WO2005125212A1

WO2005125212A1 - Method for video encoding and decoding process

Info

Publication number: WO2005125212A1
Application number: PCT/CN2005/000795
Authority: WO
Inventors: Lianhuan Xiong
Original assignee: Huawei Technologies Co., Ltd.
Priority date: 2004-06-16
Filing date: 2005-06-07
Publication date: 2005-12-29
Also published as: CN100344163C; HK1086696A1; CN1710959A

Abstract

A method for video encoding and decoding process. The method comprises the steps as following: transform sub-block dividing each MB included in each set of macro blocks; sorting out all the sub-blocks which suffered the transform sub-block process in each set of macro blocks encoding and transmitting the header information of each set of macro blocks, the header information of macro blocks and the transform coefficients of each kind of sub-block set by the transmitting side; decoding the bit streams which were exactly received to restore the image information of corresponding sub-block by the receiving side for each sub-block which produces error codes , inserting approach image information by the receiving side on the basis of the image information of the adjacent sub-blocks. The error resilience process can be more accurately accomplished for the error bit stream due to transmission channel quality by means of the present invention. The invention can reduce the distortion of video image on vision and improve the effect of video communication.

Description

TECHNICAL FIELD The present invention relates to the field of multimedia communications, and in particular, to a video codec processing method.

Background technique

In multimedia communication systems, video image quality plays a very important role in multimedia communication performance. However, often due to the problem of the quality of the transmission channel, the encoded image stream of the video image will cause packet loss and error during the transmission process, especially the wireless transmission channel has a high bit error rate and packet loss rate. And the bandwidth changes greatly, so that the video image receiver cannot completely receive the encoded video stream of the video image, and if you want to obtain a good video communication effect, you need to correct the received error code Streams (including errored code streams and discarded code streams, etc.) are subjected to error-resistant processing, so that the video images recovered from decoding can be viewed normally.

At present, the motion picture experts group MPEG (Motion picture experts group) standard MPEG-4, the low-bit-rate video communication coding H.263 standard and the advanced video compression standard H.264 have all proposed corresponding anti-error for video transmission bitstreams. Code processing methods, such as:

Resynchronization processing method ^

Reversible Variable Length Code (RVLC) processing method;

Data partition (Data Partition) processing method;

Adaptive Striping (ASO, Adaptive SLICE Ordering) processing method;

Flexible Macroblock Ordering (FMO) processing; and

Redundant SLICEs processing method, etc .;

The above-mentioned anti-error processing methods of the video transmission code stream can compensate the erroneous code stream caused by the transmission channel to a certain extent, so that the video image receiver can better decode and recover the video image processing to obtain Better video communication.

However, during the video image encoding and decoding process, the error-resistant processing method of the video transmission code stream is used, which may appropriately reduce the encoding and decoding efficiency. However, when the quality of the transmission channel is poor, The anti-error processing of the transmission stream is very important to achieve a good video communication effect. Therefore, it is often willing to appropriately reduce the encoding and decoding efficiency in order to obtain better anti-error and packet loss performance in order to achieve better errors. Compensation purpose to obtain better video communication effect.

Among them, in the advanced video compression standard H.264, the video image compression standard is an integer transform method based on performance approximating a discrete digital cosine transform (DCT, Digit Cosine Transform), and the unit object targeted for compression is the video image. Each 16 x 16 pixel-sized macroblock (MB, Macroblock). Referring to FIG. 1, this figure is a schematic diagram of a compression process and a decompression process used by H. 264 in the prior art. A process of compressing and decompressing a video image is as follows:

Image compression process of video image sender

Perform motion estimation and motion compensation on a pair of macroblocks each 16 x 16 pixels divided in the video image;

DCT transform the 4X4 sub-blocks in each macro block repeatedly;

Quantize a pair of DCT transformed transform coefficients;

After entropy encoding the quantized result repeatedly, it is assembled into a data packet and sent to the transmission channel; the image decompression process of the video image receiver

Entropy decoding of a pair of received encoded code streams;

Perform inverse quantization on a pair of decoded results;

—The inverse discrete digital cosine transform (IDCT) is then performed on the inverse quantized result;

A pair of IDCT transformed results are motion compensated and the decoded video image is restored. The process of dividing a video image into multiple macroblocks MB includes: first performing transform macroblock division processing on the entire video image to obtain multiple macroblocks of 16 x 16 pixels, and then classifying the macroblocks, and A macroblock group composed of each type of macroblock is defined as a video clip (SLICE). The coding structure of each macroblock MB obtained by performing macroblock division processing on each video image is shown in Figure 2. For H.264, each macroblock MB includes 16 4 × 4 pixel-sized luminance blocks. Y (coded 0 to 15 block areas), 4 4X4 pixel-sized chroma blocks Cb (coded 16 to 19 block areas) and 4 4X4 pixel-sized chroma blocks Cr (coded 20 To block area coded as 23). In order to obtain better video communication effects, the H. 264 standard uses the Flexible Macroblock Ordering (FM0) video coding processing method proposed by POLYCOM. Referring to FIG. 3, which is a schematic diagram of the processing principle of the FM0 video encoding processing method in the prior art. In this figure, it is assumed that a macroblock MB identified by “0” belongs to the video segment SLICEO, and a macroblock MB identified by “1” It belongs to the video segment SLICE1 (of course, in addition to the case where the two kinds of macroblock ordering schemes are used to divide different SLICEs as shown in the figure, it can also include various cases where the arbitrary macroblock order is used to divide different SLICEs). As shown in the figure, the video image sender first performs compression encoding processing on the macroblock MB identified as "0", and then encapsulates it into a stream data packet for transmission; and then performs compression encoding on the macroblock MB identified as "1". Process, and then encapsulate and send it as a stream data packet. On the video image receiver side, first decode the stream data packet identified as "0", and then decode the stream data packet identified as "1"; if the transmission channel quality problem causes the flag to be " 0 "bitstream data packets (that is, the video segment SLICEO) are dropped on the transmission channel. According to the decoded and restored macroblock image information of the adjacent identifier" 1 "in the corresponding macro identified by" 0 " The approximate image information is inserted in the block, so as to achieve the purpose of masking errors.

Since the FM0 video encoding processing technology uses the macroblock MB as the basic processing unit for encoding and decoding, although it has a good concealment effect on the bit error code stream caused by the quality of the transmission channel, during the video image restoration process, According to the restored image information of other adjacent macroblocks MB, the approximated image inserted in the macroblock that caused the bit error still has a large distortion in the visual angle, so that the video communication cannot achieve the expected effect.

Summary of the invention

The technical problem to be solved by the present invention is to propose a video codec processing method to more accurately perform error-resistant processing on the bit error stream caused by the quality of the transmission channel, and reduce the visual distortion of the video image. Improve the effectiveness of video communications.

To solve the above problems, the present invention provides a video codec processing method, which includes steps:

B. Perform classification processing on all the sub-blocks after each macroblock group transformation process;

C. The sender encodes each macroblock group header information, the macroblock header information, and the transform coefficients of each type of subblock group and sends them separately; D. The receiver separately decodes the correctly received bitstream to recover the image information of the corresponding sub-block. For each sub-block that generates an error, the receiver inserts the image based on the image information of the corresponding neighboring sub-block. Approaching image information processing.

After the step B, the method further includes the following steps: Class sub-block groups belong to independent macro block groups, respectively, to split each corresponding macro block group into multiple independent macro block groups.

After the step B, the method further includes the following steps: The sub-block groups of the class belong to the sub-macro block groups of the corresponding macro-block group respectively, so that each corresponding macro-block group originally includes multiple sub-macro block groups.

In steps C and D, the sender and receiver also perform deblocking processing on adjacent sub-blocks belonging to the same type.

In steps C and D, the sender and receiver also perform intra prediction on the image information in the hand block belonging to the same type.

In the step B, classifying all the sub-blocks after each macroblock group transformation process is specifically performed as follows:

All sub-blocks after each macroblock group transformation process are divided into two types of sub-block groups according to the sub-block transformation coefficients.

Wherein each sub-block is 4 x 4 pixels in size. The arrangement modes of the two types of sub-block groups in each macro block include:

Two sub-blocks belonging to different types are distributed diagonally;

Sub-blocks belonging to different types are distributed in an interlaced state; or

Distributed in a state of columns.

Wherein each sub-block is 8 x 8 pixels in size. The arrangement modes of the two types of sub-block groups in each macro block include:

Two sub-blocks belonging to different types are distributed diagonally;

The sub-blocks belonging to different types are distributed in a row state; or

Stated distribution. The beneficial effects achieved by the present invention are as follows:

Since the video codec processing method of the present invention no longer assigns all subblock coding information in the macroblock MB to the same macroblock group SLICE, but classifies the subblocks in the MB, so that the various types of subblock coding information belong to different Macroblock group SLICE, or a submacroblock group SLICE belonging to the same macroblock group SLICE, so that during the video image decoding and restoration process, for the subblocks that have generated errors, the image information of other neighboring subblocks can be referred to, Insert the corresponding approximation image information, thereby achieving the purpose of higher accuracy of the inserted approximation information and better error concealment effect, thereby reducing the visual effect of the error code stream caused by the quality of the transmission channel. The image is distorted, which improves the effect of video communication.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic diagram of a compression process and a decompression process used by H. 264 in the prior art;

2 is a schematic diagram of a coding structure of each macroblock MB in the prior art;

3 is a schematic diagram of a processing principle of an FMO video encoding processing method in the prior art; FIG. 4 is a flowchart of an implementation process of a video encoding and decoding processing method of the present invention;

FIG. 5 is a first arrangement pattern diagram of dividing each MB into two types of sub-blocks of a size of 4 x 4 pixels in the video encoding and decoding processing method of the present invention;

6 is a second arrangement pattern diagram of dividing each MB into two types of sub-blocks of a size of 4 x 4 pixels in the video encoding and decoding processing method of the present invention;

FIG. Ί is a third arrangement pattern diagram of dividing each MB into two types of sub-blocks of 4 × 4 pixels in the video encoding and decoding processing method of the present invention;

FIG. 8 is a first arrangement pattern diagram of dividing each MB into two types of sub-blocks of 8 × 8 pixels size in the video encoding and decoding processing method of the present invention; FIG.

FIG. 9 is a second arrangement pattern diagram of dividing each MB into two types of sub-blocks of 8 × 8 pixels in the video encoding and decoding processing method of the present invention;

FIG. 10 is a third arrangement pattern diagram of dividing each MB into two types of sub-blocks of 8 × 8 pixels in the video encoding and decoding processing method of the present invention; FIG.

FIG. 11 is a schematic diagram of the coding prediction principle when a macro block acts as a SLICE; FIG. 12 is a schematic diagram of the coding prediction principle when two macro blocks act as a SLICE. DETAILED DESCRIPTION The basic design idea of the video encoding and decoding processing method of the present invention is: classify the sub-blocks in the macro block MB, so that the encoding information of each type of sub-block belongs to different macro block groups SLICE, or belongs to the same macro block group SLICE The sub-macro block group SLICE, so that during the image restoration and restoration process, the error code sub-block group caused by the transmission channel quality problem can be based on the image information of other adjacent sub-blocks around each of the error sub-blocks. The error sub-block performs the process of inserting the approximate image information, so that the inserted approximate image information has higher accuracy and better error concealment effect, so as to reduce d and the distortion degree of the video image in the visual angle. Corresponding to Flexible Macroblock Ordering (FMO) video coding processing method proposed by POLYCOM company, here will be

Ordering) encoding and decoding processing method, that is, FBO encoding and decoding processing method.

Please refer to FIG. 4, which is a flowchart of an implementation process of the video codec processing method of the present invention; the FBO codec processing process of the present invention includes the following steps:

Step S10: Transform subblock division processing is performed on each macroblock included in each macroblock group (that is, each image segment SLICE) after the video image is subjected to transform macroblock processing, so as to divide the Each macroblock MB is divided into multiple sub-blocks. If the size of each MB is 16 × 16 pixels, you can perform transform sub-block processing on each MB to obtain 4 X 4 pixels or 8 pixels. x 8 pixel-sized sub-blocks;

Step S20: Classify all the subblocks obtained after the transformation processing of each macroblock group in step S10, so that all the subblocks after the transformation processing in each macroblock group are grouped together according to type to form different Type sub-block groups, for example, all sub-blocks obtained after transform sub-block processing of each macro block group can be classified and processed according to the transform coefficients of each sub-block obtained from the transformation; as a result, various types of sub-block groups can be separately assigned To separate macroblock groups (that is, to form independent video segments SLICE), to split each corresponding macroblock group into multiple independent macroblock groups, such as each macroblock in a certain macroblock group SLICE0 is transformed After the sub-blocks are processed, all the corresponding sub-blocks are divided into 3 groups. Each group of sub-blocks is defined as a macro block group. That is, the original macro block group SLICE0 is split into 3 independent macro block groups. SLICE1, SLICE2, and SLICE3. As a result of the processing, each type of sub-block group is respectively assigned to the sub-macro block group of the corresponding macro-block group, so that each corresponding macro-block group originally has multiple sub-macro-block groups. For example, after each macroblock in a certain macroblock group SLICE0 is subjected to transform subblock processing, and then all the corresponding subblocks are divided into three groups, each group of subblocks is defined as one of the original macroblock group SLICE0. The sub-macroblock group, that is, the original macroblock group SLICE0 is composed of three sub-macroblock groups, which are SLICE0_1, SLICE0-2, and SLICE0_3.

In step S30, the video image sender encodes and sends each macroblock group header information, macroblock header information, and transform coefficients of each type of subblocks after the video image is divided. As shown in the above example, the video codec processing method of the present invention is In the video image encoding process, if the macro block group SLICE0 is encoded and sent, the header information of the macro block group SLICE0, the macro block header information contained in SLICE0, and the transform coefficients of each type of subblock included in SLICE0 are separately coded. .

In step S40, the video image receiver performs decoding processing on the correctly received code stream to recover and restore the image information of the corresponding subblock. In some cases, such as the poor quality of the transmission channel, the video image encoding code is often caused. The stream cannot reach the receiver completely correctly, but some stream data packets will be lost or some bit stream data packets will have error information, so on the receiver side, the corresponding sub-blocks where these error packets are located will have errors. Image information or image information with a high degree of distortion;

Step S50: After decoding and restoring the video image, for the sub-blocks that generate errors, according to the image information of other neighboring sub-blocks around the sub-blocks that generate errors, in the sub-blocks that generate errors Information processing for inserting and approximating images is performed to perform error concealment processing on a sub-block in which an error occurs.

The following description is based on dividing all the subblocks obtained after each macroblock group (SLICE) transformation process in a video image into two types to illustrate the two types of subblocks in each macroblock in the FBO encoding and decoding process of the present invention. Arrange pattern of groups.

The size of each sub-block obtained after transforming sub-blocks for each MB can be 4 X 4 pixels:

As shown in FIG. 5, this figure is a first arrangement pattern diagram of dividing each MB into two types of sub-blocks of 4 x 4 pixel size in the video encoding and decoding processing method of the present invention; among them, sub-blocks identified by a five-pointed star Forms a type of sub-block group, and the sub-blocks without a pentagram mark form another type of sub-block group, that is, respectively Two sub-blocks belonging to different types are distributed diagonally; in this way, a group of sub-blocks identified by a pentagram can belong to one SLICE alone, and a group of sub-blocks not identified by a pentagram can belong to another SLICE alone; or there are five-pointed stars The identified sub-block group and the sub-block group without the pentagram mark respectively form the sub-SLICEs under the original macro-block group SLICE, so that one SLICE is equivalent to being divided into two SLICEs.

Please refer to FIG. 6. This figure is a second arrangement pattern diagram of dividing each MB into two types of sub-blocks of 4 x 4 pixel size in the video encoding and decoding processing method of the present invention; One type of sub-block group, and the sub-blocks without a five-pointed star mark constitute another type of sub-block group, that is, sub-blocks belonging to different types are distributed in an interlaced state;

Please refer to FIG. 7, which is a third arrangement pattern diagram of dividing each MB into two types of sub-blocks with a size of 4 x 4 pixels in the video encoding and decoding processing method of the present invention; One type of sub-block group, and the sub-blocks without a five-pointed star mark form another type of sub-block group, that is, the sub-blocks belonging to different types are distributed in an alternate column state.

In addition, the size of each sub-block obtained after transforming sub-blocks for each MB can also be 8 X 8 pixels:

Please refer to FIG. 8. This figure is a first arrangement pattern diagram of dividing each MB into two types of sub-blocks of 8 x 8 pixel size in the video encoding and decoding processing method of the present invention; One type of sub-block group, and the sub-blocks without a five-pointed star mark constitute another type of sub-block group, that is, two sub-blocks respectively belonging to different types are distributed diagonally.

Please refer to FIG. 9, which is a second arrangement pattern diagram of dividing each MB into two types of sub-blocks of 8 x 8 pixel size in the video encoding and decoding processing method of the present invention; One type of sub-block group, and the sub-blocks without a five-pointed star mark form another type of sub-block group, that is, the sub-blocks belonging to different types are distributed in a row state.

Please refer to FIG. 10, which is a third arrangement pattern diagram of dividing each MB into two types of sub-blocks of the size of 8 x 8 pixels in the video encoding and decoding processing method of the present invention; One type of sub-block group, and the sub-blocks without a five-pointed star mark form another type of sub-block group, that is, sub-blocks belonging to different types are distributed in a column state.

Of course, regardless of whether each MB is divided into 4 x 4 pixel-sized subblocks or 8 x 8 pixel-sized subblocks, the different types of subblock arrangement modes are not limited to those mentioned above. In the embodiment, according to specific situations, there may be other arrangement modes, and other different types of sub-block arrangement modes are also within the scope of the present invention.

Moreover, the above-mentioned different types of subblock arrangement modes are described based on the structure of the luminance block y, chrominance block Cr, and chrominance block Cb in the macro block MB in the video image format 4: 2: 0. The image format is two cases of 4: 2: 2 and 4: 4: 4, the principle is the same as 4: 2: 0, so I won't go into details here.

Since the video encoding and decoding processing method of the present invention is based on FBO, that is, the video encoding and decoding operations are performed in the SLICE manner in which subblocks in a macroblock MB belong to different macroblock groups, it will cause corresponding changes in some standard processes in the encoding and decoding process. Such as:

1) The video image sender and the video image receiver need to perform de-blocking on adjacent sub-blocks of the same type, as shown in FIG. 6, FIG. 7, FIG. 8, FIG. 9, and FIG. 10 De-blocking processing is performed on adjacent 4X4 pixel sub-blocks with a five-pointed star mark, and De-blocking processing is performed on adjacent 4X4 pixel sub-blocks without a five-pointed star mark; and the 4X4 pixel sub-blocks in the arrangement pattern shown in FIG. There is no need to perform De-blocking, because in this arrangement mode, no sub-blocks of the same type exist adjacently.

2) The video image sender and the video image receiver should perform intra prediction on the image information in the subblocks belonging to the same type.

3) The codec motion estimation and motion compensation of the video image sender and the video image receiver are basically unaffected, but it is necessary to refer to the motion vectors of the same type of subblocks when performing motion vector prediction.

4) The video image sender and the video image receiver can only use the information of the same type of sub-block part when considering the sub-block transform coefficient encoding and context consideration.

5) The problem of redundant parameter information, as in the above example of dividing an original SLICE into two independent SLICEs, because for each independent SLICE part, only half of the image information of the original SLICE part is encoded, so the relative In terms of the FBO video codec method, the number of macroblocks included in a SLICE can be nearly doubled compared to the regular macroblocks. However, the header information of the macroblock obviously repeats some common parts and belongs to different types of sub-blocks. The block header information can be separately encoded in the corresponding independent SLICE part, such as block motion vector information. 6) The encoding efficiency problem can be considered from two aspects:

On the one hand, because the FBO codec processing method of the present invention assigns different types of subblocks of each macroblock MB to different macroblock groups SLICE, it is obvious that the correlation of adjacent blocks in the same SLICE part is reduced to a certain extent. The coding efficiency will definitely be reduced to some extent.

On the other hand, referring to FIG. 11, this figure is a schematic diagram of the principle of encoding prediction when a macroblock acts as a SLICE. In this case, when one macroblock line is encoded one by one, generally only the left macroblock information can be referred to. Make predictions; refer to FIG. 12, which is a schematic diagram of the principle of encoding prediction when two macroblocks act as one SLICE; for the case where two macroblocks act as one SLICE, the macroblocks in the following macroblock row can refer to the macroblock The top left corner macroblock information, the top right macroblock information, the top right corner macroblock information, and the left side macroblock information are used to perform coding prediction. It can be seen that due to the improvement of the coding prediction performance, the coding efficiency can be improved to a certain extent (the above reasons are Because in order to improve the efficiency of H.264 intra-frame coding, the spatial correlation of adjacent macroblocks is fully utilized in a given frame. Adjacent macroblocks usually contain similar attributes. Therefore, when encoding a given macroblock, First of all, it can be predicted based on the surrounding macroblocks. Typically, it is based on the macroblock in the upper left corner, because this macroblock has been encoded and processed. The difference between the measured value and the actual value is encoded, so that, with respect to the direct coding for this frame, can be greatly reduced rate).

Of course, considering the impact of the two aspects above, it seems that the encoding efficiency of FBO encoding and decoding methods will still be reduced to a certain extent.

In specific practical applications, according to specific applications, FMO and FBO codec processing methods can be selectively used, and at the same time, the arrangement modes of different types of sub-blocks can be selectively used, such as dividing multiple sub-blocks according to the foregoing For two types of cases, the following settings can be made: '

1. Establish an FBO logo "FBO— label"

When FBO_ label = 0, the FMO codec processing method of the prior art is selected, which is mainly applied to transmission channels with lower bit error rates;

When FBO_label = 1, the FBO encoding and decoding processing method of the present invention is selected, which is mainly applied to a transmission channel with a poor bit error rate;

2. For a SLICE in a video image, set up a main logo Sub- SLICE, with values of 0 and 1 respectively corresponding to the two sub-SLICEs that are split, or they can be split. The last two independent SLICE parts are respectively set to Sub_SLICE;

3. Establish an FBO mode identifier "FBO_mode". The value of FBO_mode 0 ~ 5 corresponds to the six FBO arrangement modes shown in Figure 5 to Figure 10.

4. The selection of six FBO arrangement modes can be selected according to the comprehensive consideration of system coding performance requirements and error resistance performance requirements. Of course, considering the calculation complexity in the specific processing process, only six FBO arrangement modes can be selected. For example, only the FBO arrangement patterns shown in FIG. 5 and FIG. 8 may be taken.

It can be seen that, in the FBO video codec processing method of the present invention, when the image information is restored and restored, for a distorted image due to bit errors, block information of the size of 4 x 4 pixels or 8 x 8 pixels can be used as the reference information pair. The processing of inserting the approximation information into the wrong block is performed. Compared with the prior art, which uses a 16 x 16 pixel size macroblock to perform the information approximation processing, the accuracy of the inserted approximation information will be higher, and the image restoration effect will be more accurate. a little better.

The above mainly describes the example of dividing all sub-blocks obtained after transforming sub-block processing on the same macro block group into two types, where all sub-blocks obtained after transforming sub-block processing on the same macro block group are divided The implementation for multiple types is similar to the case where it is divided into two types, which is not described in detail here, but its implementation for dividing all sub-blocks into multiple types is also included in the scope of the present invention.

The FBO codec processing method of the present invention is a modification based on the H.264 reference program JM76. For some test sequences, the FBO codec processing method proposed by the present invention is tested, and the results show that better errors can be obtained. The masking effect reduces the image distortion caused by the transmission channel quality problem, and improves the video communication effect better.

The above is only a preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the technical principles of the present invention, several improvements and retouches can be made. These improvements and retouches also It should be regarded as the protection scope of the present invention.

Claims

Rights request

1. A video codec processing method, comprising:

C. The sender performs encoding processing on each macroblock group header information, the macroblock header information, and the transformation coefficient of each type of subblock group and sends them;

D. The receiver separately decodes the correctly received bitstream to recover the image information of the corresponding sub-block. For each sub-block that generates an error, the receiver inserts the image based on the image information of the corresponding neighboring sub-block. Approaching image information processing.

2. The video codec processing method according to claim 1, further comprising the step after the step B: class sub-block groups are respectively belonged to independent macro block groups, so that each original macro block group is correspondingly Split into multiple independent macroblock groups.

3. The video codec processing method according to claim 1, further comprising steps after the step B:

According to the classification processing result of all the sub-blocks after each macro block group transformation process, each type of sub-block group is respectively assigned to the sub-macro block group of the corresponding macro block group, so that each corresponding macro block group originally contains multiple Sub-macroblock groups.

4. The video codec processing method according to claim 1, wherein in steps C and D, the sender and receiver further perform deblocking processing on adjacent sub-blocks belonging to the same type.

5. The video codec processing method according to claim 1, wherein in steps C and D, the sender and receiver further perform intra prediction on image information in subblocks belonging to the same type.

6. The video codec processing method according to claim 1, wherein the sub-blocks after each macroblock group transform process are divided into two types of sub-block groups according to the sub-block transform coefficients.

7. The video codec processing method according to claim 6, wherein the size of each sub-block is 4 x 4 pixels.

8. The video codec processing method according to claim 7, wherein the arrangement mode of the two types of subblock groups in each macroblock comprises:

Two sub-blocks belonging to different types are distributed diagonally;

Distributed in a state of columns.

9. The video codec processing method according to claim 6, wherein the size of each sub-block is 8 x 8 pixels.

10. The video codec processing method according to claim 9, wherein the arrangement mode of the two types of subblock groups in each macroblock comprises:

Two sub-blocks belonging to different types are distributed diagonally;

The sub-blocks belonging to different types are distributed in a row state; or

Stated distribution.