JP3948442B2 - Motion compensation decoding apparatus - Google Patents

Description

  The present invention decodes a code sequence that has been subjected to high-efficiency coding that converts image information into a digital signal with a smaller code amount in order to efficiently transmit, store, and display the image. The present invention relates to a method for performing motion compensation decoding on a code string on which motion vector information is variably encoded.

<Motion compensated video coding>
In motion picture encoding, motion compensated inter-picture prediction is generally performed as represented by MPEG and the like. At this time, the motion vector (MV) information used in the motion compensation is variable-length coded, multiplexed with the prediction residual code, and transmitted.

  In the MV encoding, the difference from the MV of the adjacent block is taken, and the horizontal difference value and the vertical difference value are each encoded with a variable length code. This variable length code is composed of a Huffman code or the like in accordance with the frequency of occurrence of the difference value.

<Conventional Motion Compensation Encoding Device>
A configuration example of a conventional motion compensation encoding apparatus will be described below with reference to FIG.

  The motion picture signal coming from the picture input terminal 1 is subtracted from the prediction signal given from the motion compensation predictor 10 in the subtracter 2 and given to the DCT 3 as a prediction residual. The DCT 3 performs DCT (Discrete Cosine Transform) conversion processing on the prediction residual and gives the obtained coefficient to the quantizer 4. The quantizer 4 quantizes the coefficient with a predetermined step width, and supplies the coefficient that has become a fixed-length code to the variable-length encoder 5 and the inverse quantizer 7. The variable length encoder 5 compresses the fixed-length prediction residual with the variable length code, and the resulting code is supplied to the multiplexer 61.

  On the other hand, the inverse quantizer 7 and the inverse DCT 13 perform the inverse processing of the quantizer 4 and the DCT 3 to reproduce the prediction residual. The obtained reproduction prediction residual is added with a prediction signal by an adder 12 to form a reproduction image, which is given to the image memory 11. The reproduced image stored in the image memory 11 is given to the motion compensation predictor 10.

  The motion compensation predictor 10 performs motion compensation in accordance with the MV supplied from the MV detector 6 and supplies the obtained prediction signal to the subtracter 2 and the adder 12. The MV detector 6 obtains the spatial movement amount of the reference frame with respect to the encoded frame for each 16 × 16 pixel or 8 × 8 pixel block, and provides it to the motion compensation predictor 10 and the MV encoder 62 as MV. The MV encoder 62 takes the difference from the previous MV and encodes the difference value with a predetermined variable length code. The MV code is multiplexed with the prediction residual code by the multiplexer 61 and output from the code output terminal 9.

<Conventional Motion Compensation Decoding Device>
FIG. 7 shows a configuration of a conventional example of a motion compensation decoding apparatus corresponding to the motion compensation encoding apparatus of FIG.

  A code sequence of motion-compensated predictive coding that comes from the code input terminal 21 is separated into a prediction residual code and an MV code by a demultiplexer 71, and the prediction residual code is converted into a variable-length decoder 23. The MV code is provided to the MV decoder 72. The variable length decoder 23 converts the variable length code back to a fixed length code and provides it to the inverse quantizer 7.

  The fixed-length code becomes a reproduction DCT coefficient value of the prediction residual by the inverse quantizer 7 and is given to the inverse DCT 13. The inverse DCT 13 converts 8 × 8 coefficients into a reproduction prediction residual signal and supplies it to the adder 12. The adder 12 adds the prediction signal to the reproduction prediction residual signal to form a reproduction image. The reproduced image signal obtained in this way is output from the image output terminal 24 and given to the image memory 11. The motion compensation predictor 10 performs motion compensation on the image stored in the image memory 11 based on the MV, and forms a prediction signal. The obtained prediction signal is given to the adder 12. On the other hand, the MV decoder 72 performs reverse processing of the MV encoder 62 of FIG. 6 to obtain an MV value from the MV code. The obtained MV is given to the motion compensation predictor 10.

  A conventional motion compensation coding apparatus uses one type of code table for variable length coding of a motion vector (MV). The code table used there is created according to the MV difference distribution (probability of occurrence of each difference value), but the MV difference varies greatly depending on the situation of the image movement, and the distribution matches the code table depending on the image part. In many cases, the generated code amount is large. In particular, if the motion compensation block is made finer and the accuracy is increased, the ratio of the MV information amount increases, so the influence on the total code amount becomes significant.

The present invention has been made paying attention to the above points, and has a plurality of types of variable-length code tables for MV encoding, integrates a predetermined number of MVs, and makes one image (one frame or field) a plurality. A motion compensation code capable of reducing the MV code amount by performing MV coding using a code table selected in units of divided MV groups, transmitting the code table selection information, and using a common code table for decoding It is an object of the present invention to provide a motion compensation decoding apparatus corresponding to the conversion.

In order to solve the above problems, the present invention provides the following apparatus .

(1) In a motion compensation decoding apparatus that decodes a motion picture code sequence in which a motion picture is motion compensation prediction-encoded with a motion vector for each motion compensation block,
An incoming moving image code string includes the motion in each integrated unit obtained by dividing one image into a plurality of regions by integrating units in which the motion compensation blocks are integrated within a range of about 16 to 64. Code table selection information indicating a variable length code table selected for encoding the motion vector from a plurality of variable length coding tables for each motion vector group obtained by integrating vectors, and the selected variable The motion vector information obtained by variable-length coding the motion vector using a long code table is multiplexed,
The variable length code table for encoding the motion vector is in accordance with a comparison result in which a value related to a code amount when the motion vector is encoded by the plurality of variable length code tables is compared for each motion vector group. , Selected for each of the motion vector groups from the plurality of variable length code tables,
Demultiplexing means for separating the motion vector information and the code table selection information from the incoming video code sequence,
Motion vector decoding means for variable-length decoding the motion vector information using a variable length code table selected according to the code table selection information;
Decoding means for obtaining the moving picture by performing motion compensation predictive decoding on the incoming moving picture code string using the decoded motion vector;
A motion compensation decoding apparatus comprising:
(2) In a motion compensation decoding apparatus that decodes a motion picture code sequence in which a motion picture is motion compensation prediction-encoded with a motion vector for each motion compensation block,
An incoming moving image code string includes the motion in each integrated unit obtained by dividing one image into a plurality of regions by integrating units in which the motion compensation blocks are integrated within a range of about 16 to 64. Code table selection information indicating a variable length code table selected for encoding the motion vector from a plurality of variable length coding tables for each motion vector group obtained by integrating vectors, and the selected variable The motion vector information obtained by variable-length coding the motion vector using a long code table is multiplexed,
The variable length code table for encoding the motion vector has a value related to a code amount when the motion vector is encoded by one variable length code table preselected from the plurality of variable length code tables. Obtained for each motion vector group, and selected for each motion vector group from the plurality of variable length code tables according to the relationship between the value related to the code amount for each motion vector group and a preset threshold value. Yes,
Demultiplexing means for separating the motion vector information and the code table selection information from the incoming video code sequence,
Motion vector decoding means for variable-length decoding the motion vector information using a variable length code table selected according to the code table selection information;
Buffer means for holding the decoded motion vector;
Decoding means for obtaining the moving picture by receiving the decoded motion vector corresponding to each motion compensation block from the buffer means and performing motion compensation predictive decoding on the incoming moving picture code string;
A motion compensation decoding apparatus comprising:

(Work)
The present invention accurately performs motion compensation decoding on a code string encoded by the following motion compensation encoding. In motion compensated video coding, there are multiple types of variable length code tables for motion vector (MV) coding, MVs are integrated in a predetermined number, and one image (one frame or field) is divided into multiple By performing MV coding using a code table selected in units of MV groups, transmitting the code table selection information, and using a common code table for decoding, an appropriate code table can be selected from a plurality of code tables. Is used. Since a plurality of types of code tables are prepared according to different MV difference occurrence distributions, a code table suitable for the MV group is selected even if the distribution in the MV group of the MV difference is significantly different from the average distribution. The selected code table is composed of variable length codes adapted to the MV difference distribution of the MV group, and the code amount of the MV group is smaller than that of the average code table.

  In each MV group, a code table suitable for the MV group is selected, and the MV code amount is reduced as a whole. On the other hand, since the code table selection information accompanies, the amount of code increases accordingly, but since it is 1 to 2 bits in the MV group, only a slight increase is required compared with the reduction effect.

  In the present invention, it is possible to accurately perform motion compensation decoding on a code string encoded by the following motion compensation encoding.

  In motion compensated video coding, there is a plurality of variable length code tables for motion vector (MV) coding, a predetermined number of MVs are integrated, and one image (one frame or field) is divided into a plurality of MVs. MV coding is performed using a code table selected in units of groups, and the code table selection information is transmitted and a common code table is used for decoding, so that an appropriate code table can be used from among a plurality of code tables. A plurality of types of code tables are prepared in accordance with different MV difference occurrence distributions, and those that match the MV difference distribution of the MV group and that match the partial state of the image are selected. Less.

  Since the code amount of the prediction residual does not change, the total generated code amount is reduced. In particular, if the motion compensation block is made finer and the accuracy is increased, the ratio of the MV code amount is relatively increased, so the effect of reducing the MV code amount greatly contributes to the reduction of the total code amount as it is.

<First Example of Motion Compensation Encoding Device Corresponding to the Present Invention>
A first example of a motion compensation coding apparatus corresponding to the motion compensation decoding apparatus of the present invention (generating a code string accurately decoded in the motion compensation decoding apparatus of the present invention) will be described below.

  FIG. 1 shows the configuration, and the same components as those in the conventional example of FIG. 6 are denoted by the same reference numerals. Compared with FIG. 6, an MV integrator 15, a provisional MV encoder 16, a code amount calculator 17, a code table determiner 18, and a delay unit 19 are added to FIG. 1. The operations of the MV encoder 14 and the multiplexer 8 are different from those of the MV encoder 62 and the multiplexer 61 in FIG.

  In this example, the motion vector (MV) encoding method is greatly different from the conventional example. The inter-picture prediction process and the encoding of the prediction residual are basically the same.

  The motion picture signal coming from the picture input terminal 1 is subtracted from the prediction signal given from the motion compensation predictor 10 in the subtracter 2 and given to the DCT 3 as a prediction residual. The DCT 3 performs DCT (Discrete Cosine Transform) conversion processing on the prediction residual, and gives the obtained coefficient to the quantizer 4. The quantizer 4 quantizes the coefficient with a predetermined step width, and supplies the coefficient that has become a fixed-length code to the variable-length encoder 5 and the inverse quantizer 7. The variable-length encoder 5 compresses the fixed-length prediction residual with the variable-length code, and the code string generated by the compression is supplied to the multiplexer 8.

  On the other hand, the inverse quantizer 7 and the inverse DCT 13 perform the inverse processing of the DCT 3 and the quantizer 4 to reproduce the prediction residual. The obtained reproduction prediction residual is added with a prediction signal by an adder 12 to form a reproduction image, which is given to the image memory 11.

  The reproduced image stored in the image memory 11 is given to the motion compensation predictor 10. The motion compensation predictor 10 creates a prediction signal according to the MV supplied from the MV detector 6 and supplies the prediction signal to the subtracter 2 and the adder 12.

  The MV detector 6 obtains the spatial movement amount of the reference frame with respect to the encoded frame for each 16 × 16 pixel or 8 × 8 pixel block, and provides it to the motion compensation predictor 10 and the MV integrator 15 as MV.

<MV encoding>
One embodiment of the MV coding part that contributes to the features of the present invention is described below.

  In the MV integrator 15 shown in FIG. 1, about 16 to 64 MVs are integrated to form an MV group in which one frame is divided into a plurality. The MV group is preferably two-dimensional, such as 4 vertical × 4 horizontal, as shown in FIG. 4, from the viewpoint of encoding characteristics, but may be one-dimensional such as 16 in a horizontal row. In the case of the two-dimensional case, it has a memory for MV values and performs array conversion.

  Each MV of the MV group from the MV integrator 15 is given to the temporary MV encoder 16 and the delay unit 19. The provisional MV encoder 16 performs provisional encoding on all types of code tables included in the MV encoder 14, and provides the obtained code string to the next code amount calculator 17. .

  Here, specific examples of the code table are shown in Table 1 and Table 2, respectively.

  The first variable length code table in Table 1 is suitable when the change in MV is relatively small, and the second variable length code table in Table 2 is suitable when the change in MV is relatively large.

  An example of MV when the change is large and small is shown in FIG. In the table, s is a symbol indicating the polarity (+/−) of MV. The variable length code in the table is applied to each horizontal and vertical component of the MV difference. In addition, when MV has ½ pixel accuracy, it is applied to a value that is doubled to be an integer.

符号量算出器１７は、それぞれの符号表の符号列毎に符号量が計数され、ＭＶ群単位で符号量を符号表判定器１８に与える。なお、ここで最終的に必要なのは符号量だけであるので、必ずしも正式な符号列を得る必要はなく、仮ＭＶ符号化器１６は各ＭＶ差分の符号長を出力し、符号量算出器１７でそれをＭＶ群単位に積算しても良い。

The code amount calculator 17 counts the code amount for each code string of each code table, and provides the code amount to the code table determiner 18 in units of MV groups. Since only the code amount is finally required here, it is not always necessary to obtain a formal code string. The provisional MV encoder 16 outputs the code length of each MV difference, and the code amount calculator 17 It may be integrated in MV group units.

  The code table determination unit 18 compares the code amounts of the code tables, and determines and selects the code table with the smallest code amount from either Table 1 or Table 2, for example. The determination result is given to the MV encoder 14 and the multiplexer 8 as code table selection information.

  On the other hand, each MV in the MV group is delayed by the delay unit 19 until the code table selection information is output, and then supplied to the MV encoder 14.

  In the MV encoder 14, the basic encoding method is the same as that of the conventional example of FIG. 7, but there are a plurality of code tables, and one of them is selected according to the code table selection information and used for encoding. Is done. The obtained code is given to the multiplexer 8.

  The multiplexer 8 multiplexes the prediction residual code provided from the variable length encoder 5 and the MV code provided from the MV encoder 14, in addition to the code table provided from the code table determination unit 18. The selection information is inserted into the header of the MV code.

  The code string multiplexed here is output from the code output terminal 9.

<Second Example of Motion Compensation Encoding Device Corresponding to the Present Invention>
A second example of a motion compensation coding apparatus corresponding to the motion compensation decoding apparatus of the present invention (generating a code string accurately decoded in the motion compensation decoding apparatus of the present invention) will be described below.

  FIG. 3 shows the configuration, and the same components as those in the first example of FIG. In FIG. 3, an activity detector 31 is present instead of the temporary MV encoder 16 and the code amount calculator 17 as compared with FIG. The operation of the code table determiner 32 is different from that of the code table determiner 18 of FIG.

  In the second example, the code table determination method is different from the first example, and other processing methods are basically the same, and only different parts will be described below.

  For each MV of the MV group given from the MV integrator 15, the activity detector 31 obtains the MV activity and gives it to the code table determination unit 32. The activity is obtained for each MV group, such as the sum of absolute values of MV difference values. Here, if the absolute value of the MV difference value is logarithmically converted and then added, the correlation with the MV code amount increases, which is convenient.

  The code table determination unit 32 determines an appropriate code table from the relationship between the activity and a preset threshold value.

  If there are two types of code tables, Table 1 and Table 2, medium activity is set as a threshold, Table 1 is set when the activity is smaller, and Table 2 is set when the activity is larger. In addition to the above-mentioned activities, the activities may be multidimensional that shows the distribution status more accurately. Specifically, the frequency of a relatively small MV difference (absolute value is 1 to 3) other than 0 and the frequency of a relatively large MV difference (absolute value is 4 or more) are obtained, and two-dimensional for two parameters. Is provided with a threshold value.

<Third Example of Motion Compensation Encoding Device Corresponding to the Present Invention>
A third example of a motion compensation coding apparatus corresponding to the motion compensation decoding apparatus of the present invention (generating a code string accurately decoded in the motion compensation decoding apparatus of the present invention) will be described below.

  The third example is a combination of the first example and the second example, and its configuration is similar to that of FIG.

  In the third example, the code table determination method is different from the other examples, and the other processes are basically the same. Therefore, only different parts will be described below with reference to FIG.

  The temporary MV encoder 16 and the code amount calculator 17 in FIG. 1 obtain a code amount for each MV group with respect to one type of code table, and give it to the code table determiner 18. The code table used at that time is the average of a plurality of code tables.

  The operation of the code table determiner 18 is similar to that of the code table determiner 32 of FIG. 3, and an appropriate code table is determined from the relationship between the code amount and a threshold value set in advance, and the code table selection information is obtained. Output. If the code amount is small, select a code table suitable for the MV difference with a small change as shown in Table 1. If the code amount is large, select a code table suitable for the MV difference with a large change as shown in Table 2. To do.

<Motion Compensated Decoding Device of Embodiment>
An embodiment of the motion compensation decoding apparatus of the present invention corresponding to the motion compensation encoding apparatus of each example described above will be described below with reference to the drawings.

  The configuration of an embodiment of the motion compensation decoding apparatus of the present invention is shown in FIG.

  A code sequence of motion compensated predictive coding coming from the code input terminal 21 is separated into a prediction residual code, MV code, and code table selection information by a multiplexing / separating unit 22, and the prediction residual code is variable. The MV code and code table selection information are supplied to the MV decoder 25 to the long decoder 23. The variable length decoder 23 converts the variable length code back to a fixed length code and provides it to the inverse quantizer 7.

  The prediction residual of the fixed length code becomes a coefficient value in the inverse quantizer 7 and is given to the inverse DCT 13. The inverse DCT 13 converts 8 × 8 coefficients into a reproduction prediction residual signal and supplies it to the adder 12. The adder 12 adds the prediction signal to the reproduction prediction residual signal to form a reproduction image. The reproduced image signal obtained in this way is output from the image output terminal 24 and given to the image memory 11.

  The motion compensation predictor 10 performs motion compensation on the image stored in the image memory 11 based on the MV, and forms a prediction signal. The obtained prediction signal is given to the adder 12.

  On the other hand, the MV decoder 25 performs reverse processing of the MV encoder 14 of FIG. 1 from the MV code and code table selection information, and gives the obtained MV value to the MV buffer 26.

  A code table used for MV decoding is selected and used from a plurality of types according to code table selection information in units of MV groups.

  The MV buffer 26 holds the MV and gives it when needed by the motion compensated predictor 10. This is because when the MV group is integrated in two dimensions, the block order of the main decoding process is different from the unit of the MV group.

It is a figure which shows the structure of the 1st example of the motion compensation encoding apparatus corresponding to the motion compensation decoding apparatus of this invention. It is a figure which shows the structure of one Example of the motion compensation decoding apparatus of this invention. It is a figure which shows the structure of the 2nd example of the motion compensation encoding apparatus corresponding to the motion compensation decoding apparatus of this invention. It is a figure which shows the mode of the MV group by MV integration. It is a figure which shows the mode of MV corresponding to the degree of a motion change. It is a figure which shows the structural example of the conventional motion compensation encoding apparatus. It is a figure which shows the structural example of the conventional motion compensation decoding apparatus.

Explanation of symbols

1 Image input terminal 2 Subtractor 3 DCT
4 Quantizer 5 Variable Length Encoder 6 MV Detector 7 Inverse Quantizer 8, 61 Multiplexer 9 Code Sequence Output Terminal 10 Dynamic Compensation Predictor 11 Image Memory 12 Adder 13 Inverse DCT
14, 62 MV encoder 15 MV integrator 16 Temporary MV encoder 17 Code amount calculator 18, 32 Code table determiner 19 Delay unit 21 Code string input terminal 22, 71 Demultiplexer 23 Variable length decoder 24 Image output terminals 25, 72 MV decoder 26 MV buffer 31 Activity detector

Claims (2)

In a motion compensation decoding apparatus for decoding a motion picture code sequence in which a motion picture is motion compensation prediction encoded by a motion vector for each motion compensation block,
An incoming moving image code string includes the motion in each integrated unit obtained by dividing one image into a plurality of regions by integrating units in which the motion compensation blocks are integrated within a range of about 16 to 64. Code table selection information indicating a variable length code table selected for encoding the motion vector from a plurality of variable length coding tables for each motion vector group obtained by integrating vectors, and the selected variable The motion vector information obtained by variable-length coding the motion vector using a long code table is multiplexed,
The variable length code table for encoding the motion vector is in accordance with a comparison result in which a value related to a code amount when the motion vector is encoded by the plurality of variable length code tables is compared for each motion vector group. , Selected for each of the motion vector groups from the plurality of variable length code tables,
Demultiplexing means for separating the motion vector information and the code table selection information from the incoming video code sequence,
Motion vector decoding means for variable-length decoding the motion vector information using a variable length code table selected according to the code table selection information;
Decoding means for obtaining the moving picture by performing motion compensation predictive decoding on the incoming moving picture code string using the decoded motion vector;
A motion compensation decoding apparatus comprising: In a motion compensation decoding apparatus for decoding a motion picture code sequence in which a motion picture is motion compensation prediction encoded by a motion vector for each motion compensation block,
An incoming moving image code string includes the motion in each integrated unit obtained by dividing one image into a plurality of regions by integrating units in which the motion compensation blocks are integrated within a range of about 16 to 64. Code table selection information indicating a variable length code table selected for encoding the motion vector from a plurality of variable length coding tables for each motion vector group obtained by integrating vectors, and the selected variable The motion vector information obtained by variable-length coding the motion vector using a long code table is multiplexed,
The variable length code table for encoding the motion vector has a value related to a code amount when the motion vector is encoded by one variable length code table preselected from the plurality of variable length code tables. Obtained for each motion vector group, and selected for each motion vector group from the plurality of variable length code tables according to the relationship between the value related to the code amount for each motion vector group and a preset threshold value. Yes,
Demultiplexing means for separating the motion vector information and the code table selection information from the incoming video code sequence,
Motion vector decoding means for variable-length decoding the motion vector information using a variable length code table selected according to the code table selection information;
Buffer means for holding the decoded motion vector;
Decoding means for obtaining the moving picture by receiving the decoded motion vector corresponding to each motion compensation block from the buffer means and performing motion compensation predictive decoding on the incoming moving picture code string;
A motion compensation decoding apparatus comprising:

Images