CN113411584A

CN113411584A - Video coding and decoding method and device

Info

Publication number: CN113411584A
Application number: CN202010188807.8A
Authority: CN
Inventors: 周川; 吕卓逸; 欧阳晓
Original assignee: Samsung Electronics Co Ltd
Current assignee: Beijing Samsung Telecom R&D Center; Samsung Electronics Co Ltd
Priority date: 2020-03-17
Filing date: 2020-03-17
Publication date: 2021-09-17
Also published as: EP4122198A1; KR102585236B1; EP4122198A4; KR20220124287A; WO2021187855A1

Abstract

The invention provides a coding and decoding method and device. A method of decoding comprising: obtaining a prediction sample matrix of a current prediction block; and obtaining a reconstructed sample matrix according to the predicted sample matrix, wherein the element values of the predicted sample matrix are obtained according to the sample values in the reconstructed sample matrix after the current image is filtered.

Description

Video coding and decoding method and device

Technical Field

The present invention relates to the field of video coding and decoding, and in particular to block copy intra prediction for intra prediction mode.

Background

In a video encoding and decoding method, an image is generally divided into a plurality of image blocks, and then each image block is encoded or decoded. For each image block, the encoding step can be divided into prediction, transformation, quantization and entropy encoding, wherein the prediction is to predict the current image block to be encoded by using the reconstructed pixel values (these pixels are referred to as reference pixels) of the image block that has been encoded before to derive a predicted value, and then encode the difference between the actual value and the predicted value of the current image block into the code stream. During decoding, the decoder also needs to predict the current image block to be decoded by using the reconstructed pixel value of the image block that has been decoded before to derive a predicted value, and then adds the difference value obtained by decoding from the code stream to the predicted value to obtain the reconstructed value of the decoded image block. In order to ensure the consistency of the codec, the codec must use the same reference pixels and the same prediction method when performing prediction. There are many specific prediction methods, and the encoder selects a current image block and then writes information about the selected prediction method into the code stream to tell the decoder, so that the decoder can predict the current decoded block using the same prediction method.

Disclosure of Invention

Technical problem

There is a need for improved techniques for block copy intra prediction for intra prediction modes in video coding.

Solution scheme

There is provided a method of decoding, comprising: obtaining a prediction sample matrix of a current prediction block; and obtaining a reconstructed sample matrix according to the predicted sample matrix, wherein the element values of the predicted sample matrix are obtained according to the sample values in the reconstructed sample matrix after the current image is filtered.

There is provided an apparatus for decoding, comprising: a memory in which a computer program is stored, and a processor which, when running the computer program, performs the method of decoding as described above.

There is provided a method of encoding comprising: determining an identifier indicating whether deblocking filtering is used for the current frame according to whether the current frame to be coded is screen content; and writing the identification into a code stream.

There is provided an apparatus for encoding, comprising: a memory in which a computer program is stored, and a processor which, when running the computer program, executes the method of encoding as described above.

There is provided a method for decoding, comprising: determining whether to perform deblocking filtering on a prediction compensated sample of a current decoding unit according to whether the current decoding unit is screen content; and according to the determined result, performing deblocking filtering on the prediction compensated sample of the current decoding unit to obtain a reconstructed sample after deblocking filtering.

A computer readable storage medium is provided for storing non-transitory computer readable instructions which, when executed by a computer, cause the computer to perform the method as previously described.

Advantageous effects

The present invention provides improved techniques for block copy intra prediction for intra prediction modes in video coding. Other aspects, advantages and salient features of the invention will become apparent to those skilled in the art from the following detailed description, which, taken in conjunction with the annexed drawings, discloses exemplary embodiments of the invention.

Drawings

Fig. 1 shows a schematic diagram of dividing an image into a maximum decoding unit and a decoding unit.

Fig. 2 shows a schematic diagram of block copy intra prediction modes.

Fig. 3 shows a flow chart of a method of decoding according to an embodiment of the invention.

Fig. 4 shows a schematic diagram of a reference block at different positions when the filtered reconstructed sample matrix includes decoded prediction-compensated samples in the current maximum decoding unit and samples obtained by deblocking filtering and sample offset compensation of the prediction-compensated samples of the decoded maximum decoding unit.

Fig. 5 is a schematic diagram showing that the reference block is located at different positions when the filtered reconstructed sample matrix includes samples obtained by deblocking filtering the decoded prediction compensated samples in the current maximum decoding unit and samples obtained by deblocking filtering and sample offset compensation the prediction compensated samples of the decoded maximum decoding unit.

Fig. 6 shows a schematic diagram of the reference block being located at different positions when the filtered reconstructed sample matrix includes samples obtained by sample offset compensation of the prediction compensated samples decoded in the current maximum decoding unit and the prediction compensated samples of the decoded maximum decoding unit.

Fig. 7 shows a block diagram of an apparatus for decoding according to an embodiment of the present invention.

Fig. 8 shows a flow diagram of a method of encoding according to an embodiment of the invention.

Fig. 9 shows a schematic diagram of filter block boundary samples.

Fig. 10 shows a block diagram of an apparatus for encoding according to an embodiment of the present invention.

Fig. 11 shows a flow diagram of a method of decoding according to an embodiment of the invention.

Fig. 12 shows a block diagram of an apparatus for decoding according to an embodiment of the present invention.

Detailed Description

Embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

The terms and words used in the following description are not intended to be limited to the literal meanings, but are merely used to enable a clear and consistent understanding of the invention. Accordingly, one skilled in the art would understand that: the following description of exemplary embodiments of the present invention is provided for the purpose of illustration only and not for the purpose of limiting the invention as defined by the claims and their equivalents.

It should be understood that: the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise.

Fig. 1 shows a schematic diagram of dividing an image into a maximum decoding unit and a decoding unit. As shown in the left diagram of fig. 1, the image is divided into a series of maximum decoding units, and the maximum decoding units are decoded in sequence in the raster scanning order within the slice. As shown in the right diagram of fig. 1, the maximum decoding unit adopts a basic block partition structure of QT + BT + EQT, QT divides one decoding unit into four sub-decoding units, BT can divide one decoding unit into two sub-decoding units, left and right/up and down, EQT includes two horizontal and vertical i-shaped partition modes, and divides one decoding unit into 4 sub-decoding units. The prediction unit comprises a luminance prediction block and a corresponding chrominance prediction block which are obtained by dividing the decoding unit. If the decoding unit type of the decoding unit is 'I _2M _ 2N', I _2M _ hN ', I _2M _ nU', I _2M _ nD ', I _ hM _ 2N', 'I _ nL _ 2N' or 'I _ nR _ 2N', the prediction type is normal intra prediction, and the prediction type of the prediction block included in the decoding unit is normal intra prediction; otherwise, if the decoding unit type is 'IBC _2M _ 2N', the prediction type is block copy intra prediction, and the prediction type of the prediction block included therein is block copy intra prediction; otherwise, the prediction type of the decoding unit is inter prediction, and the prediction types of the prediction unit and the prediction block included in the decoding unit are inter prediction.

For a sequence of screen contents such as characters and graphics, a lot of repeated textures exist in the same frame, namely, the spatial correlation is strong. If the coded region of the current frame can be referred to when the current block is coded, the coding efficiency (aiming at the characteristic of strong spatial correlation of the screen image) can be greatly improved. This technique, referred to as Intra Block Copy (IBC), refers to a block that has been encoded for the current block and is used to predict the current block. Fig. 2 shows a schematic diagram of block copy intra prediction modes. Block copy intra prediction is similar to inter picture prediction, except that the prediction block of block copy intra prediction is generated from a reconstructed block of the current encoded picture frame.

Deblocking filter (DBF) is used to reduce blocking artifacts; sample Adaptive Offset (SAO) is used to improve ringing, and these two filters can effectively improve the subjective and objective quality of video.

In the existing method, the values of IBC prediction sample matrix elements are obtained from sample values in a reconstructed sample matrix of a current image without filtering. The non-filtered reconstructed samples are not processed by deblocking filtering and sample offset compensation, which results in large distortion of predicted sample values, resulting in reduced quality and coding efficiency of prediction blocks using IBC.

In order to solve the above problems, the present invention provides a block copy intra prediction method, in which the value of IBC prediction sample matrix element is obtained according to the sample value in the reconstructed sample matrix after the current image is filtered, so as to further improve the screen content image coding efficiency.

Fig. 3 shows a flow chart of a method of decoding according to an embodiment of the invention. The method comprises the following steps: obtaining a prediction sample matrix of a current prediction block (S301); and obtaining a reconstructed sample matrix according to the predicted sample matrix, wherein the element values of the predicted sample matrix are obtained according to the sample values in the reconstructed sample matrix after the current image is filtered (S302). Therefore, the distortion of the prediction sample value can be reduced, the quality of the prediction block adopting IBC is improved, and the coding efficiency is improved.

The element values of the prediction sample matrix are obtained according to sample values in the reconstructed sample matrix after the current image is filtered, and the method comprises the following steps: the element values of the luma prediction sample matrix are the sample values in the integer-pixel-precision sample matrix reconstructed after filtering the current image, and the element values of the chroma prediction sample matrix are the sample values in the 1/2-precision chroma sample matrix reconstructed after filtering the current image.

According to one embodiment, the filtered reconstructed sample matrix comprises: the sample after prediction compensation decoded in the current maximum decoding unit and the sample obtained by deblocking filtering and sample value offset compensation of the sample after prediction compensation of the decoded maximum decoding unit. The prediction compensated sample is the sum of the prediction sample and the residual sample. Therefore, the distortion of the prediction sample value can be reduced, the quality of the prediction block adopting IBC is improved, and the coding efficiency is improved.

For example, when the filtered reconstructed sample matrix includes a sample obtained by deblocking filtering and sample offset compensation of a prediction compensated sample decoded in a current maximum decoding unit and a prediction compensated sample of the decoded maximum decoding unit, at least one of the following cases exists: if the reference block pointed by the block vector is completely located outside the current maximum decoding unit (i.e. located inside the decoded maximum decoding unit), all the element values of the prediction sample matrix are obtained according to the sample values obtained by performing deblocking filtering and sample offset compensation on the prediction compensated sample of the decoded maximum decoding unit, as shown in the left diagram of fig. 4; if a part of the reference block pointed by the block vector is located outside the current maximum decoding unit (i.e. inside the decoded maximum decoding unit) and another part is located inside the current maximum decoding unit, then a part of the element values of the prediction sample matrix is obtained from the sample values obtained by deblocking filtering and sample offset compensation of the prediction compensated samples of the decoded maximum decoding unit, and another part is obtained from the prediction compensated sample values decoded in the current maximum decoding unit, as shown in the diagram of fig. 4; alternatively, if the reference block pointed to by the block vector is located completely within the current maximum decoding unit, the element values of the prediction sample matrix are all obtained from the prediction compensated sample values decoded in the current maximum decoding unit, as shown in the right diagram of fig. 4.

According to another embodiment, the filtered reconstructed sample matrix comprises: the sample obtained by deblocking filtering the prediction compensated sample decoded in the current maximum decoding unit, and the sample obtained by deblocking filtering and sample value offset compensation of the prediction compensated sample of the decoded maximum decoding unit. Therefore, the distortion of the prediction sample value can be reduced, the quality of the prediction block adopting IBC is improved, and the coding efficiency is improved.

For example, when the filtered reconstructed sample matrix includes a sample obtained by deblocking filtering the decoded prediction compensated sample in the current maximum decoding unit and a sample obtained by deblocking filtering and sample offset compensating the prediction compensated sample of the decoded maximum decoding unit, at least one of the following conditions exists: if the reference block pointed by the block vector is completely located outside the current maximum decoding unit (i.e. located inside the decoded maximum decoding unit), all the element values of the prediction sample matrix are obtained according to the sample values obtained by deblocking filtering and sample offset compensation of the prediction compensated sample of the decoded maximum decoding unit, as shown in the left diagram of fig. 5; if a part of the reference block pointed by the block vector is located outside the current maximum decoding unit (i.e. inside the decoded maximum decoding unit) and another part is located inside the current maximum decoding unit, a part of the element values of the prediction sample matrix is obtained from the sample values obtained by deblocking filtering and sample offset compensation of the prediction compensated samples of the decoded maximum decoding unit, and another part is obtained from the sample values obtained by deblocking filtering of the prediction compensated samples of the decoded maximum decoding unit, as shown in fig. 5; or, if the reference block pointed by the block vector is located completely within the current maximum decoding unit, the element values of the prediction sample matrix are all obtained according to the sample values obtained by deblocking filtering of the prediction compensated samples decoded in the current maximum decoding unit, as shown in the right diagram of fig. 5.

According to an embodiment, the reconstructed sample values may or may not be deblock filtered, indicated by an indication of whether or not deblocking filtering is used at the frame level obtained from the bitstream. Thus, for example, the method of decoding further comprises obtaining from the codestream an indication of whether or not deblocking filtering is to be used for the current frame. For example, 0 indicates that all blocks in the current frame to be decoded do not use the deblocking filter, and 1 indicates that all blocks in the current frame to be decoded use the deblocking filter.

According to an embodiment, when the indication indicates that no deblocking filtering is used for the current frame (e.g. the indication is 0), the filtered reconstructed sample matrix comprises: the sample after prediction compensation decoded in the current maximum decoding unit and the sample obtained by sample value offset compensation of the sample after prediction compensation of the decoded maximum decoding unit. Therefore, the distortion of the prediction sample value can be reduced, the quality of the prediction block adopting IBC is improved, and the coding efficiency is improved.

For example, when the filtered reconstructed sample matrix includes samples obtained by sample offset compensation of the prediction compensated samples decoded in the current maximum decoding unit and the prediction compensated samples of the decoded maximum decoding unit, at least one of the following situations exists: if the reference block pointed by the block vector is completely located outside the current maximum decoding unit (i.e. located inside the decoded maximum decoding unit), all the element values of the prediction sample matrix are obtained according to the sample values obtained by sample offset compensation of the prediction compensated samples of the decoded maximum decoding unit, as shown in the left diagram of fig. 6; if a part of the reference block pointed by the block vector is located outside the current maximum decoding unit (i.e. inside the decoded maximum decoding unit) and another part is located inside the current maximum decoding unit, a part of the element values of the prediction sample matrix is obtained from the sample values obtained by sample offset compensation of the prediction compensated samples of the decoded maximum decoding unit, and another part is obtained from the prediction compensated sample values decoded in the current maximum decoding unit, as shown in the diagram of fig. 6; or if the reference block pointed by the block vector is completely located in the current maximum decoding unit, the element values of the prediction sample matrix are all obtained according to the prediction compensated sample values decoded in the current maximum decoding unit, as shown in the right diagram of fig. 6.

When the flag indicates that deblocking filtering is used for the current frame (e.g., the flag is 1), there are two cases.

According to one embodiment, when the indication indicates that deblocking filtering is used for the current frame (e.g., the indication is 1), the filtered reconstructed sample matrix includes: the sample after prediction compensation decoded in the current maximum decoding unit and the sample obtained by deblocking filtering and sample value offset compensation of the sample after prediction compensation of the decoded maximum decoding unit. As described above, this can reduce distortion of prediction sample values, improve the quality of prediction blocks using IBC, and improve coding efficiency. Although deblocking filtering is used at the frame level, a decoding unit in the current maximum decoding unit may not use deblocking filtering.

As described above, for example, when the filtered reconstructed sample matrix includes the decoded prediction-compensated samples in the current maximum decoding unit and the samples obtained by deblocking filtering and sample offset compensation of the prediction-compensated samples of the decoded maximum decoding unit, at least one of the following cases exists: if the reference block pointed by the block vector is completely located outside the current maximum decoding unit (i.e. located inside the decoded maximum decoding unit), all the element values of the prediction sample matrix are obtained according to the sample values obtained by performing deblocking filtering and sample offset compensation on the prediction compensated sample of the decoded maximum decoding unit, as shown in the left diagram of fig. 4; if a part of the reference block pointed by the block vector is located outside the current maximum decoding unit (i.e. inside the decoded maximum decoding unit) and another part is located inside the current maximum decoding unit, then a part of the element values of the prediction sample matrix is obtained from the sample values obtained by deblocking filtering and sample offset compensation of the prediction compensated samples of the decoded maximum decoding unit, and another part is obtained from the prediction compensated sample values decoded in the current maximum decoding unit, as shown in the diagram of fig. 4; alternatively, if the reference block pointed to by the block vector is located completely within the current maximum decoding unit, the element values of the prediction sample matrix are all obtained from the prediction compensated sample values decoded in the current maximum decoding unit, as shown in the right diagram of fig. 4.

According to another embodiment, when the indication indicates that deblocking filtering is used for the current frame (e.g. the indication is 1), the filtered reconstructed sample matrix comprises: the sample obtained by deblocking filtering the prediction compensated sample decoded in the current maximum decoding unit, and the sample obtained by deblocking filtering and sample value offset compensation of the prediction compensated sample of the decoded maximum decoding unit. As described above, this can reduce distortion of prediction sample values, improve the quality of prediction blocks using IBC, and improve coding efficiency.

As mentioned above, for example, when the filtered reconstructed sample matrix includes samples obtained by deblocking filtering the decoded prediction compensated samples in the current maximum decoding unit and samples obtained by deblocking filtering and sample offset compensation the prediction compensated samples in the decoded maximum decoding unit, at least one of the following conditions may exist: if the reference block pointed by the block vector is completely located outside the current maximum decoding unit (i.e. located inside the decoded maximum decoding unit), all the element values of the prediction sample matrix are obtained according to the sample values obtained by deblocking filtering and sample offset compensation of the prediction compensated sample of the decoded maximum decoding unit, as shown in the left diagram of fig. 5; if a part of the reference block pointed by the block vector is located outside the current maximum decoding unit (i.e. inside the decoded maximum decoding unit) and another part is located inside the current maximum decoding unit, a part of the element values of the prediction sample matrix is obtained from the sample values obtained by deblocking filtering and sample offset compensation of the prediction compensated samples of the decoded maximum decoding unit, and another part is obtained from the sample values obtained by deblocking filtering of the prediction compensated samples of the decoded maximum decoding unit, as shown in fig. 5; or, if the reference block pointed by the block vector is located completely within the current maximum decoding unit, the element values of the prediction sample matrix are all obtained according to the sample values obtained by deblocking filtering of the prediction compensated samples decoded in the current maximum decoding unit, as shown in the right diagram of fig. 5.

Fig. 7 shows a block diagram of an apparatus for decoding according to an embodiment of the present invention. The decoding apparatus 700 includes: a memory (701) in which a computer program is stored, and a processor (702) which, when running the computer program, performs the method of decoding as described hereinbefore. The device can reduce the distortion of the prediction sample value, improve the quality of the prediction block adopting IBC and improve the coding efficiency.

Fig. 8 shows a flow diagram of a method of encoding according to an embodiment of the invention. The present embodiment relates to the method for obtaining the indication of whether the previous frame level uses deblocking filtering. The method comprises the following steps: determining an identifier indicating whether deblocking filtering is used for the current frame according to whether the current frame to be encoded is screen content (S801); and writing the identification into the code stream (S802). This can improve coding efficiency.

The screen content is an image directly captured from an image display unit of a computer or a mobile terminal or the like. For example, computer graphics, text images, images in which natural images are mixed with graphics and text, computer-generated animation images, and the like, and such screen content images are ubiquitous in application scenes such as desktop collaboration, desktop sharing, cloud games, and the like. The main difference between a screen content image and a natural image captured with a camera is that the screen content image is noise free, discrete in tone and sharp in edge, whereas a natural image is typically noisy, continuous in tone and complex in texture. The method for determining whether the screen content is calculated may be one of the following methods, or may be another method, which is not limited herein. One possible implementation method is that a hash (hash) table is constructed before each frame is coded, for each hash value, whether the value corresponds to 2 or more blocks is calculated, and when the probability of the occurrence of the above conditions is greater than a certain threshold value, the current frame is considered as screen content; another possible implementation method is that each frame is convolved with sobel operator before encoding, and when the calculated sobel value of the whole frame is greater than a certain threshold, the current frame is the screen content.

According to an embodiment, step S801 comprises at least one of: the identification indicates that no deblocking filtering is used for the current frame (e.g., identified as 0) if the current frame to be encoded is screen content, or indicates that deblocking filtering is used for the current frame (e.g., identified as 1) if the current frame to be encoded is not screen content.

According to an embodiment, the method of encoding further comprises: and determining whether to carry out deblocking filtering on the prediction compensated samples of the current frame to be coded based on the identification. When the identification indicates that no deblocking filtering is used for the current frame (e.g., identification is 0), no deblocking filtering is used for the filtering block of the current frame; and when the flag indicates that deblocking filtering is used for the current frame (e.g., flag 1), it is determined whether deblocking filtering is used for the filtering block of the current frame.

According to an embodiment, the step of determining whether to deblock filter the prediction compensated samples of the current frame to be encoded based on the identification comprises calculating a filter strength of a filter block boundary, the calculating a filter strength of a filter block boundary comprising at least one of: if the boundary is judged to need filtering, when the absolute value of (p0-q0) is greater than or equal to four times of a first parameter, the filtering strength is 0, wherein p0 is the 1 st sample on the left side or the upper side of the boundary, q0 is the 1 st sample on the right side or the lower side of the boundary, and the first parameter is obtained through calculation according to the quantization parameter and the bit depth; when the smoothness of the boundary is equal to 6, if the absolute value of (p0-p1) is less than or equal to one fourth of the second parameter and the absolute value of (q0-q1) is less than or equal to one fourth of the second parameter and the absolute value of (p0-p3) is less than or equal to one half of the second parameter and the absolute value of (q0-q3) is less than or equal to one half of the second parameter and the absolute value of (p0-q0) is less than the first parameter, the filter strength is equal to 4, wherein p1 is the 2 nd sample on the left or upper side of the boundary, q1 is the 2 nd sample on the right or lower side of the boundary, p3 is the 4 th sample on the left or upper side of the boundary, q3 is the 4 th sample on the right or lower side of the boundary, the second parameter is obtained by quantization parameter and bit depth calculation; or when the smoothness of the boundary is equal to 5, if p0 is equal to p1 and q0 is equal to q1 and the absolute value of (p2-q2) is less than the first parameter, the filtering strength is equal to 3, where p2 is the 3 rd sample on the left or upper side of the boundary and q2 is the 3 rd sample on the right or lower side of the boundary.

Fig. 9 shows a schematic diagram of filter block boundary samples. In fig. 9, black bold lines indicate a certain segment of filter block boundary, 8 samples on both sides of the segment are respectively denoted as p0, p1, p2, p3, q0, q1, q2, and q3, that is, p0 is the 1 st sample on the left side or the upper side of the boundary, p1 is the 2 nd sample on the left side or the upper side of the boundary, p2 is the 3 rd sample on the left side or the upper side of the boundary, p4 is the 4 th sample on the left side or the upper side of the boundary, q0 is the 1 st sample on the right side or the lower side of the boundary, q1 is the 2 nd sample on the right side or the lower side of the boundary, q2 is the 3 rd sample on the right side or the lower side of the boundary, and q3 is the 4 th sample on the right side or the lower side of the boundary. And judging whether each boundary of the filtering block needs to be filtered or not. If filtering is required, the filtering strength of each segment boundary of the strip boundary is calculated. Performing deblocking filtering according to the filtering strength of the section boundary; otherwise, the value of the compensated sample is directly used as the value of the sample after deblocking filtering.

For example, the derivation process of the deblocking boundary filtering strength Bs is as follows, the greater the value of the boundary filtering strength Bs, the greater the filtering strength, Bs is 0, which means no filtering is performed:

if it is judged that filtering is required, Bs is equal to 0 when the absolute value of (p0-q0) is greater than or equal to four times the first parameter calculated from the Quantization Parameter (QP) and the bit depth (bitdepth). Otherwise, the process of calculating the boundary filtering strength Bs is as follows:

computing the smoothness of the boundary fS:

if the absolute value of (p0-p1) is less than the second parameter, the boundary left/top smoothness is increased by 2, and the second parameter is calculated according to the quantization parameter and the bit depth. If the absolute value of (p0-p2) is less than the second parameter, the boundary left/top smoothness is increased by 1. The right/bottom smoothness is obtained in the same way. The smoothness of the boundary fS is equal to the sum of the left/top and right/bottom smoothness.

Secondly, obtaining Bs according to the value of the smoothness fS of the boundary:

when fS is equal to 6, Bs is equal to 4 if the absolute value of (p0-p1) is less than or equal to one-fourth of the second parameter and the absolute value of (q0-q1) is less than or equal to one-fourth of the second parameter and the absolute value of (p0-p3) is less than or equal to one-half of the second parameter and the absolute value of (q0-q3) is less than or equal to one-half of the second parameter and the absolute value of (p0-q0) is less than the first parameter; else Bs equals 3.

When fS is equal to 5, Bs is equal to 3 if p0 is equal to p1 and q0 is equal to q1 and the absolute value of (p2-q2) is less than the first parameter; else Bs equals 2.

Fig. 10 shows a block diagram of an apparatus for encoding according to an embodiment of the present invention. The apparatus 1000 for encoding comprises: a memory (1001) in which a computer program is stored, and a processor (1002) which, when running the computer program, performs the method of encoding as described hereinbefore. The coding device can improve coding efficiency.

Fig. 11 shows a flow diagram of a method of decoding according to an embodiment of the invention. The method relates to whether deblocking filtering is used at the block level. The method comprises the following steps: determining whether to perform deblocking filtering on the prediction-compensated samples of the current decoding unit according to whether the current decoding unit is screen content (S1101); and according to the determined result, performing deblocking filtering on the prediction compensated sample of the current decoding unit to obtain a reconstructed sample after deblocking filtering (S1102). This can improve the decoded image quality.

According to an embodiment, step S1101 comprises at least one of: it is determined to use deblocking filtering on reconstructed sample values of the current decoded block if the current decoded block is not screen content, or to not use deblocking filtering on reconstructed sample values of the current decoded block if the current decoded block is screen content.

The current decoding unit may be a maximum decoding block or a decoding block, which is not limited herein.

The method for determining whether the screen content is calculated may be one of the following methods, or may be another method, which is not limited herein. One possible implementation method is that the current decoded block is convolved by a sobel operator, and when the calculated sobel value of the current decoded block is greater than a certain threshold value, the current decoded block is screen content; another possible implementation method is to count the histogram of the current decoded block, and when the calculated sample value type of the current decoded block is less than a certain threshold, the current decoded block is the screen content.

Fig. 12 shows a block diagram of an apparatus for decoding according to an embodiment of the present invention. The decoding apparatus 1200 includes: a memory (1201) in which a computer program is stored, and a processor (1202) which, when running the computer program, performs the method of decoding described with reference to fig. 11. The decoding device can improve the quality of decoded images.

The present invention also provides a computer readable storage medium for storing non-transitory computer readable instructions which, when executed by a computer, cause the computer to perform a method as described above.

The example embodiments described herein are not meant to be limiting. The aspects of the present disclosure, as generally described herein, and illustrated in the figures, can be arranged, substituted, combined, separated, and designed in a wide variety of different configurations, all of which are contemplated herein. Furthermore, the features shown in each figure may be used in combination with each other, unless the context indicates otherwise. Thus, the drawings are to be generally regarded as forming a part of one or more general embodiments, but it is to be understood that not all illustrated features are required for each embodiment.

While the invention has been shown and described with reference to certain embodiments thereof, it will be understood by those skilled in the art that various changes and modifications in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims

1. A method of decoding, comprising:

obtaining a prediction sample matrix of a current prediction block;

obtaining a reconstructed sample matrix according to the prediction sample matrix,

wherein, the element value of the prediction sample matrix is obtained according to the sample value in the reconstructed sample matrix after the current image is filtered.

2. The method of claim 1, wherein the filtered reconstructed sample matrix comprises:

the prediction compensated samples decoded in the current maximum decoding unit, and

and the prediction compensated sample of the decoded maximum decoding unit is subjected to deblocking filtering and sample value offset compensation to obtain a sample.

3. The method of claim 2, wherein the values of the elements of the prediction sample matrix are obtained from sample values in the current image filtered reconstructed sample matrix comprising at least one of:

if the reference block pointed by the block vector is completely positioned outside the current maximum decoding unit, all the element values of the prediction sample matrix are obtained according to the sample values obtained by deblocking filtering and sample offset compensation of the prediction compensated samples of the decoded maximum decoding unit,

if one part of the reference block pointed by the block vector is positioned outside the current maximum decoding unit and the other part is positioned inside the current maximum decoding unit, one part of the element values of the prediction sample matrix is obtained according to the sample values obtained by the deblocking filtering and the sample value offset compensation of the prediction compensated sample of the decoded maximum decoding unit, and the other part is obtained according to the decoded prediction compensated sample value in the current maximum decoding unit, or

If the reference block pointed by the block vector is completely positioned in the current maximum decoding unit, all the element values of the prediction sample matrix are obtained according to the prediction compensation sample values decoded in the current maximum decoding unit.

4. The method of claim 1, wherein the filtered reconstructed sample matrix comprises:

a sample obtained by deblocking filtering the prediction compensated sample decoded in the current maximum decoding unit, and

5. The method of claim 4, wherein the values of the elements of the prediction sample matrix are obtained from sample values in the current image filtered reconstructed sample matrix comprising at least one of:

if one part of the reference block pointed by the block vector is positioned outside the current maximum decoding unit and the other part is positioned inside the current maximum decoding unit, one part of the element values of the prediction sample matrix is obtained according to the sample values obtained by the deblocking filtering and the sample value offset compensation of the prediction compensated sample of the decoded maximum decoding unit, and the other part is obtained according to the sample values obtained by the deblocking filtering of the prediction compensated sample in the current maximum decoding unit, or

And if the reference block pointed by the block vector is completely positioned in the current maximum decoding unit, all the element values of the prediction sample matrix are obtained according to the sample values obtained by deblocking filtering of the decoded prediction compensated samples in the current maximum decoding unit.

6. The method of claim 1, wherein the filtered reconstructed sample matrix comprises:

and the sample obtained by sample value offset compensation of the prediction compensated sample of the decoded maximum decoding unit.

7. The method of claim 6, wherein predicting the values of the elements of the sample matrix is obtained from sample values in a reconstructed sample matrix filtered from the current image comprises at least one of:

if the reference block pointed by the block vector is completely positioned outside the current maximum decoding unit, all the element values of the prediction sample matrix are obtained according to the sample values obtained by sample value offset compensation of the prediction compensated samples of the decoded maximum decoding unit,

if one part of the reference block pointed by the block vector is positioned outside the current maximum decoding unit and the other part is positioned inside the current maximum decoding unit, one part of the element values of the prediction sample matrix is obtained according to the sample values obtained by sample value offset compensation of the prediction compensated samples of the decoded maximum decoding unit, and the other part is obtained according to the decoded prediction compensated sample values in the current maximum decoding unit, or

8. An apparatus for decoding, comprising:

a memory in which a computer program is stored, and

a processor which, when running the computer program, performs the method according to one of claims 1 to 7.

9. A method of encoding, comprising:

determining an identifier indicating whether deblocking filtering is used for the current frame according to whether the current frame to be coded is screen content; and

and writing the identification into a code stream.

10. The method of claim 9, wherein determining the indication indicating whether to use deblocking filtering for the current frame comprises at least one of:

the indication indicates that no deblocking filtering is used for the current frame if the current frame to be encoded is screen content, or

The indication indicates that deblocking filtering is to be used for the current frame if the current frame to be encoded is not screen content.

11. The method of claim 9, further comprising:

and determining whether to carry out deblocking filtering on the prediction compensated samples of the current frame to be coded based on the identification.

12. The method of claim 11, wherein the determining whether to deblock filter the prediction compensated samples of the current frame to be encoded based on the identifying comprises calculating a filter strength of a filter block boundary, the calculating a filter strength of a filter block boundary comprising at least one of:

if the boundary is judged to need filtering, when the absolute value of (p0-q0) is greater than or equal to four times of a first parameter, the filtering strength is 0, wherein p0 is the 1 st sample on the left side or the upper side of the boundary, q0 is the 1 st sample on the right side or the lower side of the boundary, and the first parameter is obtained through calculation according to the quantization parameter and the bit depth;

when the smoothness of the boundary is equal to 6, if the absolute value of (p0-p1) is less than or equal to one fourth of the second parameter and the absolute value of (q0-q1) is less than or equal to one fourth of the second parameter and the absolute value of (p0-p3) is less than or equal to one half of the second parameter and the absolute value of (q0-q3) is less than or equal to one half of the second parameter and the absolute value of (p0-q0) is less than the first parameter, the filter strength is equal to 4, wherein p1 is the 2 nd sample on the left or upper side of the boundary, q1 is the 2 nd sample on the right or lower side of the boundary, p3 is the 4 th sample on the left or upper side of the boundary, q3 is the 4 th sample on the right or lower side of the boundary, the second parameter is obtained by quantization parameter and bit depth calculation; or

When the smoothness of the boundary is equal to 5, if p0 is equal to p1 and q0 is equal to q1 and the absolute value of (p2-q2) is less than the first parameter, the filtering strength is equal to 3, where p2 is the 3 rd sample on the left or upper side of the boundary and q2 is the 3 rd sample on the right or lower side of the boundary.

13. An apparatus of encoding, comprising:

a memory in which a computer program is stored, and

processor which, when running the computer program, performs the method according to one of claims 9 to 12.

14. A method of decoding, comprising:

determining whether to perform deblocking filtering on a prediction compensated sample of a current decoding unit according to whether the current decoding unit is screen content; and

and according to the determined result, performing deblocking filtering on the prediction compensated sample of the current decoding unit to obtain a reconstructed sample after deblocking filtering.

15. The method of claim 14, wherein determining whether to use deblocking filtering on reconstructed sample values of a current decoded block based on whether the current decoded block is screen content comprises at least one of:

determining to use deblocking filtering on reconstructed sample values of the current decoded block if the current decoded block is not screen content, or

If the current decoded block is screen content, then it is determined not to use deblocking filtering on reconstructed sample values of the current decoded block.

16. An apparatus for decoding, comprising:

a memory in which a computer program is stored, and

a processor which, when running the computer program, performs the method of claim 14 or 15.

17. A computer readable storage medium storing non-transitory computer readable instructions which, when executed by a computer, cause the computer to perform the method of one of claims 1-7, 9-12, or 14-15.