WO2012120910A1

WO2012120910A1 - Moving image coding device and moving image coding method

Info

Publication number: WO2012120910A1
Application number: PCT/JP2012/001671
Authority: WO
Inventors: 安倍　清史; 一仁木村; 秀之大古瀬; 荒川　博; 耕治有村; 重里　達郎
Original assignee: パナソニック株式会社
Priority date: 2011-03-10
Filing date: 2012-03-09
Publication date: 2012-09-13
Also published as: JP2014103429A

Abstract

A moving image coding device (100) generates a coefficient code string by variable-length coding a residual coefficient, generates a header code string including at least prediction information used when a prediction image is generated, and outputs an intermediate code string configured from the coefficient code string and the header code string in a binarization coding unit, and arithmetically codes the intermediate code string outputted by the binarization coding unit to generate an output code string in an arithmetic coding unit. When outputting the intermediate code string to the arithmetic coding unit, the binarization coding unit limits the code quantity of the intermediate code string to a predefined specific code quantity or less.

Description

Moving picture coding apparatus and moving picture coding method

The present invention relates to a moving image encoding apparatus and a moving image encoding method for encoding an input moving image by dividing it into blocks.

In recent years, with the development of multimedia applications, it has become common to handle all media information such as images, sounds and texts in a unified manner. Also, since a digitized image has a huge amount of data, an image information compression technique is indispensable for storage and transmission. On the other hand, in order to interoperate compressed image data, standardization of compression technology is also important. For example, as a standard for image compression technology, H.264 of ITU-T (International Telecommunication Union, Telecommunication Standardization Division). 261, H.H. 263, H.M. H.264, ISO / IEC (International Organization for Standardization) MPEG-1, MPEG-3, MPEG-4, MPEG-4AVC, and the like. At present, standardization activities for a next-generation screen coding method called HEVC in cooperation with ITU-T and ISO / IEC are in progress.

In such moving picture coding, each picture to be coded is divided into coding unit blocks, and the amount of information is compressed by reducing redundancy in the time direction and the spatial direction for each block. In inter-frame predictive coding for the purpose of reducing temporal redundancy, motion is detected and a predicted image is created in block units with reference to the front or rear picture, and the resulting predicted image and encoding target are obtained. A difference image from the input image of the block is acquired. In addition, in the intra prediction encoding for the purpose of reducing spatial redundancy, a prediction image is generated from pixel information of surrounding encoded blocks, and the obtained prediction image and a block to be encoded are input. A difference image from the image is acquired. Further, orthogonal transformation such as discrete cosine transformation and quantization are performed on the obtained difference image, and a code string is generated using variable length coding, thereby compressing the information amount. At this time, a coding method called CAVLC that generates a code string only by variable-length coding, and first, an intermediate code string is generated by variable-length coding (binarization), and an arithmetic code is further generated for the intermediate code string There is an encoding method called CABAC that generates a code string by performing encoding.

In decoding, prediction information and residual coefficient information are obtained by analyzing the code string generated by the encoding process, and prediction is performed by performing inter-screen prediction decoding and intra-screen prediction decoding using the prediction information. An image is generated, a difference image is generated by performing inverse quantization and inverse orthogonal transform on the residual coefficient information, and a final output image is restored by adding the obtained predicted image and the difference image.

H. In H.264 (Non-Patent Document 1), in order to restrict the upper limit of the processing amount in block units, the maximum value of the generated code amount in block units is defined (specifically, 3200 bits). If the normal encoding process described above is performed, a code string exceeding the maximum value of the generated code amount may be generated depending on the properties of the input image and the conditions of the quantization process. By using the encoding mode, it is possible to always keep within the maximum value.

Unlike the normal encoding mode, IPCM is a mode in which pixel values of an input image are described in a code string as they are as a bit string without generating a difference image by intra-screen / inter-screen prediction, or performing orthogonal transform / quantization. is there. Using this mode, for example, if the format of the input image is YUV 4: 2: 0 with 8 bits per pixel, the luminance component block is 16 × 16 pixels, and the two color difference component blocks are 8 × 8 pixels respectively. Therefore, the total is 384 bytes, and even if necessary information is included in the header, it can be surely stored within the maximum value of 3200 bits.

ITU-T H.264: Advanced video coding for generic audiovisual services (03/2010)

Many moving image encoding / decoding devices realize encoding / decoding processing by an integrated circuit called LSI. Such an encoding / decoding device has a configuration that enables a parallel operation called a pipeline in order to increase processing speed. Specifically, the processing proceeds simultaneously by starting the processing of the next block before the processing of one block is completed.

FIG. 13 (a) shows an example of a pipeline in encoding using CABAC. For block 1, each process of pixel reading, mode determination (determination of inter-screen prediction mode or intra-screen prediction mode), inter-screen / intra prediction, transform / quantization, and binary coding Are applied in order, and the same processing is applied to block 2 as well. At this time, the block 2 starts the processing immediately after the pixel reading of the block 1 is completed, thereby performing the processing in parallel while delaying the processing timing by one step. Arithmetic coding is performed at an independent timing different from the pipeline. H. In the H.264 or HEVC encoding / decoding process, the process is performed while referring to the information of blocks previously encoded / decoded. Therefore, as shown in the figure, the block 2 needs to perform processing while referring to the prediction information, pixel information, encoding information, etc. determined in the block 1.

However, it cannot be determined whether or not the generated code amount in units of blocks is within the maximum value unless the code amount at the time of completion of arithmetic coding is checked. Therefore, if it is determined that the maximum value is exceeded, the code string must be regenerated by switching to the IPCM at that time.

Fig. 13 (b) shows an example of a pipeline when switching to IPCM occurs. Assume that switching to the IPCM is determined in the arithmetic coding process of block 1. However, at this time, the encoding process of block 2 has already progressed while referring to prediction information, pixel information, and the like when block 1 performs normal encoding. For this reason, it is necessary to return to the mode determination of block 1 and update the information to be referred to by replacing the block 1 with the assumption that the block 1 has been encoded with the IPCM, and repeat the processing of the block 2. Moreover, the arithmetic encoding process is performed at an independent timing different from the pipeline in which the other processes are performed. For this reason, encoding processing of a large number of subsequent blocks has already progressed at the time when arithmetic coding of block 1 is performed, and a huge amount of processing may have to be traced.

For reference, FIG. 14A and FIG. 14B show examples of pipelines in encoding using CAVLC. In CAVLC, there is no arithmetic coding process. Therefore, even when switching to IPCM occurs, the amount of processing going back in the pipeline is constant, and the control is simple compared to CABAC.

As described above, when the IPCM occurs in the encoding using CABAC, the control going back in the pipeline requires very complicated processing control. In addition, if the number of IPCMs that occur in the target picture increases and the number of retroactive times increases, processing speed is delayed, and the encoding process of the target picture cannot be completed within the required time.

On the other hand, FIG. 15A shows an example of a pipeline in decoding using CABAC. Code block analysis, inverse transform / quantization, inter-screen / intra prediction, and output image generation are applied to block 1 in sequence, and similar processing is applied to block 2 as well. At this time, the block 2 starts the processing immediately after the code string analysis of the block 1 is completed, thereby performing the processing in parallel while delaying the processing timing by one step. Arithmetic decoding is performed at a timing independent from the pipeline.

H. In H.264, as described above, the maximum code amount of a code string in block units is defined. However, only the code amount of the code string before arithmetic decoding is limited, and the code amount of the intermediate code string generated as a result of arithmetic decoding is not defined. Therefore, the code amount of the intermediate code string may become extremely large depending on conditions.

FIG. 15B shows an example of a pipeline when the code amount of the intermediate code string becomes extremely large in the block 2. In the code string analysis process of block 2, many intermediate code strings must be read and processed. For this reason, the time required for processing has increased compared to other processing. At this time, the processing of other blocks operating in parallel also cannot proceed to the next processing until the processing of the code string analysis of block 2 is completed.

Thus, if the code amount of the intermediate code string becomes large in decoding using CABAC, the processing time of the code string analysis process increases, causing a delay in the processing speed of the entire decoding process, and the required time. It becomes impossible to complete the decoding process of the target picture.

SUMMARY OF THE INVENTION The present invention solves the above-described problem, and in a coding apparatus corresponding to arithmetic coding having a pipeline structure, simplifies the process of tracing the pipeline in order to keep the generated code amount below a limit value. In a decoding apparatus that supports arithmetic decoding with a pipeline structure at the same time, the time required for the code string analysis process to read the intermediate code string after arithmetic decoding is always less than or equal to a certain time, It is an object of the present invention to provide a new control method that can avoid the occurrence of a line processing delay and replace the conventional method.

The moving image encoding device according to the first aspect of the present invention is a moving image encoding device that encodes an input encoding target image in units of blocks, and generates a predicted image corresponding to the encoding target image. A prediction image generation unit, a subtracter that generates a difference image between the encoding target image and the generated prediction image, and an orthogonal transform process and a quantization process are performed on the output of the subtractor to generate a residual coefficient Generated by a prediction residual encoding unit, a prediction residual decoding unit that generates a residual decoded image by performing inverse quantization processing and inverse orthogonal transform processing on the residual coefficient, and a prediction image generation unit An adder that generates a reconstructed image by adding the prediction image and the residual decoded image generated by the prediction residual decoding unit, and a variable coefficient for the residual coefficient generated by the prediction residual encoding unit When generating a coefficient code string by encoding and generating a prediction image A header code string including at least the prediction information used, and a binary code part that outputs a coefficient code string and an intermediate code string composed of the header code string and the binary code part An arithmetic encoding unit that arithmetically encodes the intermediate code sequence and generates an output code sequence, and the binarization encoding unit outputs the intermediate code sequence when outputting the intermediate code sequence to the arithmetic encoding unit. The code amount is limited to a predetermined specific code amount or less.

The moving image encoding device according to the second aspect of the present invention is a moving image encoding device that encodes an input encoding target image in units of blocks, and generates a predicted image corresponding to the encoding target image. A prediction image generation unit, a subtracter that generates a difference image between the encoding target image and the generated prediction image, and an orthogonal transform process and a quantization process are performed on the output of the subtractor to generate a residual coefficient Generated by a prediction residual encoding unit, a prediction residual decoding unit that generates a residual decoded image by performing inverse quantization processing and inverse orthogonal transform processing on the residual coefficient, and a prediction image generation unit An adder that generates a reconstructed image by adding the prediction image and the residual decoded image generated by the prediction residual decoding unit, and a residual generated by the prediction residual encoding unit in the first mode The difference coefficient is variable-length encoded to generate a coefficient code string, and A header code sequence including at least prediction information used when generating an image is generated and an intermediate code sequence composed of a coefficient code sequence and a header code sequence is output. In the second mode, instead of a residual coefficient The intermediate image signal obtained in the signal processing process of the encoding target image is directly used as a coefficient code string without variable length coding, and at least an identifier indicating that the coefficient code string and the coefficient code string are intermediate image signals A binary encoding unit that outputs an intermediate code string including a header code string including the arithmetic code for the intermediate code string output in the first mode to generate an output code string; In the intermediate code sequence output in step 1, only the header code sequence is arithmetically encoded to generate an output code sequence composed of the header code sequence and the coefficient code sequence after the arithmetic encoding. The binary encoding unit, when outputting the intermediate code sequence to the arithmetic encoding unit, limits the code amount of the intermediate code sequence to a predetermined code amount or less. .

Note that the present invention can also realize processing equivalent to each means included in such a moving image coding apparatus as a program or an integrated circuit.

According to the moving picture coding apparatus in each aspect of the present invention, even when an encoding method including arithmetic coding is realized using a pipeline structure, a specific code defined in advance at the time of generating an intermediate code string The amount can be controlled to be less than the amount. Therefore, it is possible to simplify the process of tracing back the pipeline in order to keep the generated code amount below the limit value. Also, in the corresponding video decoding device, since the code amount of the intermediate code string is limited to a predetermined specific code amount or less, in order to read the intermediate code string after arithmetic decoding in the code string analysis process It is possible to always set the time required for the time to a certain time or less. Therefore, it is possible to avoid the occurrence of pipeline processing delay in the video decoding process.

In addition, in the conventional moving image encoding device, it is necessary to use a high-capacity element in the entire encoding process in order to realize a process that goes back through a complicated pipeline. However, in the present invention, the process of going back the pipeline is simplified by limiting the code amount of the intermediate code string to a certain value or less, so the processing is realized even with an element having a lower capability than the conventional one in the moving picture coding apparatus. It becomes possible to do.
Further, in the conventional moving picture decoding apparatus, in order to reduce the occurrence of processing delay such as code string analysis processing, it is necessary to use a high-capacity element for a portion that performs code string analysis processing or the like. However, in the present invention, by limiting the code amount of the intermediate code string to a certain value or less, a delay hardly occurs in the video decoding device in the first place. Therefore, it is not necessary to use an element with a particularly high capability in the moving picture decoding apparatus, and the moving picture decoding apparatus can be simply configured.

It is a block diagram which shows the structure of the moving image encoder which concerns on Embodiment 1. FIG. 3 is a flowchart of a code string generation process according to the first embodiment. 6 is a conceptual diagram for explaining an example of a syntax of a code string generated according to Embodiment 1. FIG. It is a conceptual diagram for demonstrating the pipeline control of the moving image encoder which concerns on Embodiment 1. FIG. 6 is a flowchart of a code string generation process according to another example of the first embodiment. It is a block diagram which shows the structure of the moving image encoder which concerns on Embodiment 2. FIG. 10 is a flowchart of a code string generation process according to the second embodiment. 10 is a conceptual diagram for explaining an example of a syntax of a code string generated by Embodiment 2. FIG. It is a conceptual diagram for demonstrating the pipeline control of the moving image encoder which concerns on Embodiment 2. FIG. It is a block diagram which shows the structure of the moving image decoding apparatus which concerns on Embodiment 3, 4. FIG. 10 is a flowchart of code string analysis processing according to the third embodiment. 10 is a flowchart of code string analysis processing according to the fourth embodiment. It is a conceptual diagram for demonstrating the pipeline control of the conventional moving image encoder. It is a conceptual diagram for demonstrating another pipeline control of the conventional moving image encoder. It is a conceptual diagram for demonstrating the pipeline control of the conventional moving image decoding apparatus.

(Embodiment 1)
Hereinafter, Embodiment 1 will be described with reference to the drawings.

1. Configuration of Video Encoding Device FIG. 1 is a block diagram illustrating a configuration of a video encoding device 100 according to the first embodiment. The moving image encoding apparatus 100 divides a moving image input in units of pictures into blocks, performs an encoding process in units of blocks, and generates a code string.

The moving picture encoding apparatus 100 includes a picture memory 101, a prediction residual encoding unit 102, a prediction residual decoding unit 103, a local buffer 104, a prediction encoding unit 105, and a quantization value determination unit 106. And a binary encoding unit 107 and an arithmetic encoding unit 110.

The picture memory 101 stores the input image signal 151 input in units of pictures in the order of display by rearranging the pictures in the order of encoding. Next, when the picture memory 101 receives a read command from the difference calculation unit 111 or the predictive coding unit 105, the picture memory 101 outputs an input image signal related to the read command. At this time, each picture is divided into coding units composed of a plurality of pixels called coding units (hereinafter referred to as CU). The CU is, for example, a horizontal 64 × vertical 64 pixel block, a horizontal 32 × vertical 32 pixel block, a horizontal 16 × vertical 16 pixel block, or the like. That is, any configuration may be used as long as it is a group of pixels composed of a plurality of pixels. As described above, the block size is not limited to a square shape, and may be a rectangular block size. In the moving picture encoding apparatus 100 according to the present embodiment, the subsequent processing is performed in units of CUs.

The prediction residual encoding unit 102 performs orthogonal transformation on the difference image signal 152 output from the difference calculation unit 111. Further, the prediction residual encoding unit 102 compresses the image information by performing quantization on the obtained orthogonal transform coefficient of each frequency component, and generates a residual encoded signal 153. Then, the generated residual encoded signal 153 is output to the prediction residual decoding unit 103 and the coefficient code string generation unit 109. At this time, the prediction residual encoding unit 102 quantizes the orthogonal transform coefficient using the quantized value signal 158 determined by the quantized value determining unit 106.

The prediction residual decoding unit 103 generates a residual decoded signal 154 by performing inverse quantization and inverse orthogonal transform on the residual encoded signal 153 output from the prediction residual encoding unit 102. Then, the generated residual decoded signal 154 is output to the addition operation unit 112.

The local buffer 104 stores the reconstructed image signal 155 output from the addition operation unit 112. The reconstructed image signal 155 is used for predictive coding processing in coding of a picture subsequent to a picture that is currently being coded. That is, the reconstructed image signal 155 is referred to as pixel data when a picture subsequent to the current picture to be coded is coded. The local buffer 104 outputs the stored reconstructed image signal 155 to the prediction encoding unit 105 as pixel data in response to a read command from the prediction encoding unit 105.

The predictive encoding unit 105 generates a predictive image signal 156 using intra prediction or inter prediction based on the input image signal output from the picture memory 101. Then, the predictive encoding unit 105 outputs the generated predicted image signal 156 to the difference calculation unit 111 and the addition calculation unit 112. Note that the prediction encoding unit 105 uses the reconstructed image signal 155 of a past picture that has already been encoded and stored in the local buffer 104 when using inter-screen prediction. When using intra prediction, the reconstructed image signal 155 of the current picture of an already encoded CU adjacent to the encoding target CU is used. The mode determination method of using intra-screen prediction or inter-screen prediction is performed by predicting which prediction method can reduce the information amount of the residual signal.

The quantization value determination unit 106 sets a quantization value when the difference image signal 152 is quantized by the prediction residual encoding unit 102 based on information such as a picture stored in the picture memory 101. Then, the set quantization value is output to prediction residual encoding section 102 and header code string generation section 107. As a quantization value setting method in the quantization value determination unit 106, a quantization value is set so that the bit rate of the code string signal 160 approaches a target bit rate, so-called quantization based on rate control. A value setting method or the like may be used. Note that the information for determining the quantization value in the above may include, for example, a virtual buffer retention amount, a preset bit rate, and the like.

The binarization encoding unit 107 includes a header code string generation unit 108 and a coefficient code string generation unit 109. The binarization information generated by applying each process is used as an intermediate code string 159 as an arithmetic encoding unit. To 110.

The header code string generation unit 108 is a variable length code for the prediction information signal 157 output from the prediction encoding unit 105, the quantization value signal 158 output from the quantization value determination unit 106, and other control information related to encoding control. To generate a header intermediate code string. Note that the prediction information included in the prediction information signal 157 includes, for example, information indicating an intra prediction mode, information indicating an inter prediction mode, information indicating a motion vector, information indicating a reference picture, and the like. Further, the control information is information that can be acquired before the processing in the coefficient code string generation unit 109, and is information that indicates the encoding condition applied at the time of CU encoding. For example, information indicating a block encoding type, block Division information and the like are included.

The coefficient code string generation unit 109 performs variable-length coding on the residual encoded signal 153 output from the prediction residual encoding unit 102 to generate a coefficient intermediate code string. Then, the coefficient code string generation unit 109 generates the intermediate code string signal 159 by appending the generated coefficient intermediate code string after the header intermediate code string generated by the header code string generation unit 108 (hereinafter referred to as the first code string signal 159). Called mode).

On the other hand, the coefficient code string generation unit 109 converts the coefficient intermediate code string obtained without variable length encoding of the input image signal output from the picture memory 101 into the header intermediate code string generated by the header code string generation unit 108. Subsequently, the intermediate code string signal 159 is generated by additionally writing (hereinafter referred to as the second mode).

The arithmetic encoding unit 110 performs arithmetic encoding on a part of the intermediate code sequence 159 output from the binary encoding unit 107, and outputs a code that is output from the video encoding device 100 A column signal 160 is generated.

The difference calculation unit 111 generates a difference image signal 152 that is a difference value between the input image signal read from the picture memory 101 and the prediction image signal 156 that is the output of the prediction encoding unit 105, and generates a prediction residual code. To the conversion unit 102.

The addition operation unit 112 adds the residual decoded signal 154 output from the prediction residual decoding unit 103 and the prediction image signal 156 output from the prediction encoding unit 105 to add the reconstructed image signal 155. Generated and output to the local buffer 104 and the prediction encoding unit 105.

2. Code String Signal Generation Method A method of generating the code string signal 160 in the binary coding section 107 and the arithmetic coding section 110 will be specifically described with reference to the flowchart of FIG.

First, the binary encoding unit 107 uses the input residual encoded signal 153 to determine whether or not the intermediate code amount after binarization of the encoding target CU may exceed a specified value. Determination is made (S601).

If it is determined in step S601 that there is no possibility of exceeding, the prediction information signal 157, the quantized value signal 158, and other coding control information generated as a result of performing the normal coding process described above are input. To generate a header intermediate code string by performing variable length coding (S602). Further, similarly to the conventional coding, the input residual coded signal 153 is variable-length coded (Residual mode) to generate a coefficient intermediate code string (S603).

On the other hand, if it is determined in step S601 that there is a possibility of exceeding, a header intermediate code string is generated by variable-length encoding only PCM mode information indicating that encoding has been performed in the PCM mode (S605). . Further, a coefficient intermediate code string is generated by describing the input image signal as it is in the code string as it is without variable length coding (PCM mode) (S606).

Subsequently, when the input intermediate code sequence is generated in the Residual mode, the arithmetic encoding unit 110 performs final encoding by performing arithmetic encoding on the header intermediate code sequence and the coefficient intermediate code sequence. A code string is generated (S604). When the input intermediate code sequence is generated by the PCM mode, arithmetic encoding is performed on the header intermediate code sequence including only the PCM mode information, and arithmetic encoding is performed on the coefficient intermediate code sequence. A final code string is generated by adding the code string as it is without performing the process (S607).

In step S601, it is determined whether there is a possibility that the intermediate code amount of the encoding target CU exceeds the specified value using the input residual encoded signal 153, but other methods are used. It may be determined whether there is a possibility that the intermediate code amount exceeds the specified value. For example, there is a method of determining whether the code amount exceeds a predetermined value based on the generated intermediate code string signal 159. In this case, the determination is made when the processing of step S602 and step S603 is completed, and the intermediate code string has already been generated in the Residual mode. For this reason, if it is determined that the number has been exceeded, the intermediate code string generated is replaced with the intermediate code string regenerated in the PCM mode by performing the processing in steps S605 and S606 instead of the generated intermediate code string.

Note that an input image signal is input to the coefficient code string generation unit 109, and in step S606, the input image signal is input to generate a coefficient intermediate code string in the PCM mode. However, instead of the input image signal, a reconstructed image is generated. A signal intermediate code string may be generated in the PCM mode by inputting the signal 155. Alternatively, the coefficient intermediate code sequence may be generated in the PCM mode by inputting the difference image signal 152 or the residual decoded signal 154.

Further, although the determination is performed in units of CUs in step S601, the determination may be performed in units of a plurality of CUs, blocks smaller than the CU, or other block units.

3. Syntax FIG. 3 is a diagram showing an example of the syntax: coding_unit () of the CU unit of the intermediate code string generated by the present embodiment.

FIG. 3A shows the syntax when the intermediate code string is generated in the Residual mode described with reference to FIG. At the beginning of the syntax, a code sequence generated by the header code sequence generation unit 108, which is variable length encoded information such as prediction mode: pred_mode, prediction information: prediction_unit (), quantization value: qp_value, is described. Yes. Subsequently, a code string obtained by variable-length encoding the residual encoded signal 153 is described as residual_data ().
The same applies to the syntax when the intermediate code string is generated in the Residual mode described in FIG.

FIG. 3B shows the syntax when the intermediate code string is generated in the PCM mode described with reference to FIG. At the beginning of the syntax, a code string obtained by variable-length encoding pred_mode, which is PCM mode information indicating that encoding is performed in the PCM mode, is described. Subsequently, in pcm_data (), the input image signal is described as a code string that is not subjected to variable-length encoding but is a pixel bit string as it is.

At this time, the code amount of the generated intermediate code string cannot be uniquely specified in the Residual mode because the code length varies depending on the variable length coding conditions. However, the PCM mode can be uniquely specified by the size of the CU except for pred_mode. For example, if the image format is YUV 4: 2: 0 with 8 bits for each pixel and the size of the encoding target CU is 32 × 32 pixels, the pixel value of the input image signal or the reconstructed image signal 155 is directly encoded. The amount of code required for describing as a column is 1536 bytes. Therefore, even if the maximum code amount and margin that can be generated when pred_mode is variable-length encoded, the entire CU does not exceed 12500 bits.

That is, the specified value described in step S601 in FIG. 2 indicates that the pixel value of the input image signal of the encoding target CU is encoded as it is as a code string and is encoded in the PCM mode. By setting the maximum code amount that can be generated when information is variable-length encoded and the margin amount (12500 bits in the above example), the generated intermediate code amount always uses the specified value. Guaranteed not to exceed.

In the above description of the syntax, the input image signal is described in pcm_data (), but the reconstructed image signal 155, the difference image signal 152, and the residual decoded signal 154 are input to the coefficient code string generation unit 109. If so, the reconstructed image signal 155, the difference image signal 152, and the residual decoded signal 154 are described in accordance with the input signal.

It should be noted that the syntax described in FIG. 3 and the numerical values used in the description of the specified values are examples for explaining the present embodiment, and the syntax and numerical values different from those described here are used. May be used to implement similar functions.

4). Pipeline Improvement Effect An example of a pipeline of the moving picture coding apparatus 100 according to the present embodiment will be shown using FIG.

FIG. 4A is a diagram illustrating pipeline control when an intermediate code string is generated in the Residual mode as a result of the determination in step S601 of FIG. Processing is performed in exactly the same flow as in the conventional control described with reference to FIG.

On the other hand, FIG. 4B is a diagram illustrating pipeline control when an intermediate code string is generated in the PCM mode as a result of the determination in step S601 of FIG. Unlike the conventional control described with reference to FIG. 13B, it can be determined that the PCM mode should be set at the time of the binary encoding process before the arithmetic encoding process. Conventionally, since it is determined that the PCM mode should be set at the time of arithmetic coding processing, a huge amount of retroactive processing has occurred in the previous pipeline. However, in this embodiment, since it is possible to determine that the PCM mode should be set at the time of the binary coding process, it is possible to reduce the amount of processing going back in the pipeline and make it constant. The process control can be simplified.

In addition, the pipeline processing control in FIGS. 4A and 4B is controlled by the pipeline processing in the case of encoding using the CAVLC described in FIGS. 14A and 14B. It is exactly the same. That is, the moving picture coding apparatus 100 according to the present embodiment performs processing using exactly the same pipeline control method whether coding is performed using CABAC or coding is performed using CAVLC. It becomes possible to do.

Furthermore, the code sequence generated by the moving picture encoding apparatus 100 according to the present embodiment is guaranteed that the generated code amount of the intermediate code sequence is equal to or less than the specific code amount. For this reason, in the decoding pipeline described with reference to FIG. 15, the code amount of the intermediate code string read by the code string analysis process is less than or equal to the specific amount, and the processing time of the code string analysis process is always as shown in FIG. Can be kept within the prescribed value, and the occurrence of a delay in the decoding process can be avoided.

As described above, the moving picture coding apparatus 100 according to the present embodiment is not limited to the identification that is defined in advance at the time of generating the intermediate code string even when the coding method using arithmetic coding is realized using the pipeline structure. It is possible to control so as to be less than or equal to the code amount. Therefore, it is possible to simplify the process of tracing back the pipeline in order to keep the generated code amount below the limit value. Furthermore, in the decoding device that decodes the generated code string, the time required for the code string analysis process to read the intermediate code string after arithmetic decoding can always be less than or equal to a certain time. It is possible to avoid the occurrence of a pipeline processing delay in the processing apparatus.

In addition, in the conventional moving image encoding apparatus, it is necessary to use a high-performance element in the entire encoding process in order to realize a process that goes back through a complicated pipeline. However, the present invention simplifies the process of going back through the pipeline by limiting the code amount of the intermediate code string to a certain value or less, so the processing is realized even with an element having a lower capacity than the conventional one in the moving picture coding apparatus. It becomes possible to do.
Further, in the conventional moving picture decoding apparatus, in order to reduce the occurrence of processing delay such as code string analysis processing, it is necessary to use a high-capacity element for a portion that performs code string analysis processing or the like. However, in the present invention, by limiting the code amount of the intermediate code string to a certain value or less, a delay hardly occurs in the video decoding device in the first place. Therefore, it is not necessary to use an element having a particularly high capability in the moving picture decoding apparatus, and the moving picture decoding apparatus can be simply configured.

5. Summary The moving image encoding apparatus 100 according to the present embodiment encodes an input encoding target image in units of blocks. The moving image encoding device 100 includes a prediction encoding unit 105 that generates a prediction image corresponding to an encoding target image, a difference calculation unit 111 that generates a difference image between the encoding target image and the generated prediction image, A prediction residual encoding unit 102 that performs orthogonal transformation processing and quantization processing on the output of the difference calculation unit 111 to generate a residual coefficient, and performs inverse quantization processing and inverse orthogonal transformation processing on the residual coefficient. A prediction residual decoding unit 103 that generates a residual decoded image, and adds the prediction image generated by the prediction encoding unit 105 and the residual decoded image generated by the prediction residual decoding unit 103 In addition, in the first mode, the addition operation unit 112 that generates a reconstructed image and the residual coefficient generated by the prediction residual encoding unit 102 are variable-length encoded to generate a coefficient code string, and the prediction image is Prediction information used when generating A header code string including at least a header code string is generated and an intermediate code string 159 composed of a coefficient code string and a header code string is output. In the second mode, in the signal processing process of an encoding target image instead of a residual coefficient The obtained intermediate image signal is directly converted into a coefficient code string without variable length coding, and an intermediate code composed of the coefficient code string and a header code string including at least an identifier indicating that the coefficient code string is an intermediate image signal A binary encoding unit 107 that outputs a sequence, and arithmetically encodes the intermediate code sequence output in the first mode to generate an output code sequence, while the intermediate code sequence output in the second mode Arithmetic coding unit 110 that performs arithmetic coding only on the header code string and generates an output code string composed of the header code string and coefficient code string after arithmetic coding , Comprising a. When the binary coding unit 107 outputs the intermediate code string to the arithmetic coding unit 110, the binary coding unit 107 limits the code amount of the intermediate code string to a predetermined specific code amount or less.

Preferably, the specific code amount defined in advance includes a code amount necessary for describing the pixel value of the intermediate image signal as it is as a code string, and all information that may be described in the header code string. Is a code amount that is a combination of the maximum code amount required for encoding and a margin amount.

Preferably, the binarization encoding unit 107 uses the second mode when the code amount of the output intermediate code sequence 159 may exceed a predetermined specific code amount. Is output.

Preferably, the binarization encoding unit 107 outputs the intermediate code string 159 in the first mode, and as a result, when the code amount of the intermediate code string 159 exceeds a predetermined code amount, Instead of the intermediate code sequence 159 generated in the first mode, the intermediate code sequence 159 generated using the second mode is output.

6). Other Examples In the present embodiment, the intermediate code amount after binarization of the encoding target CU is set to a predetermined value or less by switching between the Residual mode and the PCM mode in the flowchart of FIG. However, the intermediate code amount after binarization of the encoding target CU can be limited to a specified value or less without using the PCM mode. This example will be described below. The configuration of the moving image encoding apparatus is substantially the same as the configuration shown in FIG. 1 except that an input image signal input from the picture memory 101 to the coefficient code string generation unit 109 is not necessary. The syntax is as shown in FIG. Hereinafter, in this example, a method of generating the code string signal 160 by the binary encoding unit 107 and the arithmetic encoding unit 110 will be described with reference to the flowchart of FIG.

First, the binary encoding unit 107 uses the input residual encoded signal 153 to determine whether or not the intermediate code amount after binarization of the encoding target CU may exceed a specified value. Determination is made (S1501).

If it is determined in step S1501 that there is no possibility of exceeding, the prediction information signal 157, the quantized value signal 158, and other encoding control information generated as a result of performing the normal encoding process described above are input. To generate a header intermediate code string by performing variable length coding (S1502). Furthermore, a coefficient intermediate code string is generated by variable-length encoding the input residual encoded signal 153 (Residual mode) in the same manner as in the conventional encoding (S1503).

On the other hand, if it is determined in step S1501 that there is a possibility of exceeding, the prediction residual encoding process in the prediction residual encoding unit 102 is performed again (S1505). At this time, at least the quantization value used in the quantization process is updated to a value larger than the set value (large quantization width) and applied, so that the residual encoded signal 153 to be regenerated is less information amount Control to be The process of step S1501 is performed again using the updated residual encoded signal 153, and the process of step S1505 is repeated until it is determined that there is no possibility of exceeding.

Subsequently, the arithmetic encoding unit 110 generates a final code sequence by performing arithmetic encoding on the input header intermediate code sequence and the coefficient intermediate code sequence (S1504).

Here, in step S1501, it is determined whether or not the intermediate code amount of the CU to be encoded may exceed the specified value using the input residual encoded signal 153. A method may be used to determine whether there is a possibility that the intermediate code amount exceeds a specified value. For example, there is a method of determining whether the code amount exceeds a predetermined value based on the generated intermediate code string signal 159. In this case, the determination is made when the processing of step S1502 and step S1503 is completed, and the intermediate code string has already been generated. For this reason, if it is determined that the number has been exceeded, it is replaced with the regenerated intermediate code string by performing the processing in step S1505 instead of the generated intermediate code string.

The moving image encoding apparatus 100 according to this other example encodes an input encoding target image in units of blocks. The moving image encoding device 100 includes a prediction encoding unit 105 that generates a prediction image corresponding to an encoding target image, a difference calculation unit 111 that generates a difference image between the encoding target image and the generated prediction image, An orthogonal transform process and a quantization process are performed on the output of the difference calculation unit 111 to generate a residual coefficient, and a dequantization process and an inverse orthogonal transform process are performed on the residual coefficient. A prediction residual decoding unit 103 that generates a residual decoded image, and adds the prediction image generated by the prediction encoding unit 105 and the residual decoded image generated by the prediction residual decoding unit 103 Then, the addition calculation unit 112 that generates a reconstructed image and the residual coefficient generated by the prediction residual encoding unit 102 are variable-length encoded to generate a coefficient code string, which is used when generating a predicted image Including at least A binarization encoding unit 107 that generates a code sequence and outputs an intermediate code sequence composed of a coefficient code sequence and a header code sequence, and an intermediate code sequence output by the binarization encoding unit 107 An arithmetic encoding unit 110 that performs arithmetic encoding and generates an output code string. When the binary coding unit 107 outputs the intermediate code string to the arithmetic coding unit 110, the binary coding unit 107 limits the code amount of the intermediate code string to a predetermined specific code amount or less.

According to this example, the intermediate code amount after binarization of the encoding target CU can be limited to a predetermined value or less without using the PCM mode. In particular, since the PCM mode is not used, it is not necessary to mount a processing circuit related to the PCM mode, and the circuit configuration can be simplified.

(Embodiment 2)
Next, a moving picture coding apparatus according to Embodiment 2 will be described with reference to the drawings.

1. Configuration of Moving Image Encoding Device FIG. 6 is a block diagram of a moving image encoding device 100-1 according to the second embodiment. The moving image encoding apparatus 100-1 divides a moving image input in units of pictures into blocks, performs encoding processing in units of blocks, and generates a code string.

This moving image coding apparatus 100-1 includes a binary coding unit 107, a header code string generation unit 108, a coefficient code string generation unit 109, and an arithmetic coding unit 110 of the moving image coding apparatus 100 according to the first embodiment. Instead of this, a binary coding section 107-1, a header code string generation section 108-1, a coefficient code string generation section 109-1, and an arithmetic coding section 110-1 are provided.

Hereinafter, for convenience of explanation, detailed description of the same configuration as that of the first embodiment will be omitted. Further, in FIG. 6, the same numbers are assigned to blocks having the same functions as those in FIG.

The binarization encoding unit 107-1 includes a header code sequence generation unit 108-1 and a coefficient code sequence generation unit 109-1, and converts the binarization information generated by applying each processing into an intermediate code sequence 159-1 is output to arithmetic coding section 110-1.

The header code string generation unit 108-1 changes the prediction information signal 157 output from the prediction encoding unit 105, the quantization value signal 158 output from the quantization value determination unit 106, and other control information related to encoding control. A header intermediate code string is generated by performing long encoding. Note that the prediction information includes, for example, information indicating an intra-screen prediction mode, information indicating an inter-screen prediction mode, information indicating a motion vector, information indicating a reference picture, and the like. Further, the control information is information that can be acquired before processing in the coefficient code string generation unit 109-1, and is information that indicates the encoding conditions applied at the time of CU encoding. For example, information indicating the block encoding type , Block division information and the like are included.

The coefficient code sequence generation unit 109-1 performs variable length encoding on the residual encoded signal 153 output from the prediction residual encoding unit 102 to generate a coefficient intermediate code sequence. Then, the coefficient code string generation unit 109-1 generates the intermediate code string signal 159-1 by adding the generated coefficient intermediate code string after the header intermediate code string generated by the header code string generation unit 108-1. (Hereinafter referred to as the first mode).

On the other hand, the coefficient code string generation unit 109 generates the intermediate code string generated by the header code string generation unit 108 from the coefficient intermediate code string obtained without performing variable length coding on the reconstructed image signal 155 output from the addition calculation unit 112. The code string is added after the code string to generate an intermediate code string signal 159-1 (hereinafter referred to as a second mode).

The arithmetic coding unit 110-1 performs arithmetic coding on the intermediate code string 159-1 output from the binary coding unit 107-1, and becomes an output of the moving picture coding apparatus 100-1. A code string signal 160-1 is generated.

2. Code String Generation Method A method for generating the code string signal 160-1 in the binary coding section 107-1 and the arithmetic coding section 110-1 will be specifically described with reference to the flowchart of FIG.

First, the binary encoding unit 107-1 receives the prediction information signal 157, the quantized value signal 158, and other encoding control information generated as a result of the above encoding process, and performs variable length encoding. By doing so, a header intermediate code string is generated (S1001).

Next, it is determined by using the input residual encoded signal 153 whether there is a possibility that the intermediate code amount of the encoding target CU exceeds the specified value (S1002).

If it is determined in step S1002 that there is no possibility of exceeding, the coefficient intermediate code string is obtained by variable-length encoding the residual encoded signal 153 (Residual mode) as in the conventional encoding. Generate (S1003).

On the other hand, if it is determined in step S1002 that there is a possibility of exceeding, the intermediate code is obtained by describing the input reconstructed image signal 155 as it is in the code string without performing variable length coding (PCM mode). A column is generated (S1005).

Subsequently, when the input intermediate code sequence is generated in the Residual mode, the arithmetic encoding unit 110-1 performs arithmetic encoding on the header intermediate code sequence and the coefficient intermediate code sequence. A final code string is generated (S1004). Further, when the input intermediate code string is generated by the PCM mode, arithmetic coding is performed only on the header intermediate code string, and arithmetic coding is not performed on the coefficient intermediate code string. The final code string is generated by adding the code string as it is (S1006).

Here, in step S1002, it is determined whether there is a possibility that the intermediate code amount of the encoding target CU exceeds the specified value using the input residual encoded signal 153. A method may be used to determine whether there is a possibility that the intermediate code amount exceeds a specified value. For example, there is a method of determining whether the code amount exceeds a predetermined value based on the generated intermediate code string signal 159-1. In this case, the determination is made when the process of step S1003 is completed, and the coefficient intermediate code string has already been generated in the Residual mode. Therefore, if it is determined that the number has been exceeded, the processing is performed by replacing the generated coefficient intermediate code sequence with the coefficient intermediate code sequence regenerated in the PCM mode by performing the process of step S1005. Made.

Note that the reconstructed image signal 155 is input to the coefficient code sequence generation unit 109, and the reconstructed image signal 155 is input in step S1005 to generate a coefficient intermediate code sequence in the PCM mode. Alternatively, the coefficient intermediate code string may be generated in the PCM mode using the input image signal as an input. Alternatively, the coefficient intermediate code sequence may be generated in the PCM mode with the difference image signal 152 or the residual decoded signal 154 as an input.

Note that although the determination is performed in units of CUs in step S1002 here, the determination may be performed in units of a plurality of CUs, blocks smaller than the CU, or other block units.

3. Syntax FIG. 8 is a diagram illustrating an example of the syntax: coding_unit () of the intermediate code string and the code string of the CU unit generated according to the present embodiment.

FIG. 8A shows the syntax when the intermediate code string is generated in the Residual mode described in FIG. At the top of the syntax, a code string generated by the header code string generation unit 108-1 is encoded with variable length coding of information such as prediction mode: pred_mode, prediction information: prediction_unit (), and quantization value: qp_value. Has been. Subsequently, pcm_flag indicating whether the coefficient information is encoded in the Residual mode or the PCM mode is described, and further, a code string obtained by variable-length encoding the residual encoded signal 153 is described as residual_data (). ing.

FIG. 8B shows the syntax when the intermediate code string is generated in the PCM mode described in FIG. Similarly to FIG. 8A, information such as prediction mode: pred_mode, prediction information: prediction_unit (), quantization value: qp_value, etc., generated by the header code string generation unit 108-1, is used at the beginning of the syntax. A variable-length encoded code string is described. Subsequently, pcm_flag indicating whether the encoding is performed in the Residual mode or the PCM mode is described, and further, as the pcm_data (), the reconstructed image signal 155 is not subjected to variable length encoding, and is a pixel bit string as it is. A code string is described.

Note that the code amount of the generated intermediate code string is variable-length encoded for all information in the Residual mode. For this reason, the code length varies depending on the variable length coding conditions, and therefore cannot be uniquely identified. However, in the PCM mode, pcm_data () is not variable-length coded and can be uniquely specified by the CU size. For example, when the image format is YUV 4: 2: 0 with 8 bits for each pixel and the size of the encoding target CU is 32 × 32 pixels, the pixel value of the input image signal or the reconstructed image signal is directly used as the code string. The code amount required when describing as 1536 bytes. Since the variable length coding is performed except for pcm_data (), the code length cannot be uniquely specified. However, the number of pieces of information to be encoded using variable length encoding is limited, and the maximum amount of code that can be generated is not so large. In other words, even if the maximum code amount and margin that can be generated when variable length coding is performed on anything other than pcm_data () is added to the code amount of pcm_data (), the entire CU does not exceed 13000 bits.

That is, as the specified value described in step S1002 of FIG. 7, the code amount necessary for describing the pixel value of the reconstructed image signal 155 of the encoding target CU as a code string as it is and the header information are variable-length encoded. By adding the maximum code amount that can be generated and the margin amount (13000 bits in the above example), it is guaranteed that the generated intermediate code amount does not always exceed the specified value. The

In the above description of the syntax, the reconstructed image signal 155 is described in pcm_data (), but the input image signal, the difference image signal 152, and the residual decoded signal 154 are included in the coefficient code string generation unit 109-1. Are input image signal, difference image signal 152, and residual decoded signal 154 according to the input signal.

Note that the syntax described in FIG. 8 and the numerical values used in the description of the specified values are examples for explaining the present embodiment, and the syntax and numerical values different from those described here are used. May be used to implement similar functions.

4). Pipeline Improvement Effect An example of a pipeline of the moving picture coding apparatus 100-1 according to the present embodiment will be shown using FIG.

FIG. 9A is a diagram illustrating pipeline control when an intermediate code string is generated in the Residual mode as a result of the determination in step S1002 of FIG. Processing is performed in exactly the same flow as in the conventional control described with reference to FIG.

On the other hand, FIG. 9B is a diagram illustrating pipeline control when the intermediate code string is generated using the reconstructed image signal 155 in the PCM mode as a result of the determination in step S1002 of FIG. Unlike the conventional control described with reference to FIG. 13B, it can be determined that the PCM mode should be set at the time of the binary encoding process before the arithmetic encoding process. Conventionally, since it is determined that the PCM mode should be set at the time of arithmetic coding processing, a huge amount of retroactive processing has occurred in the previous pipeline. However, in the present embodiment, it can be determined that the PCM mode should be set at the time of the binary encoding process. Here, the reconstructed image signal 155 is based on the same input image signal 151 as the residual encoded signal 153. Further, when the decoding apparatus decodes the residual encoded signal 153 using the prediction information of the signal and the set, the same signal as the reconstructed image signal 155 is obtained. That is, the pixel information of the reconstructed image signal finally generated in the decoding device is the same. For this reason, even if the block 1 is switched to the PCM mode, it is not necessary to change the prediction information described in the header code string or re-encode. For this reason, there is no influence on the processing of block 2 in which the encoding processing is proceeding while referring to them. Therefore, it is possible to generate the intermediate code string in the PCM mode without going back through the pipeline.

In addition, when processing is performed using the input image signal from the picture memory 101 instead of the reconstructed image signal 155 in step S1005 of FIG. That is, the input image signal from the picture memory 101 and the residual encoded signal 153 are based on the same input image signal 151. Therefore, even when the block 1 is switched to the PCM mode, it is not necessary to change the prediction information described in the header code string. However, when the decoding apparatus decodes the residual encoded signal 153 using the signal and a set of prediction information, a signal different from the input image signal is generated. Therefore, the pixel information of the reconstructed image signal finally obtained in the decoding apparatus differs between when encoded in the first mode and when encoded in the second mode. For this reason, the encoding process of block 2 that has already progressed while referring to the pixel information of block 1 must be performed again by replacing the pixel information. As a result, it is necessary to go back to the inter-screen / intra prediction processing of block 1. However, the amount of processing going back from the pipeline described with reference to FIG. 4B in the first embodiment is reduced, and further, it is possible to suppress a delay in processing speed.

Furthermore, the code sequence generated by the moving picture encoding apparatus 100 according to the present embodiment is guaranteed that the generated code amount of the intermediate code sequence is equal to or less than the specific code amount. For this reason, in the decoding pipeline described with reference to FIG. 15, the code amount of the intermediate code string read in the code string analysis process is equal to or less than the specific amount. Therefore, it is possible to always keep the processing time of the code string analysis processing within the specified value as shown in FIG. 15A, and it is possible to avoid the occurrence of delay in the decoding processing.

As described above, the moving picture coding apparatus 100-1 according to the present embodiment switches to the PCM mode without going back in the pipeline even when the coding method by arithmetic coding is realized using the pipeline structure. Encoding can be performed. Therefore, the generated code amount can be kept below the limit value without increasing the processing speed or increasing the processing amount. Furthermore, in a decoding apparatus that decodes a generated code string, the time required to read the intermediate code string after arithmetic decoding in the code string analysis process can always be set to a certain time or less. Therefore, it is possible to avoid the occurrence of pipeline processing delay.

5. Summary The moving image encoding apparatus 100-1 in the present embodiment encodes an input encoding target image in units of blocks. The moving image encoding apparatus 100-1 includes a prediction encoding unit 105 that generates a prediction image corresponding to an encoding target image, and a difference calculation unit 111 that generates a difference image between the encoding target image and the generated prediction image. A prediction residual encoding unit 102 that performs orthogonal transformation processing and quantization processing on the output of the difference calculation unit 111 to generate a residual coefficient, and inverse quantization processing and inverse orthogonal transformation for the residual coefficient A prediction residual decoding unit 103 that performs processing and generates a residual decoded image; a prediction image generated by the prediction encoding unit 105; a residual decoded image generated by the prediction residual decoding unit 103; And an addition operation unit 112 that generates a reconstructed image by adding, and in the first mode, a coefficient code string is generated by variable-length encoding the residual coefficient generated by the prediction residual encoding unit 102, and prediction Prediction used when generating the image Generating a header code string including at least information, and outputting a coefficient code string and an intermediate code string composed of the header code string, while in the second mode, a signal processing process of an encoding target image instead of a residual coefficient The intermediate image signal obtained in the above is directly converted into a coefficient code string without variable length coding, and a header code string including at least prediction information used when generating a predicted image is generated, and is composed of the coefficient code string and the header code string A binary encoding unit 107-1 for outputting the intermediate code string to be output, and arithmetic coding the intermediate code string output in the first mode to generate an output code string, while outputting in the second mode Arithmetic coding that generates only the header code string of the intermediate code string and generates an output code string composed of the header code string and coefficient code string after the arithmetic coding 110-1 and the binary encoding unit 107-1, when outputting the intermediate code string to the arithmetic encoding unit 110-1, the code amount of the intermediate code string is defined as a predetermined specific code amount. Restrict to:

Further preferably, the binary coding unit 107-1 uses the second mode when the code amount of the intermediate code string 159-1 to be output may exceed a predetermined specific code amount. To output an intermediate code string.

Further preferably, as a result of outputting the intermediate code string 159-1 in the first mode, the binarization encoding unit 107-1 obtains a predetermined code amount as a code amount of the intermediate code string 159-1. In the case of exceeding, the intermediate code string 159-1 generated using the second mode is output instead of the intermediate code string 159-1 generated in the first mode.

(Embodiment 3)
A moving picture decoding apparatus according to Embodiment 3 will be described with reference to the drawings.

1. Configuration of Video Decoding Device FIG. 10 is a block diagram illustrating a configuration of a video decoding device 200 according to the third embodiment. The moving picture decoding apparatus 200 decodes the code string generated by the moving picture encoding apparatus described in Embodiment 1 in units of blocks called coding units (CUs), and generates an output image.

The moving picture decoding apparatus 200 includes an arithmetic decoding unit 201, a binary decoding unit 202, a prediction residual decoding unit 205, a picture memory 206, a prediction decoding unit 207, and a quantization value determination. Part 208.

The arithmetic decoding unit 201 performs arithmetic decoding on the input code string signal 251 in units of blocks, and outputs the generated intermediate code string signal 252 to the binarization decoding unit 202.

The binarization decoding unit 202 includes a header code string analysis unit 203 and a coefficient code string analysis unit 204, and analyzes each code string by applying each process to the input intermediate code string 252. .

The header code string analysis unit 203 analyzes header information by performing variable length decoding on the header intermediate code string of the input intermediate code string signal 252. Then, the header code string analysis unit 203 outputs the prediction information signal 257 obtained by the analysis to the prediction decoding unit 207. Further, the header code string analysis unit 203 outputs the quantization value information 262 obtained by the analysis to the quantization value determination unit 208.

The coefficient code string analysis unit 204 performs a variable length decoding on the coefficient intermediate code string encoded subsequent to the header intermediate code string analyzed by the header code string analysis unit 203 to thereby generate a residual encoded signal. 253 is acquired. Then, the coefficient code string analysis unit 204 has a first mode for outputting the residual encoded signal 253 to the prediction residual decoding unit 205. Further, the coefficient code string analysis unit 204 reconstructs the reconstructed image signal without performing variable length decoding on the coefficient intermediate code string encoded subsequent to the header intermediate code string analyzed by the header code string analysis unit 203. 256 ′ is acquired, and the acquired reconstructed image signal 256 ′ is replaced with the reconstructed image signal 256 that is the output of the addition operation unit 209. When processing is performed using the second mode at this time, the generation process of the residual decoded signal 254 by the prediction residual decoding unit 205 and the generation process of the predicted image signal 255 by the prediction decoding unit 207 are described below. You don't have to.

The prediction residual decoding unit 205 generates a residual decoded signal 254 by performing inverse quantization and inverse orthogonal transform on the residual encoded signal 253 input from the coefficient code string analyzing unit 204. Then, the generated residual decoded signal 254 is output to the addition operation unit 209. At this time, the prediction residual decoding unit 205 dequantizes the residual encoded signal 253 using the quantized value signal 258 determined by the quantized value determining unit 208.

The predictive decoding unit 207 generates a predicted image signal 255 using intra prediction or inter prediction based on the prediction information signal 257 output from the header code string analysis unit 203, and outputs the prediction image signal 255 to the addition operation unit 209. . Note that the predictive decoding unit 207 uses the reconstructed image signal 256 of a past picture that has already been decoded and stored in the picture memory 206 when using inter-screen prediction. Further, when using intra prediction, the reconstructed image signal 256 of the current picture of a CU that has already been decoded and is adjacent to the decoding target CU is used. Whether to use intra prediction or inter prediction is determined according to the input prediction information signal 257.

The picture memory 206 stores the reconstructed image signal 256 sequentially input from the addition operation unit 209 or the coefficient code string analysis unit 204 in units of pictures, rearranges them in the order of output in units of pictures, and displays them as an output image signal 259. Output to.

The addition operation unit 209 adds the residual decoded signal 254 output from the prediction residual decoding unit 205 and the predicted image signal 255 output from the prediction decoding unit 207 to add the reconstructed image signal 256. Generate and output to the picture memory 206.

Here, a method of analyzing the code string signal 251 in the arithmetic decoding unit 201 and the binary decoding unit 202 will be specifically described with reference to the flowchart of FIG.

2. Code Sequence Analysis Method FIG. 11 is a flowchart when the code sequence signal generated by the video encoding apparatus 100 according to the first embodiment is analyzed.

First, the arithmetic decoding unit 201 analyzes PCM mode information described at the beginning of the input code string (S1301). Subsequently, according to the obtained PCM mode information, it is determined whether the target CU is encoded in the Residual mode or the PCM mode (S1302).

If it is determined in step S1302 that the encoding is performed in the Residual mode, the intermediate code string is generated by performing arithmetic decoding on the header code string and the coefficient code string belonging to the input code string (S1303). .

On the other hand, if it is determined in step S1302 that the encoding is performed in the PCM mode, the bit string acquired without performing arithmetic decoding on the coefficient code string belonging to the input code string is used as the code string as it is. The intermediate code string is generated by describing the code string of the previously analyzed PCM mode information (S1306).

Next, when the input intermediate code string is generated in the Residual mode, the binary decoding unit 202 analyzes header information by performing variable length decoding on the header intermediate code string. The generated prediction information signal 257, quantization value information, and other decoding control information are output to each processing block of FIG. 10 (S1304).

Subsequently, the residual encoded signal 253 obtained by performing variable length decoding on the coefficient intermediate code string is output to the prediction residual decoding unit 205 (S1305).

On the other hand, when the input intermediate code string is generated in the PCM mode, the binary decoding unit 202 does not perform variable length decoding on the coefficient intermediate code string, and directly converts the acquired bit string. The reconstructed image signal 256 ′ is output and replaced with the reconstructed image signal 256 output from the addition operation unit 209, and the subsequent processing is performed (S1307). At this time, since the intermediate code string does not include header information other than the PCM mode information, it is not necessary to analyze the header intermediate code string.

Note that in the video encoding apparatus 100 described in Embodiment 1, even when a code string is generated using the reconstructed image signal 155 instead of the input image signal in step S606 in FIG. It is not necessary to distinguish at all as a decoding process for. That is, it is possible to output as a reconstructed image signal in step S1307 and perform subsequent processing.

3. Syntax In this embodiment, the syntax of the code string to be decoded and the maximum code amount of the generated intermediate code string are exactly the same as those in the first embodiment.

4). Pipeline Improvement Effect A moving picture coding apparatus that generates a code sequence corresponding to the moving picture decoding apparatus 200 according to the present embodiment can have the configuration described in the first embodiment. As shown in FIG. 4B, the PCM mode can be determined at the time of the binary encoding process. Therefore, it is not necessary to consider the timing difference between the pipeline processing and the arithmetic coding processing, and the amount of processing going back in the pipeline can always be made constant. Therefore, it is possible to simplify the process control when the PCM mode occurs. Furthermore, it is guaranteed that the generated code amount of the generated intermediate code string is equal to or less than a specific code amount. Therefore, in the decoding pipeline described with reference to FIG. 15, it is possible to always keep the processing time of the code string analysis processing within the specified value as shown in FIG. That is, it is possible to avoid the occurrence of a delay in the decoding process.

(Embodiment 4)
A moving picture decoding apparatus according to Embodiment 4 will be described with reference to the drawings.

1. Configuration of Video Decoding Device The configuration of the video decoding device 200 according to the fourth embodiment is the same as the configuration of the video decoding device 200 according to the third embodiment described with reference to FIG. Therefore, explanation is omitted.

2. Code Sequence Analysis Method FIG. 12 is a flowchart when analysis is performed on a code string signal generated by the video encoding apparatus 100-1 according to the second embodiment. This flowchart is almost the same as the flowchart of FIG. 11, but steps S1401 and S1403 are used instead of steps S1301 and S1303, and step S1404 is newly added.

First, the arithmetic decoding unit 201 performs arithmetic decoding on the header code string belonging to the input code string to generate a header intermediate code string (S1401). At this time, a signal indicating whether the target CU is encoded in the Residual mode or the PCM mode, which is information described in the header code string, is sent to Step S1302 and used for the determination process.

If it is determined in step S1302 that the encoding is performed in the Residual mode, arithmetic decoding is performed on the coefficient code string belonging to the input code string. Then, an intermediate code string is generated by describing the header intermediate code string generated earlier (S1403).

On the other hand, if it is determined in step S1302 that the coding is performed in the PCM mode, the acquired bit string is directly used without performing arithmetic decoding on the coefficient code string belonging to the input code string. An intermediate code string is generated by describing the generated header intermediate code string after the header intermediate code string (S1306).

Next, the binary decoding unit 202 performs variable-length decoding on the header code string in the same manner whether the input intermediate code string is generated in the Residual mode or the PCM mode. Header information is analyzed. Then, the generated prediction information signal 257, quantization value information, and other decoding control information are output to each processing block of FIG. 10 (S1304 and S1404).

Subsequently, when the input intermediate code sequence is generated by the Residual mode, the prediction residual decoding unit converts the residual encoded signal 253 obtained by performing variable length decoding on the coefficient intermediate code sequence. It outputs to 205 (S1305).

On the other hand, when the input intermediate code string is generated by the PCM mode, the acquired bit string is output as it is as the reconstructed image signal 256 ′ without performing variable length decoding on the coefficient intermediate code string. Then, it is replaced with the reconstructed image signal 256 output from the addition operation unit 209, and the subsequent processing is performed (S1307).

Note that in the moving picture encoding apparatus 100-1 described in the second embodiment, even when a code string is generated using an input image signal instead of the reconstructed image signal 155 in step S1005 in FIG. It is not necessary to distinguish the decoding process for the code string. That is, it is possible to output as a reconstructed image signal in step S1307 and perform subsequent processing.

4). Pipeline improvement effect By using the moving picture decoding apparatus 200 according to the present embodiment, a moving picture encoding apparatus that generates a code string corresponding to the moving picture decoding apparatus 200 can have the configuration described in the second embodiment. As shown in FIG. 9B, the PCM mode can be determined at the time of the binary encoding process. Therefore, it is not necessary to consider the timing difference between pipeline processing and arithmetic coding processing. Further, it is not necessary to change the prediction information described in the header code string or the pixel information of the finally generated reconstructed image. Therefore, it is possible to perform processing control when the PCM mode occurs without going back in the pipeline. Furthermore, it is guaranteed that the generated code amount of the generated intermediate code string is equal to or less than a specific code amount. Therefore, in the decoding pipeline described with reference to FIG. 15, it is possible to always keep the processing time of the code string analysis processing within the specified value as shown in FIG. That is, it is possible to avoid the occurrence of a delay in the decoding process.

(Other embodiments)
By recording a program having the same function as each unit included in the moving image encoding device and the moving image decoding device described in each of the above embodiments on a recording medium such as a flexible disk, The processing shown in the form can be easily performed in an independent computer system. The recording medium is not limited to a flexible disk, but may be any medium that can record a program, such as an optical disk, an IC card, and a ROM cassette.

Also, the functions equivalent to the means included in the moving picture coding apparatus and the moving picture decoding apparatus shown in the above embodiments may be realized as an LSI which is an integrated circuit. These may be integrated into one chip so as to include a part or all of them. An LSI may also be called an IC, a system LSI, a super LSI, or an ultra LSI depending on the degree of integration.

Further, the method of circuit integration is not limited to LSI, and implementation with a dedicated circuit or a general-purpose processor is also possible. An FPGA (Field Programmable Gate Array) that can be programmed after manufacturing the LSI, or a reconfigurable processor that can reconfigure the connection and setting of circuit cells inside the LSI may be used.

Furthermore, if integrated circuit technology that replaces LSI or the like appears due to progress in semiconductor technology or other derived technology, the functional blocks may naturally be integrated using this technology.

Further, the present invention is applied to a broadcast wave recording apparatus such as a DVD recorder or a BD recorder that compresses and records a broadcast wave broadcast from a broadcast station, including the above-described moving picture encoding apparatus and moving picture decoding apparatus. It doesn't matter.

Further, at least a part of the functions of the moving picture coding apparatus and the moving picture decoding apparatus according to the above-described embodiment, or a modification thereof may be combined.

The present invention, for example, in a video camera, a digital camera, a video recorder, a mobile phone, a personal computer, etc., a moving image encoding device that encodes each picture constituting an input image and outputs it as moving image encoded data, The present invention is useful as a moving picture decoding apparatus that generates a decoded picture by decoding the moving picture encoded data.

100, 100-1 Video coding apparatus 101 Picture memory 102 Prediction residual encoding unit 103 Prediction residual decoding unit 104 Local buffer 105 Prediction encoding unit 106 Quantized value determination unit 107, 107-1 Binary code Conversion unit 108, 108-1 header code sequence generation unit 109, 109-1 coefficient code sequence generation unit 110, 110-1 arithmetic encoding unit 111 difference calculation unit 112 addition calculation unit 151 input image signal 152 difference image signal 153 residual Coded signal 154 Residual decoded signal 155 Reconstructed image signal 156 Predicted image signal 157 Predicted information signal 158 Quantized value signal 159, 159-1 Intermediate code sequence signal 160, 160-1 Code sequence signal 200 Video decoding device 201 Arithmetic decoding unit 202 Binary decoding unit 203 Header code string analysis unit 204 Coefficient code Sequence analysis unit 205 Prediction residual decoding unit 206 Picture memory 207 Prediction decoding unit 208 Quantized value determination unit 209 Addition operation unit 251 Code sequence signal 252 Intermediate code sequence signal 253 Residual encoded signal 254 Residual decoded signal 255 Predicted image signal 256 Reconstructed image signal 257 Predicted information signal 258 Quantized value signal 259 Output image signal

Claims

A video encoding device that encodes an input encoding target image in units of blocks,
A predicted image generation unit that generates a predicted image corresponding to the encoding target image;
A subtractor that generates a difference image between the encoding target image and the generated predicted image;
A prediction residual encoding unit that performs orthogonal transform processing and quantization processing on the output of the subtractor to generate a residual coefficient;
A prediction residual decoding unit that performs an inverse quantization process and an inverse orthogonal transform process on the residual coefficient to generate a residual decoded image;
An adder that generates a reconstructed image by adding the prediction image generated by the prediction image generation unit and the residual decoded image generated by the prediction residual decoding unit;
The residual coefficient generated by the prediction residual encoding unit is variable-length encoded to generate a coefficient code string, and a header code string including at least prediction information used when generating the predicted image is generated, A binary coding unit that outputs a coefficient code string and an intermediate code string composed of the header code string;
An arithmetic encoding unit that arithmetically encodes the intermediate code sequence output by the binary encoding unit and generates an output code sequence, and
When the binary coding unit outputs an intermediate code string to the arithmetic coding unit, the code amount of the intermediate code string is limited to a predetermined specific code amount or less.
Video encoding device.
When the code amount of the intermediate code string to be output may exceed the predetermined specific code amount, the binarization encoding unit has a larger quantization width in the prediction residual encoding unit. To regenerate the residual coefficient by performing the quantization process again, and output an intermediate code string using the regenerated residual coefficient,
The moving image encoding apparatus according to claim 1.
When the code amount of the intermediate code sequence exceeds the predetermined specific code amount as a result of outputting the intermediate code sequence, the binary coding unit outputs a larger quantum in the prediction residual coding unit. A residual coefficient is regenerated by performing quantization processing again by applying a quantization width, and an intermediate code string generated using the regenerated residual coefficient is output instead of the output intermediate code string The moving picture encoding apparatus according to claim 1.
A video encoding device that encodes an input encoding target image in units of blocks,
A predicted image generation unit that generates a predicted image corresponding to the encoding target image;
A subtractor that generates a difference image between the encoding target image and the generated predicted image;
A prediction residual encoding unit that performs orthogonal transform processing and quantization processing on the output of the subtractor to generate a residual coefficient;
A prediction residual decoding unit that performs an inverse quantization process and an inverse orthogonal transform process on the residual coefficient to generate a residual decoded image;
An adder that generates a reconstructed image by adding the prediction image generated by the prediction image generation unit and the residual decoded image generated by the prediction residual decoding unit;
In the first mode, a header code string including at least prediction information used when generating a coefficient code string by variable-length coding the residual coefficient generated by the prediction residual coding unit and generating the predicted image And the intermediate code string composed of the coefficient code string and the header code string is output, and obtained in the signal processing process of the encoding target image instead of the residual coefficient in the second mode. An intermediate code sequence composed of a header code sequence including at least an identifier indicating that the coefficient code sequence and the coefficient code sequence are intermediate image signals without changing the length of the intermediate image signal. A binary encoding unit to output;
The intermediate code string output in the first mode is arithmetically encoded to generate an output code string, while the intermediate code string output in the second mode is arithmetically encoded only for the header code string An arithmetic coding unit that generates an output code string composed of a header code string after arithmetic coding and the coefficient code string,
The binary coding unit is a moving picture coding device that restricts a code amount of an intermediate code string to a predetermined code amount or less when outputting the intermediate code string to the arithmetic coding unit.
The specific code amount defined in advance includes a code amount required when the pixel value of the intermediate image signal is directly described as a code string, and all information that may be described in the header code string. 5. The moving picture encoding apparatus according to claim 1, wherein the encoding amount is a code amount that is a sum of a maximum code amount necessary for encoding and a margin amount. 6.
The binarization encoding unit outputs the intermediate code sequence using the second mode when the code amount of the output intermediate code sequence may exceed the predetermined specific code amount. The moving image encoding apparatus according to claim 4.
When the code amount of the intermediate code sequence exceeds the predetermined specific code amount as a result of outputting the intermediate code sequence in the first mode, the binarization encoding unit generates the first code mode. The moving picture coding apparatus according to claim 4, wherein an intermediate code string generated using the second mode is output instead of the intermediate code string.
A moving image encoding method for encoding an input encoding target image in units of blocks,
Generating a predicted image corresponding to the encoding target image;
Generating a difference image between the encoding target image and the generated predicted image;
An orthogonal transformation process and a quantization process are performed on the output of the subtractor to generate a residual coefficient,
Performing an inverse quantization process and an inverse orthogonal transform process on the residual coefficient to generate a residual decoded image;
Generating a reconstructed image by adding the generated predicted image and the generated residual decoded image;
The generated residual coefficient is variable-length encoded to generate a coefficient code string, a header code string including at least prediction information used when generating the predicted image is generated, the coefficient code string, and the header Output an intermediate code sequence consisting of code sequences,
Arithmetically encoding the output intermediate code string to generate an output code string;
A moving picture encoding method for limiting the code amount of an intermediate code sequence to a predetermined code amount or less when outputting the intermediate code sequence.
A moving image encoding method for encoding an input encoding target image in units of blocks,
Generating a predicted image corresponding to the encoding target image;
Generating a difference image between the encoding target image and the generated predicted image;
An orthogonal transformation process and a quantization process are performed on the difference image to generate a residual coefficient,
Performing an inverse quantization process and an inverse orthogonal transform process on the residual coefficient to generate a residual decoded image;
Generating a reconstructed image by adding the generated predicted image and the generated residual decoded image;
In the first mode, the generated residual coefficient is variable-length encoded to generate a coefficient code string, a header code string including at least prediction information used when generating the predicted image is generated, and the coefficient code Output an intermediate code sequence composed of a sequence and the header code sequence, and in the second mode, instead of the residual coefficient, an intermediate image signal obtained in a signal processing process of the encoding target image is a variable-length code Output an intermediate code string composed of a header code string including at least an identifier indicating that the coefficient code string and the coefficient code string are intermediate image signals,
The intermediate code string output in the first mode is arithmetically encoded to generate an output code string, while the intermediate code string output in the second mode is arithmetically encoded only for the header code string Generating an output code string composed of the header code string after arithmetic coding and the coefficient code string;
A moving picture coding method for restricting a code amount of an intermediate code sequence to a predetermined code amount or less when outputting the intermediate code sequence.