prediction value of an upper left CU of the smallest image block to which the CU belongs, and predict the QP The sum of the value and the QP difference of the CU is taken as the QP of the CU, and the QP of the CU is used as the QP of all the CUs in the minimum image block;
- the fourth calculating module 14044 is configured to: for a CU whose size is smaller than the minimum image block unit size, if all CUs in the smallest image block to which the CU belongs do not satisfy the preset condition, acquire an upper left CU of the smallest image block to which the CU belongs The QP prediction value, and the QP prediction value is taken as the QP of all CUs in the minimum image block.
- the encoding end may write a quantization depth parameter for indicating a minimum image block size with independent QP in the data to be encoded during encoding, and carry the CU for each CU that satisfies the preset condition.
- the QP difference is such that each LCU not only corresponds to one QP, so decoding can be performed for each CU when decoding is performed at the decoding end, without decoding for an entire LCU, thereby improving decoding efficiency.
- Video encoders, decoders are widely used in a variety of communication devices or electronic devices, such as: digital TV, set-top box, media gateway, mobile phone, wireless device, personal data assistant (PDA), handheld or portable computer, GPS receiver / navigator, camera, video player, camcorder, video recorder, surveillance equipment, video conferencing and video telephony equipment, etc.
- Such devices include processors, memory, and interfaces for transmitting data.
- the video codec can be directly implemented by a digital circuit or a chip such as a DSP (digital signal processor), or a software code can be used to drive a processor to execute a software generation. The process in the code is implemented.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Algebra (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computing Systems (AREA)
- Theoretical Computer Science (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
Description
Claims
Priority Applications (8)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP11861079.9A EP2672707A4 (en) | 2011-03-11 | 2011-10-21 | METHOD AND DEVICE FOR ENCODING, AND METHOD AND DEVICE FOR DECODING |
BR112013022073-2A BR112013022073B1 (pt) | 2011-03-11 | 2011-10-21 | Método e dispositivo de codificação, e método e dispositivo de decodificação |
KR1020137024746A KR101581097B1 (ko) | 2011-03-11 | 2011-10-21 | 인코딩 방법 및 장치와, 디코딩 방법 및 장치 |
CA2828692A CA2828692C (en) | 2011-03-11 | 2011-10-21 | Determining a quantization depth parameter for the largest coding unit of a picture, and quantization parameter differences for each coding unit within the largest coding unit which meets a preset condition |
AU2011362462A AU2011362462B2 (en) | 2011-03-11 | 2011-10-21 | Encoding method and device, and decoding method and device |
RU2013145526/08A RU2566332C2 (ru) | 2011-03-11 | 2011-10-21 | Способ и устройство кодирования и способ и устройство декодирования |
JP2013556946A JP5860067B2 (ja) | 2011-03-11 | 2011-10-21 | 符号化方法及び装置並びに復号化方法及び装置 |
US14/024,047 US9571829B2 (en) | 2011-03-11 | 2013-09-11 | Method and device for encoding/decoding with quantization parameter, block size and coding unit size |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110059194.9A CN102685478B (zh) | 2011-03-11 | 2011-03-11 | 编码方法以及装置、解码方法以及装置 |
CN201110059194.9 | 2011-03-11 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US14/024,047 Continuation US9571829B2 (en) | 2011-03-11 | 2013-09-11 | Method and device for encoding/decoding with quantization parameter, block size and coding unit size |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2012122798A1 true WO2012122798A1 (zh) | 2012-09-20 |
Family
ID=46816769
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2011/081102 WO2012122798A1 (zh) | 2011-03-11 | 2011-10-21 | 编码方法以及装置、解码方法以及装置 |
Country Status (10)
Country | Link |
---|---|
US (1) | US9571829B2 (zh) |
EP (1) | EP2672707A4 (zh) |
JP (1) | JP5860067B2 (zh) |
KR (1) | KR101581097B1 (zh) |
CN (1) | CN102685478B (zh) |
AU (1) | AU2011362462B2 (zh) |
BR (1) | BR112013022073B1 (zh) |
CA (1) | CA2828692C (zh) |
RU (1) | RU2566332C2 (zh) |
WO (1) | WO2012122798A1 (zh) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2014534749A (ja) * | 2011-11-04 | 2014-12-18 | インフォブリッジ ピーティーイー. エルティーディー. | 量子化ブロック逆量子化方法 |
JP2017063489A (ja) * | 2011-12-21 | 2017-03-30 | 株式会社Jvcケンウッド | 動画像復号装置、動画像復号方法及び動画像復号プログラム |
JP2017079483A (ja) * | 2011-12-21 | 2017-04-27 | 株式会社Jvcケンウッド | 動画像符号化装置、動画像符号化方法及び動画像符号化プログラム |
JP2018152884A (ja) * | 2011-01-13 | 2018-09-27 | キヤノン株式会社 | 画像符号化装置、画像符号化方法及びプログラム、画像復号装置、画像復号方法及びプログラム |
US10419761B2 (en) | 2011-01-24 | 2019-09-17 | Sony Corporation | Image decoding device, image encoding device, and method thereof |
JP2020065290A (ja) * | 2011-06-21 | 2020-04-23 | インテレクチュアル ディスカバリー カンパニー リミテッド | クアッドツリー構造に基づく適応的量子化パラメータ符号化及び復号化方法及び装置 |
Families Citing this family (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2017079485A (ja) * | 2011-01-24 | 2017-04-27 | ソニー株式会社 | 画像符号化装置と画像符号化方法およびプログラム |
AR092786A1 (es) * | 2012-01-09 | 2015-05-06 | Jang Min | Metodos para eliminar artefactos de bloque |
CN103338374B (zh) * | 2013-06-21 | 2016-07-06 | 华为技术有限公司 | 图像处理方法及装置 |
US9571837B2 (en) * | 2013-11-01 | 2017-02-14 | Broadcom Corporation | Color blending prevention in video coding |
EP3453172B1 (en) * | 2016-05-28 | 2023-08-09 | HFI Innovation Inc. | Methods and apparatuses of video data processing with conditionally quantization parameter information signaling |
EP3425911A1 (en) | 2017-07-06 | 2019-01-09 | Thomson Licensing | A method and a device for picture encoding and decoding |
CN109660825B (zh) | 2017-10-10 | 2021-02-09 | 腾讯科技(深圳)有限公司 | 视频转码方法、装置、计算机设备及存储介质 |
CN110519594B (zh) * | 2018-05-21 | 2021-04-20 | 华为技术有限公司 | 视频编码方法及装置 |
CN118233633A (zh) * | 2018-11-08 | 2024-06-21 | 交互数字Vc控股公司 | 基于块的表面的视频编码或解码的量化 |
JP2021103875A (ja) * | 2019-12-25 | 2021-07-15 | キヤノン株式会社 | 画像符号化装置及び画像復号装置、並びにそれらの制御方法及びプログラム、及び、撮像装置 |
CN114157722A (zh) * | 2020-08-17 | 2022-03-08 | 华为技术有限公司 | 一种数据传输方法及装置 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1212561A (zh) * | 1997-06-20 | 1999-03-31 | 松下电器产业株式会社 | 图象处理方法、图象处理装置、和数据存储媒体 |
CN101888550A (zh) * | 2010-06-28 | 2010-11-17 | 中兴通讯股份有限公司 | 一种slice头信息中量化参数编码方法和装置 |
CN101938657A (zh) * | 2010-10-07 | 2011-01-05 | 西安电子科技大学 | 高效视频编码中编码单元自适应划分方法 |
CN101978697A (zh) * | 2008-01-25 | 2011-02-16 | 惠普开发有限公司 | 针对基于块的编码的编码模式选择 |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1751518B (zh) | 2003-02-18 | 2010-12-01 | 诺基亚有限公司 | 图像编码方法 |
US7801383B2 (en) | 2004-05-15 | 2010-09-21 | Microsoft Corporation | Embedded scalar quantizers with arbitrary dead-zone ratios |
RU2395174C1 (ru) * | 2006-03-30 | 2010-07-20 | ЭлДжи ЭЛЕКТРОНИКС ИНК. | Способ и устройство для декодирования/кодирования сигнала видео |
CN101453643B (zh) | 2007-12-04 | 2011-05-18 | 华为技术有限公司 | 量化模式、图像编码、解码方法、编码器、解码器及系统 |
US8542730B2 (en) * | 2008-02-22 | 2013-09-24 | Qualcomm, Incorporated | Fast macroblock delta QP decision |
US8897359B2 (en) * | 2008-06-03 | 2014-11-25 | Microsoft Corporation | Adaptive quantization for enhancement layer video coding |
US20120114034A1 (en) * | 2010-11-08 | 2012-05-10 | Mediatek Inc. | Method and Apparatus of Delta Quantization Parameter Processing for High Efficiency Video Coding |
-
2011
- 2011-03-11 CN CN201110059194.9A patent/CN102685478B/zh active Active
- 2011-10-21 RU RU2013145526/08A patent/RU2566332C2/ru active
- 2011-10-21 JP JP2013556946A patent/JP5860067B2/ja active Active
- 2011-10-21 AU AU2011362462A patent/AU2011362462B2/en active Active
- 2011-10-21 BR BR112013022073-2A patent/BR112013022073B1/pt active IP Right Grant
- 2011-10-21 CA CA2828692A patent/CA2828692C/en active Active
- 2011-10-21 WO PCT/CN2011/081102 patent/WO2012122798A1/zh active Application Filing
- 2011-10-21 EP EP11861079.9A patent/EP2672707A4/en not_active Withdrawn
- 2011-10-21 KR KR1020137024746A patent/KR101581097B1/ko active IP Right Grant
-
2013
- 2013-09-11 US US14/024,047 patent/US9571829B2/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1212561A (zh) * | 1997-06-20 | 1999-03-31 | 松下电器产业株式会社 | 图象处理方法、图象处理装置、和数据存储媒体 |
CN101978697A (zh) * | 2008-01-25 | 2011-02-16 | 惠普开发有限公司 | 针对基于块的编码的编码模式选择 |
CN101888550A (zh) * | 2010-06-28 | 2010-11-17 | 中兴通讯股份有限公司 | 一种slice头信息中量化参数编码方法和装置 |
CN101938657A (zh) * | 2010-10-07 | 2011-01-05 | 西安电子科技大学 | 高效视频编码中编码单元自适应划分方法 |
Cited By (30)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2018152884A (ja) * | 2011-01-13 | 2018-09-27 | キヤノン株式会社 | 画像符号化装置、画像符号化方法及びプログラム、画像復号装置、画像復号方法及びプログラム |
US10419761B2 (en) | 2011-01-24 | 2019-09-17 | Sony Corporation | Image decoding device, image encoding device, and method thereof |
JP7507944B2 (ja) | 2011-06-21 | 2024-06-28 | ドルビー ラボラトリーズ ライセンシング コーポレイション | クアッドツリー構造に基づく適応的量子化パラメータ符号化及び復号化方法及び装置 |
JP7344343B2 (ja) | 2011-06-21 | 2023-09-13 | ドルビー ラボラトリーズ ライセンシング コーポレイション | クアッドツリー構造に基づく適応的量子化パラメータ符号化及び復号化方法及び装置 |
USRE49330E1 (en) | 2011-06-21 | 2022-12-06 | Dolby Laboratories Licensing Corporation | Method and apparatus for adaptively encoding and decoding a quantization parameter based on a quadtree structure |
JP2022120012A (ja) * | 2011-06-21 | 2022-08-17 | ドルビー ラボラトリーズ ライセンシング コーポレイション | クアッドツリー構造に基づく適応的量子化パラメータ符号化及び復号化方法及び装置 |
JP7087169B2 (ja) | 2011-06-21 | 2022-06-20 | ドルビー ラボラトリーズ ライセンシング コーポレイション | クアッドツリー構造に基づく適応的量子化パラメータ符号化及び復号化方法及び装置 |
JP2022032027A (ja) * | 2011-06-21 | 2022-02-24 | ドルビー ラボラトリーズ ライセンシング コーポレイション | クアッドツリー構造に基づく適応的量子化パラメータ符号化及び復号化方法及び装置 |
JP2020065290A (ja) * | 2011-06-21 | 2020-04-23 | インテレクチュアル ディスカバリー カンパニー リミテッド | クアッドツリー構造に基づく適応的量子化パラメータ符号化及び復号化方法及び装置 |
US9912950B2 (en) | 2011-11-04 | 2018-03-06 | Infobridge Pte. Ltd. | Method of deriving quantization parameter with differential and predicted quantization parameters |
JP2017130947A (ja) * | 2011-11-04 | 2017-07-27 | インフォブリッジ ピーティーイー. エルティーディー. | 量子化ブロック逆量子化方法 |
JP2016036178A (ja) * | 2011-11-04 | 2016-03-17 | インフォブリッジ ピーティーイー. エルティーディー. | 量子化ブロック逆量子化方法 |
US11825092B2 (en) | 2011-11-04 | 2023-11-21 | Gensquare Llc | Method of deriving quantization parameter with differental and predicted quantization parameters |
US9699460B2 (en) | 2011-11-04 | 2017-07-04 | Infobridge Pte. Ltd. | Method of deriving quantization parameter with differential and predicted quantization parameters |
JP2014534749A (ja) * | 2011-11-04 | 2014-12-18 | インフォブリッジ ピーティーイー. エルティーディー. | 量子化ブロック逆量子化方法 |
JP2018078636A (ja) * | 2011-11-04 | 2018-05-17 | インフォブリッジ ピーティーイー. エルティーディー. | 量子化ブロック逆量子化方法 |
US9712825B2 (en) | 2011-11-04 | 2017-07-18 | Infobridge Pte. Ltd. | Method of deriving quantization parameter with differental and predicted quantization parameters |
JP2019071668A (ja) * | 2011-11-04 | 2019-05-09 | インフォブリッジ ピーティーイー. エルティーディー. | 量子化ブロック逆量子化方法 |
US10313671B2 (en) | 2011-11-04 | 2019-06-04 | Infobridge Pte. Ltd. | Method for generating intra prediction block with most probable mode |
US11290719B2 (en) | 2011-11-04 | 2022-03-29 | Infobridge Pte. Ltd. | Method for generating intra prediction block with most probable mode |
US9712824B2 (en) | 2011-11-04 | 2017-07-18 | Infobridge Pte. Ltd. | Method of deriving quantization parameter with differential and predicted quantization parameters |
US10742983B2 (en) | 2011-11-04 | 2020-08-11 | Infobridge Pte. Ltd. | Method for generating intra prediction block with most probable mode |
JP2017175649A (ja) * | 2011-12-21 | 2017-09-28 | 株式会社Jvcケンウッド | 動画像復号装置、動画像復号方法及び動画像復号プログラム |
JP2017169234A (ja) * | 2011-12-21 | 2017-09-21 | 株式会社Jvcケンウッド | 動画像復号装置、動画像復号方法及び動画像復号プログラム |
JP2017175650A (ja) * | 2011-12-21 | 2017-09-28 | 株式会社Jvcケンウッド | 動画像復号装置、動画像復号方法及び動画像復号プログラム |
JP2017200193A (ja) * | 2011-12-21 | 2017-11-02 | 株式会社Jvcケンウッド | 動画像符号化装置、動画像符号化方法及び動画像符号化プログラム |
JP2017079483A (ja) * | 2011-12-21 | 2017-04-27 | 株式会社Jvcケンウッド | 動画像符号化装置、動画像符号化方法及び動画像符号化プログラム |
JP2017063489A (ja) * | 2011-12-21 | 2017-03-30 | 株式会社Jvcケンウッド | 動画像復号装置、動画像復号方法及び動画像復号プログラム |
JP2017188927A (ja) * | 2011-12-21 | 2017-10-12 | 株式会社Jvcケンウッド | 動画像符号化装置、動画像符号化方法及び動画像符号化プログラム |
JP2017188928A (ja) * | 2011-12-21 | 2017-10-12 | 株式会社Jvcケンウッド | 動画像符号化装置、動画像符号化方法及び動画像符号化プログラム |
Also Published As
Publication number | Publication date |
---|---|
BR112013022073B1 (pt) | 2021-12-07 |
EP2672707A4 (en) | 2014-03-05 |
RU2013145526A (ru) | 2015-04-20 |
JP5860067B2 (ja) | 2016-02-16 |
RU2566332C2 (ru) | 2015-10-20 |
US20140079117A1 (en) | 2014-03-20 |
AU2011362462B2 (en) | 2015-03-12 |
KR20130132613A (ko) | 2013-12-04 |
EP2672707A1 (en) | 2013-12-11 |
CN102685478A (zh) | 2012-09-19 |
AU2011362462A1 (en) | 2013-10-17 |
BR112013022073A2 (pt) | 2016-11-29 |
US9571829B2 (en) | 2017-02-14 |
CN102685478B (zh) | 2015-04-29 |
CA2828692C (en) | 2017-04-25 |
CA2828692A1 (en) | 2012-09-20 |
JP2014509150A (ja) | 2014-04-10 |
KR101581097B1 (ko) | 2015-12-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2012122798A1 (zh) | 编码方法以及装置、解码方法以及装置 | |
US11277630B2 (en) | Method of coding and decoding images, coding and decoding device and computer programs corresponding thereto | |
US20220312006A1 (en) | Cross-component adaptive loop filter for chroma | |
US9414086B2 (en) | Partial frame utilization in video codecs | |
CN110650349B (zh) | 一种图像编码方法、解码方法、编码器、解码器及存储介质 | |
KR20140110008A (ko) | 객체 검출 정보에 따른 인코딩 | |
CN102685483B (zh) | 解码方法 | |
CN102685485B (zh) | 编码方法以及装置、解码方法以及装置 | |
WO2022198465A1 (zh) | 一种编码方法及装置 | |
WO2021263251A1 (en) | State transition for dependent quantization in video coding | |
US11606574B2 (en) | Efficient coding of source video sequences partitioned into tiles | |
CN102685484B (zh) | 编码方法以及装置、解码方法以及装置 | |
KR20150096353A (ko) | 이미지 인코딩 시스템, 디코딩 시스템 및 그 제공방법 | |
CN112449184B (zh) | 变换系数优化方法、编解码方法、装置、介质及电子设备 | |
US11159826B2 (en) | Method for encoding and decoding images, encoding and decoding device, and corresponding computer programs | |
KR20150029398A (ko) | 이미지 인코딩 시스템, 디코딩 시스템 및 그 제공방법 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 11861079 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2828692 Country of ref document: CA |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2011861079 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 2013556946 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 20137024746 Country of ref document: KR Kind code of ref document: A |
|
ENP | Entry into the national phase |
Ref document number: 2013145526 Country of ref document: RU Kind code of ref document: A |
|
ENP | Entry into the national phase |
Ref document number: 2011362462 Country of ref document: AU Date of ref document: 20111021 Kind code of ref document: A |
|
REG | Reference to national code |
Ref country code: BR Ref legal event code: B01A Ref document number: 112013022073 Country of ref document: BR |
|
ENP | Entry into the national phase |
Ref document number: 112013022073 Country of ref document: BR Kind code of ref document: A2 Effective date: 20130829 |