JPWO2020187329A5 - - Google Patents

Description

Mutual reference to related applications This application is International Application No. PCT / CN2020 / 080632 filed on March 23, 2020, which is the International Patent Application No. PCT / CN2019 filed on March 21, 2019. It asserts the priority and interests of International Patent Application No. PCT / CN2019 / 100616 filed on / 079148 and August 14, 2019 in a timely manner. The entire disclosure of International Patent Application No. PCT / CN2019 / 079148 and International Patent Application No. PCT / CN2019 / 100616 is incorporated by reference as part of the disclosure of this application.

Technical Field This article relates to video and image coding and decoding techniques.

Digital video occupies the largest bandwidth usage on the Internet and other digital communication networks. Bandwidth demand for the use of digital video is expected to continue to increase as the number of connected user devices capable of receiving and displaying video increases.

The disclosed techniques can be used by embodiments of video or image decoders or encoders in which composite intra-inter-prediction is used.

In one exemplary aspect, a method of processing video is disclosed. This method is a step of determining the coded mode of the first video block; based on the determination of the coding mode of the first video block, constrains one or more flags to a certain operating state. A step in which the operating state is false or true; a step of performing further processing of the second video block depending on the operating state of the one or more flags. The first video block includes a step, which is a nearby video block or a reference video block in relation to the second video block.

In another example, the method of processing the video is the step of determining that pairwise prediction or combined-bi prediction is used for the first video block; paired prediction. Alternatively, it is a step of determining the operating state of a combined inter-intra prediction (CIIP) based on the determination that a compound bi-intra prediction will be used, and the operating state is enabled or disabled (enabled) or disabled (CIIP). Includes a step that is disabled); a step that performs further processing of the first video block according to the operational state of CIIP.

In another example, another method of video processing is disclosed. In this method, the conversion between a video block of a video and the coded representation of the video block is a composite intra and a composite intra where the coded representation corresponds to the intra and inter-prediction weighted averages of the video block. Containing to perform using the inter-prediction mode, the weighted average uses a weight pair from a set of weight pairs that contains less than three pairs.

In another example, another method of video processing is disclosed. In this method, the conversion between a video block of a video and the coded representation of the video block is a composite intra and a composite intra where the coded representation corresponds to the intra and inter-prediction weighted averages of the video block. Including performing using the inter-prediction mode, the weighted average uses a weight pair from a set of weight pairs determined by coding the information of one or more neighboring blocks.

In another example, another method of video processing is disclosed. This method provides one or more coding modes for one or more second blocks during the conversion between the first block in the video data and the bitstream representation of the first block. The step to determine; the step to determine the coding mode constraint of the first block based on the one or more coding modes of the one or more second blocks; the coding of at least the first block. The one or more second blocks are among adjacent blocks, non-adjacent blocks, and reference blocks of the first block, including the step of performing the transformation by applying a mode constraint. Includes at least one.

Another exemplary aspect discloses another method of video processing. This method converts between the current block and the bitstream representation of the current block in the video data by compound inter-intra-prediction (CIIP), diffuse filtering, bilateral filtering, transformation domain filtering, or diffuse filtering, by It involves performing by using at least one of the lateral filtering, and another type of post-reconstruction filtering that is different from the transformation domain filtering, where complex inter-intra-prediction (CIIP), diffusion filtering, The reconstructed neighborhood sample of the current block used in at least one of bilateral filtering, transformation domain filtering, or at least one of the other types of post-reconstruction filtering is the reconstructed neighborhood sample in one or more reference pictures. It is replaced by an approximate sample generated from the corresponding sample of the constructed neighbor sample.

Another exemplary aspect discloses another method of video processing. This method is a step of storing the CIIP mode's composite inter-intra prediction (CIIP) flag and / or intra mode, along with motion information, in a history-based motion vector prediction (HMVP) table. Includes a step of performing a conversion between the current block in the video data and the bitstream representation of the current block, at least based on the HMVP table.

Another exemplary aspect discloses another method of video processing. This method is a step of determining the prediction mode for the current block during the conversion between the current block and the bitstream representation of the current block in the video data; the current block is in Advanced Motion Vector Prediction (AMVP) mode. Or in response to the decision to be coded in merge mode, with the step of determining the applicability of compound inter-intra-predictive (CIIP) mode, which indicates that CIIP mode is valid for the current block; and application of CIIP mode. Includes a step of performing the transformation based on the possibility.

Another exemplary aspect discloses another method of video processing. This method is a step of determining the type of selected merge candidate for the current block during the conversion between the current block in the video data and the bitstream representation of the current block; the selected merge. The current block is coded in merge mode, including the step of determining the applicability of compound inter-intra prediction (CIIP) for the current block depending on the type of candidate.

Another exemplary aspect discloses another method of video processing. This method uses the CIIP flag of one or more neighboring video blocks for the current video block during conversion between the current video block in the video data and the bitstream representation of the current video block. With the steps of coding the composite inter-intra prediction (CIIP) flag for the current video block by coding based on the context model without reference to; at least the composite inter-intra prediction of the current video block ( CIIP) includes the step of performing the conversion by applying the flag.

Another exemplary aspect discloses another method of video processing. This method is a composite inter-intra prediction (CIPP) for the current video block by bypass coding during the conversion between the current video block in the video data and the bitstream representation of the current video block. It involves performing the transformation by coding the flag and applying at least the compound inter-intra prediction (CIIP) flag.

Another exemplary aspect discloses another method of video processing. This method is a step of determining the intra-prediction mode of the first video block of the video according to the rules: the first video during the process of deriving the intra-prediction mode of the first video block. A step that involves skipping checking the compound inter-intra-prediction (CIIP) flag for one or more neighboring video blocks to the block; at least based on the determined intra-prediction mode, the first. Includes steps to perform a conversion between the video block and the bitstream representation of the first video block.

Another exemplary aspect discloses another method of video processing. This method is during the conversion of the video between the current video block coded in compound intra and inter-prediction (CIIP) mode and the bitstream representation of the current video block. A weight pair containing a first weight for the first prediction result of the current video block and a second weight for the second prediction result of the current video block for one or more neighboring video blocks to the current video block. A block-based determination step in which the first prediction result is generated by the intra prediction mode and the second prediction result is generated by the inter prediction mode; with the first prediction result. It includes a step of determining the prediction result of the current block based on the weighted sum of the second prediction result.

In another exemplary aspect, the method described above may be implemented by a video encoder device equipped with a processor.

In yet another exemplary aspect, these methods may be embodied in the form of processor executable instructions and stored on a computer-readable program medium.

These and other aspects are further described in this article.

An example of the derivation process for constructing the merge candidate list is shown.

An example of the position of a spatial merge candidate is shown.

Here is an example of candidate pairs considered for redundancy checking of spatial merge candidates.

Here is an example of the location of the Nx2N and 2NxN partitions for the second PU.

An example of motion vector scaling for temporal merge candidates is shown.

An example of candidate positions for temporal merge candidates C0 and C1 is shown.

An example of a compound bi-predictive merge candidate is shown.

An example of the derivation process for motion vector prediction candidates is shown.

An example of motion vector scaling for spatial motion vector candidates is shown.

An example of alternative temporal motion vector prediction (ATMVP) for a coding unit (CU) is shown.

An example of a neighborhood sample used to derive IC parameters is shown.

An example of the ultimate motion vector expression (UMVE) search process is shown.

Here is an example of a UMVE search point.

An example of the proposed decoding flowchart by the HMVP method is shown.

An example of updating a table in the proposed HMVP method is shown.

An example of symmetric mode is shown.

Here are some examples of nearby samples of the current block and their corresponding samples in the reference picture.

It is a block diagram of an example of a video processing apparatus.

A block diagram of an exemplary implementation of a video encoder is shown.

It is a flowchart about an example of a video processing method.

It is a flowchart about an example of a video processing method.

Here is an example of the upper and left neighborhood blocks used in CIIP weight evaluation.

It is a flowchart about an example of a video processing method.

It is a flowchart about an example of a video processing method.

It is a flowchart about an example of a video processing method.

It is a flowchart about an example of a video processing method.

It is a flowchart about an example of a video processing method.

It is a flowchart about an example of a video processing method.

It is a flowchart about an example of a video processing method.

It is a flowchart about an example of a video processing method.

It is a flowchart about an example of a video processing method.

This article provides a variety of techniques that can be used by image or video bitstream decoders to improve the quality of decompressed or decoded digital video or images. For brevity, the term "video" is used herein to include both a series of pictures (traditionally referred to as video) and individual images. In addition, video encoders can also implement these techniques during the encoding process to reconstruct the decoded frames used for further encoding.

Section headings are used herein for ease of understanding and are not limited to the corresponding sections of embodiments and techniques. Thus, embodiments from one section can be combined with embodiments from other sections.

1. 1. Overview

This article is about video coding technology. Specifically, this article relates to compound intra-inter-prediction (CIIP) in video coding or decoding. This may apply to existing video coding standards such as HEVC, or to standards that will be finalized (Versatile Video Coding ). This may also be applicable to future video coding standards or video codecs.

2. 2. background

Video coding standards have evolved primarily through the development of well-known ITU-T and ISO / IEC standards. ITU-T produces H.261 and H.263, ISO / IEC produces MPEG-1 and MPEG-4 Visual, and both agencies jointly produce H.262 / MPEG-2 Video and H.264 / MPEG- 4 Created Advanced Video Coding (AVC) and H.265 / HEVC [1] standards. Since H.262, the video coding standard has been based on a hybrid video coding structure that utilizes temporal prediction and transformation coding . In 2015, VCEG and MPEG jointly established the Joint Video Exploration Team (JVET) to explore future video coding technologies beyond HEVC. Since then, many new methods have been adopted by JVET and put into reference software named JEM (Joint Exploration Model). In April 2018, a joint video expert team (JVET) was established between VCEG (Q6 / 16) and ISO / IEC JTC1 SC29 / WG11 (MPEG), which is 50% bit compared to HEVC. We are working on a VVC standard that aims to reduce rates.

FIG. 19 is a block diagram of an exemplary implementation of a video encoder. Figure 19 incorporates a feedback path in which the encoder implementation also performs the video encoder function (reconstructing a compressed representation of the video data for use in the next video data encoding). It shows that there is.

2.1 Inter-prediction in HEVC / H.265

Each inter-predicted PU has motion parameters for one or two reference picture lists. Motion parameters include motion vectors and reference picture indexes. The use of one of the two reference picture lists may be signaled using inter_pred_idc. The motion vector may be explicitly coded as a delta for the predictor.

If a CU is coded in skip mode, one PU is associated with that CU, there is no significant residual factor, and there is no delta or reference picture index for the coded motion vector. A merge mode is specified, which retrieves motion parameters for the current PU from nearby PUs, including spatial and temporal candidates. The merge mode can be applied to any inter-predictive PU, not just the skip mode. An alternative to merge mode is the explicit transmission of motion parameters, where the motion vector (more precisely, the motion vector difference compared to the motion vector predictor), the corresponding reference for each reference picture list. Picture index and reference picture list usage are explicitly signaled for each PU. Such a mode is referred to in the present disclosure as Advanced motion vector prediction (AMVP).

If the signal transduction indicates that one of the two reference picture lists should be used, the PU is generated from one block of sample. This is called "uni-prediction". Simple predictions are possible for both P and B slices.

If the signal transduction indicates that both reference picture lists should be used, the PU is generated from two blocks of the sample. This is called "bi-prediction". Bi-prediction is possible only for B slices.

The following text gives details of the inter-prediction mode specified in HEVC. The explanation starts in merge mode.

2.1.1 Merge mode

2.1.1.1 Derivation of candidates for merge mode

When the PU is predicted using merge mode, the index pointing to the entry in the merge candidate list is parsed from the bitstream and used to get the motion information. The construction of this list is specified in the HEVC standard and can be summarized by following a series of steps below.

Step 1: Derivation of initial candidates

Step 1.1: Derivation of spatial candidates

Step 1.2: Redundancy check for spatial candidates

Step 1.3: Derivation of temporal candidates

Step 2: Insert additional candidates

Step 2.1: Generate dual prediction candidates

Step 2.2: Insert Zero Movement Candidate

These steps are also outlined in Figure 1. Up to four merge candidates are selected from five differently located candidates for the derivation of spatial merge candidates. For deriving temporal merge candidates, up to one merge candidate is selected from the two candidates. The decoder assumes that the number of candidates is constant for each PU, so if the number of candidates obtained in step 1 does not reach the maximum number of merge candidates signaled in the slice header (MaxNumMergeCand). Additional candidates are generated. Since the number of candidates is constant, the index of the best merge candidates is encoded using truncated unary binarization (TU). If the CU size is equal to 8, all PUs in the current CU share the same single merge candidate list as the 2Nx2N predictive unit merge candidate list.

In the following, the operations related to the above steps will be described in detail.

2.1.1.2 Derivation of spatial candidates

In deriving the spatial merge candidates, up to four merge candidates are selected from the candidates located at the positions shown in FIG. The order of derivation is A1, B1, B0, A0, B2. Position B2 is only considered if none of the PUs at positions A1, B1, B0, A0 are available (eg, because it belongs to another slice or tile) or if it is intracoded . After the candidate at position A1 is added, the addition of the remaining candidates is subject to a redundancy check that ensures that candidates with the same motion information are excluded from the list, improving coding efficiency. To reduce computational complexity, the redundancy checks described above do not consider all possible candidate pairs. Instead, only the pairs connected by the arrows in Figure 3 are considered, and the candidates are added to the list only if the corresponding candidates used for the redundancy check do not have the same motion information. Another source of duplicate motion information is the "second PU" associated with a partition different from 2Nx2N. As an example, FIG. 4 shows a second PU for N × 2N and 2N × N respectively. If the current PU is split as Nx2N, the candidates at position A1 are not considered for list construction. In fact, adding this candidate gives us two predictive units with the same motion information. This is redundant with having only one PU per coding unit. Similarly, if the current PU is split as 2N × N, position B1 is not considered.

2.1.1.3 Derivation of temporal candidates

At this step, only one candidate is added to the list. Specifically, in deriving this temporal merge candidate, a scaled motion vector is based on the co-positioned PU belonging to the picture that has the smallest POC difference from the current picture in a given reference picture list. Derived. The reference picture list used to derive the co-positioned PU is explicitly signaled within the slice header. A scaled motion vector for temporal merge candidates is obtained as shown by the dotted line in Figure 5, which is from the co-positioned PU motion vector using the POC distance, tb and td. Scaled. Here, tb is defined as the POC difference between the reference picture of the current picture and the current picture, and td is the POC difference between the reference picture of the co-positioned picture and the co-positioned picture. Defined. The reference picture index for temporal merge candidates is set to zero. The practical implementation of the scaling process is described in the HEVC specification. For B slices, two motion vectors, one for reference picture list 0 and the other for reference picture list 1, are obtained and combined to create bipredictive merge candidates.

In the co-positioned PU (Y) belonging to the reference frame, a position for the temporal candidate is selected between candidate C0 and candidate C1 as shown in FIG. Position C1 is used if the PU at position C0 is not available, is intracoded , or deviates from the current CTU line. Otherwise, position C0 is used to derive the temporal merge candidate.

2.1.1.4 Inserting additional candidates

In addition to spatial and temporal merge candidates, there are two additional types of merge candidates: compound bi-predictive merge candidates and zero merge candidates. Complex bi-predictive merge candidates are generated by utilizing spatial and temporal merge candidates. The compound bi-predictive merge candidate is used only for B slices. The compound bi-prediction candidate is generated by combining the first reference picture list motion parameter of the initial candidate with the second reference picture list motion parameter of another candidate. If these two tuples provide different motion hypotheses, they form a new bi-prediction candidate. As an example, Figure 7 uses two candidates in the original list with mvL0 and refIdxL0 or mvL1 and refIdxL1 (on the left) to create a compound bi-predictive merge candidate that will be added to the final list (on the right). Shows the case. There are a number of possible combinations for generating these additional merge candidates.

Zero motion candidates are inserted to fill the remaining entries in the merge candidate list and thus reach the MaxNumMergeCand capacity. These candidates have a zero spatial displacement and a reference picture index that starts at zero and increases with each new zero motion candidate added to the list. The number of reference frames used by these candidates is 1 and 2 for unidirectional and bidirectional predictions, respectively. Finally, no redundancy check is performed on these candidates.

2.1.1.5 Motion estimation area for parallel processing

To speed up the encoding process, motion estimation can be performed in parallel, which simultaneously derives motion vectors for all predictive units in a given region. Derivation of merge candidates from spatial neighborhoods can hinder parallel processing. One prediction unit can derive motion parameters from adjacent PUs only after its associated motion estimation is complete. To ease the trade-off between coding efficiency and processing latency, HEVC uses the "log2_parallel_merge_level_minus2" syntax element in the picture parameter set to signal the size of the motion estimation region (MER). ) Is defined. If MER is defined, merge candidates in the same area are marked as unavailable and are therefore not considered in list construction.

2.1.2 AMVP

AMVP utilizes the spatiotemporal correlation of motion vectors with nearby PUs, which is used for explicit transmission of motion parameters. For each reference picture list, first check the availability of temporally nearby PU positions on the left and above, remove redundant candidates, and add zero vectors to make the candidate list constant length. , The motion vector candidate list is constructed. The encoder can then select the best predictor from the candidate list and send the corresponding index indicating the selected candidate. Similarly, in merge index signaling, the index of the best motion vector candidate is encoded using the censored unilinear system. The maximum value to be encoded in this case is 2 (see Figure 8). The following sections show the details of the process for deriving motion vector prediction candidates.

2.1.2.1 Derivation of AMVP candidates

Figure 8 summarizes the derivation process for motion vector prediction candidates.

Motion vector prediction considers two types of motion vector candidates: spatial motion vector candidates and temporal motion vector candidates.

For the derivation of spatial motion vector candidates, two motion vector candidates are finally derived based on the motion vectors of each PU located at five different positions as shown in FIG. For the derivation of temporal motion vector candidates, one motion vector candidate is selected from two candidates derived based on the positions of two different co-positions. After the first list of spatial and temporal candidates is created, duplicate motion vector candidates in the list are removed. If the number of potential candidates is greater than 2, motion vector candidates with a reference picture index greater than 1 in the associated reference picture list are removed from the list. If the number of spatial-temporal motion vector candidates is less than 2, additional zero motion vector candidates are added to the list.

2.1.2.2 Spatial motion vector candidates

In deriving spatial motion vector candidates, up to two of the five potential candidates are considered. The five potential candidates are derived from PUs located at positions as shown in Figure 2, and their positions are the same as for motion merging. The order of derivation for the left side of the current PU is defined as A0, A1, and scaled A0, scaled A1. The order of derivation for the upper side of the current PU is defined as B0, B1, B2, scaled B0, scaled B1, scaled B2. Therefore, for each side, there are four cases that can be used as motion vector candidates, two cases do not need to use spatial scaling, and two cases use spatial scaling. The four different cases can be summarized as follows:

No spatial scaling

(1) Same reference picture list, same reference picture index (same POC)

(2) Different reference picture lists, but the same reference picture (same POC)

Spatial scaling

(3) Same reference picture list, but different reference pictures (different POC)

(4) Different reference picture list and different reference pictures (different POC)

Check first if there is no spatial scaling, then deal with spatial scaling. Spatial scaling is taken into account when the POC differs between the reference picture of the nearby PU and the reference picture of the current PU, regardless of the reference picture list. If all PUs in the left candidate are not available or are intracoded , scaling for the motion vector above is allowed to help parallel derivation of the left and top MV candidates. Otherwise, spatial scaling is not allowed for the motion vector above.

In the spatial scaling process, the motion vectors of nearby PUs are scaled in a manner similar to temporal scaling, as shown in Figure 9. The main difference is that the current PU reference picture list and index are given as input, and the actual scaling process is the same as for temporal scaling.

2.1.2.3 Temporal motion vector candidates

Apart from deriving the reference picture index, all processes for deriving temporal merge candidates are the same as for deriving spatial motion vector candidates (see Figure 6). The reference picture index is signaled to the decoder.

2.2 New inter-prediction method

2.2.1 Sub-CU-based motion vector prediction

In JEM with QTBT, each CU can have up to one set of motion parameters for each prediction direction. By dividing the large CU into sub-CUs and deriving motion information for all sub-CUs of the large CU, the encoder considers two sub-CU level motion vector prediction methods. The Alternative temporal motion vector prediction (ATMVP) method allows each CU to bring in multiple sets of motion information from multiple blocks smaller than the current CU in the co-located reference picture. .. In the spatial-temporal motion vector prediction (STMVP) method, the motion vector of the sub-CU is recursively derived using the temporal motion vector predictor and the motion vector in the spatial vicinity.

Motion compression for reference frames is currently disabled to maintain a more accurate motion field for subCU motion prediction.

2.2.1.1 Alternative Time Motion Vector Prediction

The alternative temporal motion vector prediction (ATMVP) method provides multiple sets of motion information (including motion vectors and reference indexes) from blocks smaller than the current CU, with motion vector temporal motion vector prediction (TMVP). Correct by fetching. As shown in Figure 10, the subCU is a square N × N block (N is set to 4 by default).

ATMVP predicts the motion vector of the sub-CU in the CU in two steps. The first step is to use so-called temporal vectors to identify the corresponding blocks in the reference picture. This reference picture is called a motion source picture. The second step is to divide the current CU into sub-CUs and get the motion vector and reference index of each sub-CU from the block corresponding to each sub-CU, as shown in FIG.

In the first step, the reference picture and the corresponding block are determined by the motion information of the spatially neighboring blocks of the current CU. The first merge candidate in the current CU merge candidate list is used to avoid the iterative scanning process of neighboring blocks. The first available motion vector and its associated reference index are set to the temporal vector and index to the motion source picture. Thus, in ATMVP, the corresponding block can be identified more accurately compared to TMVP. Here, the corresponding block (sometimes called the co-positioned block) is always in the lower right or center position with respect to the current CU.

In the second step, the corresponding blocks of the sub-CU are identified by a temporal vector in the motion source picture. This is due to adding a time vector to the coordinates of the current CU. For each sub-CU, the motion information of the corresponding block (the smallest motion grid covering the central sample) is used to derive the motion information of the sub-CU. After the motion information of the corresponding N × N block is identified, it is converted to the motion vector and reference index of the current subCU, similar to the TMVP of HEVC. Here, motion scaling and other procedures apply. For example, the decoder checks if a low latency condition (ie, the POC of all reference pictures in the current picture is less than the POC of the current picture) is met, and possibly the motion for each subCU. Use the motion vector MVx (the motion vector corresponding to the reference picture list X) to predict the vector MVy (X equals 0 or 1 and Y equals 1–X).

2.2.2 Average candidate for each pair

A pairwise average candidate is generated by averaging the predefined pairs of candidates in the current list of merge candidates. The predefined pairs are defined as {(0,1), (0,2), (1,2), (0,3), (1,3), (2,3)}. Here, the numbers indicate the merge index to the merge candidate list. The average motion vector is calculated separately for each reference list. If both motion vectors are available in one list, these two motion vectors will be averaged even if they point to different reference pictures. If only one motion vector is available, use that one motion vector directly and leave this list disabled if no motion vector is available. The average candidate for each pair replaces the combined candidate in the HEVC standard.

Table 1 summarizes the complexity analysis of the mean candidates for each pair. For the worst case of additional calculations for the mean (last column in Table 1), 4 additions and 4 shifts are required for each pair (MVx and MVy in L0 and L1), for each pair. Four reference index comparisons are required (refIdx0 is valid for L0 and L1 and refIdx1 is valid). There are 6 pairs, for a total of 24 additions, 24 shifts, and 24 reference index comparisons. Combined candidates in the HEVC standard use two reference index comparisons for each pair (refIdx0 is valid for L0, refIdx1 is valid for L1), and there are 12 pairs, for a total of 24 reference index comparisons. Become.

Table 1: Behavior analysis of average candidates for each pair

2.2.3 Local lighting compensation

Local Illumination Compensation (LIC) is based on a linear model of lighting change with scaling factors a and offset b. It is also adaptively enabled or disabled for each intermode coded coding unit (CU).

When the LIC is applied to the CU, the least squares method is used to derive the parameters a and b by using the samples in the vicinity of the current CU and their corresponding reference samples. More specifically, as shown in FIG. 11, a sample in the vicinity of the CU subsampled (2: 1 subsampling) and the corresponding sample in the reference picture (depending on the current CU or subCU motion information). Identified) is used. IC parameters are derived and applied separately for each prediction direction.

If the CU is coded in merge mode, the LIC flag is copied from the neighboring blocks in the same way as motion information copy in merge mode; otherwise, to indicate whether LIC is applied. , The LIC flag is signaled for that CU.

If LIC is enabled for a picture, an additional CU level RD check is needed to determine if LIC applies to the CU. If LIC is enabled for the CU, instead of SAD and SATD, mean-removed sum of absolute difference (MR-SAD) and mean-removed sum of mean-removed sum of absolute Hadamard-transformed difference (MR-SATD)) is used for integer pixel motion search and fractional pixel motion search, respectively.

To reduce encoding complexity, the following encoding schemes are applied in JEM.

LIC is currently disabled for the entire picture if there is no apparent lighting change between the picture and its reference picture. To identify this situation, the histogram of the current picture and all reference pictures of the current picture are calculated in the encoder. If the histogram difference between the current picture and all reference pictures of the current picture is less than a given threshold, LIC is disabled for the current picture; otherwise, LIC is enabled for the current picture.

2.2.4 Composite Intra and Inter Prediction

In some embodiments, multi-hypothesis prediction has been proposed, where compound intra-hypothesis and inter-prediction are one way to generate multiple hypotheses.

When multiple hypothesis predictions are applied to improve intramode, multiple hypothesis predictions combine one intra prediction with one merge indexed prediction. The merge CU signals one flag for the merge mode. This is to select the intra mode from the intra candidate list when the flag is true. For the luma component, the intra-candidate list is derived from four intra-prediction modes, including DC, planar, horizontal, and vertical modes, and the size of the intra-candidate list can be 3 or 4 depending on the shape of the block. If the CU width is greater than twice the CU height, horizontal mode is excluded from the intramode list, and if the CU height is greater than twice the CU height, vertical mode is removed from the intramode list. One intra-prediction mode selected by the intra-mode index and one merge indexed prediction selected by the merge index are combined using a weighted average. For chroma components, DM is always applied without extra signaling. The weights for combining the predictions are described as follows. If DC or planar mode is selected, or if the CB width or height is less than 4, equal weights apply. For CBs with a width and height of 4 or more, if horizontal / vertical mode is selected, one CB is first divided vertically / horizontally into four equal area areas. Each weight set shown as (w_intrai, w_interi) applies to the corresponding region. Here, i is from 1 to 4, (w_intra1, w_inter1) = (6,2), (w_intra2, w_inter2) = (5,3), (w_intra3, w_inter3) = (3,5), (w_intra4). , w_inter4) = (2,6). (w_intra1, w_inter1) is for the region closest to the reference sample, and (w_intra4, w_inter4) is for the region farthest from the reference sample. The combined predictions can then be calculated by adding the two weighted predictions together and shifting them 3 bits to the right. In addition, the intra-prediction mode for the predictor's intra-hypothesis can be saved for subsequent references to neighboring CUs.

Generalized bi-prediction

In conventional bi-prediction, the predictors from L0 and L1 are averaged to produce the final predictor with an equal weight of 0.5. The predictor generation formula is expressed as Eq. (3).

P _{TraditionalBiPred} = (P _L0 + P _L1 + RoundingOffset) >> shiftNum (1)

In equation (3), P _{TraditionalBiPred} is the final predictor for traditional bipredictors, P _L0 and P _L1 are predictors from L0 and L1, respectively, and Rounding Offset and shiftNum normalize the final predictors. Used for.

A generalized bi-prediction (GBI) is proposed that allows different weights to be applied to the predictors from L0 and L1. The predictor generation is shown in Eq. (4).

P _GBi = ((1-w ₁ ) * P _L0 + w ₁ * P _L1 + RoundingOffset _GBi ) >> shiftNum _GBi (2)

In equation (4), P _GBi is the final predictor of GBi, and (1-w ₁ ) and w ₁ are the selected GBI weights applied to the predictors of L0 and L1, respectively. RoundingOffsetGBi and shiftNumGBi are used to normalize the final predictors of GBi.

The supported w1 weight tables are {-1 / 4,3 / 8,1 / 2,5 / 8,5 / 4}. One equal weight set and four unequal weight sets are supported. In the case of equal weights, the process of generating the final predictor is exactly the same as in the traditional bi-predictor mode. For true bi-prediction in random access (RA) conditions, the number of candidate weight sets is reduced to 3.

For Advanced Motion Vector Prediction (AMVP) modes, if this CU is coded by bi-prediction, the weight selection in GBI is explicitly signaled at the CU level. For merge modes, the weight selection is inherited from the merge candidates. In this proposal, GBI supports DMVR to generate weighted averages for templates as well as final predictors for BMS-1.0.

2.2.6 Extreme motion vector expression

In some embodiments, an ultimate motion vector expression (UMVE) is presented. UMVE is used in either skip or merge modes using the proposed motion vector expression method.

UMVE reuses merge candidates as it does with VVC. Among the merge candidates, candidates can be selected and further extended by the proposed motion vector expression method.

UMVE provides a new motion vector expression with simplified signal transduction. This expression method includes a starting point, the magnitude of movement, and the direction of movement.

This proposed technique uses the merge candidate list as is. However, only candidates with the default merge type (MRG_TYPE_DEFAULT_N) are considered for UMVE extensions.

The base candidate index defines the starting point. The base candidate index indicates the best candidate in the list as follows.

Table 1: Base candidate IDX

If the number of base candidates is equal to 1, the base candidate IDX is not signaled.

The distance index is information on the magnitude of movement. The distance index indicates a predefined distance from the starting point information. The predefined distances are:

Table 2: Distance IDX

The directional index represents the direction of the MVD with respect to the starting point. The directional index can represent the four directions shown below.

Table 3: Direction IDX

The UMVE flag is signaled immediately after sending the skip and merge flags. If the skip and merge flags are true, the UMVE flags are parsed. If the UMVE flag is equal to 1, the UMVE syntax is parsed. However, if it is not 1, the AFFINE flag is parsed. If the AFFINE flag is equal to 1, it is in AFFINE mode, but if it is not 1, the skip / merge index is parsed for VTM skip / merge mode.

No additional line buffers due to UMVE candidates are needed. This is because software skip / merge candidates are used directly as base candidates. The input UMVE index is used to determine the supplement just before motion compensation. There is no need to hold a long line buffer for this.

2.2.7 History-based motion vector prediction

In our previous invention P1805028401H, there is one or more look-up tables containing at least one motion candidate for predicting block motion information.

A history-based MVP (HMVP) method is proposed in which HMVP candidates are defined as previously coded block motion information. A table with multiple HMVP candidates is maintained during the encode / decode process. When encountering a new slice, the table is emptied. Whenever there is an intercoding block, the relevant motion information is added to the last entry in the table as a new HMVP candidate. The overall coding flow is shown in Figure 13.

In one example, the table size is set to L (eg L = 16 or 6, or 44), which indicates that L HMVP candidates may be added to the table.

In one embodiment, if there are more than L HMVP candidates from the previously coded block, first-in-first so that the table always contains the latest previously coded L motion candidates. -Out, FIFO) rules apply. Figure 14 shows an example where FIFO rules are applied to remove HMVP candidates and add new ones to the table used in the proposed method.

In another embodiment, whenever a new motion candidate is added (eg, the block is currently interconnected and in non-affine mode), to identify if there are identical or similar motion candidates in the LUTs. First, the redundancy check process is applied.

2.2.8 Symmetric motion vector difference

In some embodiments, symmetric motion vector difference (SMVD) is proposed in order to encode the MVD more efficiently.

First, at the slice level, the variables BiDirPredFlag, RefIdxSymL0, RefIdxSymL1 are derived as follows.

The forward reference picture in reference picture list 0 that is closest to the current picture is searched. If found, RefIdxSymL0 is set equal to the reference index of the forward picture.

The backward reference picture in reference picture list 1 that is closest to the current picture is searched. If found, RefIdxSymL1 is set equal to the reference index of the back picture.

BiDirPredFlag is set to 1 if both front and back pictures are found.

Otherwise, the following applies:

The back-referenced picture in reference picture list 0 that is closest to the current one is searched. If found, RefIdxSymL0 is set equal to the reference index of the back picture.

The forward reference picture in Reference Picture List 1 that is closest to the current one is searched. If found, RefIdxSymL1 is set equal to the reference index of the forward picture.

BiDirPredFlag is set to 1 if both back and front pictures are found. Otherwise, BiDirPredFlag is set equal to 0.

Second, at the CU level, if the prediction direction for the CU is bi-prediction and the BiDirPredFlag is equal to 1, a symmetric mode flag indicating whether symmetric mode is used is explicitly signaled.

If the flag is true, only mvp_l0_flag, mvp_l1_flag, and MVD0 are explicitly signaled. The reference index is set equal to RefIdxSymL0 and RefIdxSymL1 for List 0 and List 1, respectively. MVD1 is set equal to -MVD0. The final motion vector is shown in the following equation.

Table 2 shows the modifications in the coding unit syntax.

Table 2: Modifications in coding unit syntax

2.2.9 CIIP in VTM4
In VTM4, if the CU is coded in merge mode and the CU contains at least 64 luma samples (ie, CU width x CU height is 64 or greater), then compound inter / intra prediction (CIIP) to the current CU. An additional flag is signaled to indicate whether the mode is applied.

To form the CIIP prediction, the intra prediction mode is first derived from two additional syntax elements. Up to four possible intra-prediction modes are available: DC, planar, horizontal, or vertical. The inter-prediction and intra-prediction signals are then derived using normal intra and inter-decode processes. Finally, a weighted average of the inter and intra prediction signals is performed to obtain a CIIP prediction.

1. Derivation of intra prediction mode
In CIIP mode, up to four intra-prediction modes, including DC, planar, horizontal, and vertical modes, can be used to predict luma components. Horizontal mode is not acceptable if the CU shape is very wide (ie, if the width is more than twice the height). Vertical mode is not acceptable if the shape of the CU is very narrow (ie, the height is more than twice the width). In these cases, only three intra prediction modes are allowed.

CIIP mode uses three most probable modes (MPM) for intra-prediction. The CIIP MPM candidate list is formed as follows:
-The left neighborhood block and the top neighborhood block are set as A and B, respectively.
-The intra prediction modes of block A and block B are labeled intraModeA and intraModeB, respectively, and are derived as follows:
○ Let X be either A or B ○ intraModeX is 1) if block X is not available, or 2) if block X is not predicted using CIIP mode or intramode, 3) block B is current. If it is outside the CTU, it is set to DC. ○ Otherwise, intraModeX is 1) DC or plane if the intra prediction mode of block X is DC or plane; or 2) Intra prediction mode of block X. Vertical if in "vertical" angle mode (greater than 34), or 3) set horizontally if block X's intra prediction mode is "horizontal" angle mode (34 or less)-intraModeA and intraModeB are the same case:
○ If intraModeA is planar or DC, the three MPMs are set to {planar, DC, vertical} (in that order).
○ If not, the three MPMs are set to {intraModeA, plane, DC} (in that order)
-Otherwise (intraModeA and intraModeB are different):
○ The first two MPMs are set to {intraModeA, intraModeB} (in this order) ○ Plane, DC, vertical uniqueness is checked for the first two MPM candidate modes in this order And as soon as a unique mode is found, it will be added as a third MPM.

If the shape of the CU is very wide or very narrow as defined above, the MPM flag is estimated to be 1 without signal transduction. Otherwise, the MPM flag is signaled to indicate whether the CIIP intra prediction mode is one of the CIIP MPM candidate modes.

If the MPM flag is 1, the MPM index is further signaled to indicate which MPM candidate mode is used in the CIIP intra prediction. Otherwise, if the MPM flag is 0, the intra prediction mode is set to "missing" mode in the MPM candidate list. For example, if the plane mode is not in the MPM candidate list, the plane will be in missing mode and the intra prediction mode will be set to plane. Since CIIP allows four possible intra-prediction modes and the MPM candidate list contains only three intra-prediction modes, one of the four possible modes must be the missing mode.

For chroma components, DM mode is always applied without additional signal transduction. That is, the chroma uses the same prediction mode as the luma.

The intra-prediction mode of the CIIP coding CU is preserved and used in future intra-mode coding of nearby CUs.

2.2.9.2 Combination of inter and intra prediction signals

The inter prediction signal P _inter in CIIP mode is derived using the same inter prediction process applied in normal merge mode, and the intra prediction signal P _intra is derived using CIIP intra prediction mode according to the normal intra prediction process. Will be done. The intra prediction signal and the inter prediction signal are then combined using a weighted average. Here, the weight value depends on the intra prediction mode, and the sample is located in the coding block. It looks like this:
-If the intra prediction mode is DC or planar mode, or if the block width or height is less than 4, weights equal to the intra prediction and inter prediction signals are applied.
-Otherwise, the weights are determined based on the intra prediction mode (in this case horizontal or vertical mode) and the sample position within the block. Take the horizontal prediction mode as an example (weights for the vertical mode are similarly derived, but in the orthogonal direction). Let W be the width of the block and H be the height of the block. The coding block is first divided into four equal area portions having dimensions (W / 4) × H, respectively. The weights wt for each of the four regions starting from the part closest to the intra-predictive reference sample and ending at the part farthest from the intra-predictive reference sample are set to 6, 5, 3, and 2, respectively. The final CIIP prediction signal is derived using the following equation:

CIIP in VTM5
In VTM5, when the CU is coded in merge mode, the CU contains at least 64 luma samples (ie, CU width x CU height is 64 or more), and both CU width and CU height are 128 luma. • If less than a sample, an additional flag is signaled to indicate whether combined inter / intra prediction (CIIP) mode is applied to the current CU. As the name implies, CIIP prediction combines an intra-prediction signal with an inter-prediction signal. The inter prediction signal P _inter in CIIP mode is derived using the same inter prediction process applied in normal merge mode, and the intra prediction signal P _intra follows the normal intra prediction process using plane mode. Derived. The intra-prediction signal and the inter-prediction signal are then combined using a weighted average. Here, the weight value is calculated depending on the coding mode of the upper neighborhood block and the left neighborhood block (shown in Figure 16) as follows:
-Set isIntraTop to 1 if the above neighborhood is available and intracoded, otherwise set isIntraTop to 0;
-Set isIntraLeft to 1 if the left neighborhood is available and intracoded, otherwise set isIntraLeft to 0;
-If (isIntraLeft + isIntraTop) is equal to 2, wt is set to 3;
-Otherwise, if (isIntraLeft + isIntraTop) is equal to 1, wt is set to 2;
-Otherwise, set wt to 1.

The CIIP prediction is formed as follows:

FIG. 16 shows the top and left neighborhood blocks used in CIIP weight derivation.

2.3 Intramode derivation process

In the current VTM-4.0, when coding one intracoding block, one most probable mode (MPM) flag is signaled first. If the MPM flag is true, further indexes to the MPM list will be signaled. Otherwise, the non-MPM mode index will be signaled.

The relevant syntax elements, symmetry, and decoding process in the latest VVC specification (JVET-M1001_v7) are given as follows:

Related syntax

Coding unit syntax

Meaning content

The syntax elements intra_luma_mpm_flag [x0] [y0], intra_luma_mpm_idx [x0] [y0], intra_luma_mpm_resedder [x0] [y0] specify the intra prediction mode for the luma sample. The array index x0, y0 specifies the position (x0, y0) of the upper left luma sample of the considered coding block with respect to the upper left luma sample of the picture. If intra_luma_mpm_flag [x0] [y0] is equal to 1, the intra prediction mode is estimated from the neighboring intra-predicted coding units according to Section 8.4.2.

If intra_luma_mpm_flag [x0] [y0] does not exist, it is estimated to be equal to 1.

Decoding process

注記：イントラ予測モードINTRA_LT_CCLM、INTRA_L_CCLM、INTRA_T_CCLMはクロマ成分にのみ適用できる。
IntraPredModeY[xCb][yCb]は、以下の順序付けられたステップによって導出される：
１．近傍位置(xNbA,yNbA)および(xNbB,yNbB)は、それぞれ(xCb－1,yCb＋cbHeight－1)および(xCb＋cbWidth－1,yCb－1)に等しく設定される。
２．AまたはBのいずれかで置き換えられるXについて、変数candIntraPredModeXは次のように導出される：
－6.4.X節に規定されるブロックについての利用可能性導出プロセス［編注（BB）：近傍ブロック利用可能性検査プロセスtbd］は、(xCb,yCb)に等しいと設定された位置(xCurr,yCurr)と、(xNbX,yNbX)に等しいと設定された近傍位置(xNbY,yNbY)とを入力として呼び出され、出力はavailableXに割り当てられる。
－候補イントラ予測モードcandIntraPredModeXは以下のように導出される：
－下記の条件の一つまたは複数が真の場合、candIntraPredModeXはINTRA_PLANARに等しく設定される。
－変数avalableXはFALSEに等しい。
－CuPredMode[xNbX][yNbX]はMODE_INTRAと等しくなく、ciip_flag[xNbX][yNbX]は1と等しくない。
－pcm_flag[xNbX][yNbX]は1に等しい。
－XはBに等しく、yCb－1は((yCb>>CtbLog2SizeY)<<CtbLog2SizeY)より小さい。
－それ以外の場合、candIntraPredModeXはIntraPredModeY[xNbX][yNbX]に等しく設定される。
３．変数ispDefaultMode1およびispDefaultMode2は次のように定義される：
－IntraSubPartitionsSplitTypeがISP_HOR_SPLITに等しい場合、ispDefaultMode1はINTRA_ANGULAR18に等しく設定され、ispDefaultMode2はINTRA_ANGULAR5に等しく設定される。
－そうでない場合、ispDefaultMode1はINTRA_ANGULAR50に等しく設定され、ispDefaultMode2はINTRA_ANGULAR63に等しく設定される。
４．x＝0…5として、candModeList[x]は次のように導出される：
－candIntraPredModeBがcandIntraPredModeAに等しく、candIntraPredModeAがINTRA_DCより大きい場合、x＝0…5として、candModeList[x]は次のように導出される：
－IntraLumaRefLineIdx[xCb][yCb]が0に等しく、IntraSubPartitionsSplitTypeがISP_NO_SPLITに等しい場合、下記が適用される：
candModeList[0]＝candIntraPredModeA (8-9)
candModeList[1]＝INTRA_PLANAR (8-10)
candModeList[2]＝INTRA_DC (8-11)
candModeList[3]＝2＋((candIntraPredModeA＋61)%64) (8-12)
candModeList[4]＝2＋((candIntraPredModeA－1)%64) (8-13)
candModeList[5]＝2＋((candIntraPredModeA＋60)%64) (8-14)
－それ以外の場合（IntraLumaRefLineIdx[xCb][yCb]が0に等しくないか、IntraSubPartitionsSplitTypeがISP_NO_SPLITに等しくない場合）、下記が適用される：
candModeList[0]＝candIntraPredModeA (8-15)
candModeList[1]＝2＋((candIntraPredModeA＋61)%64) (8-16)
candModeList[2]＝2＋((candIntraPredModeA－1)%64) (8-17)
－下記の条件：
－IntraSubpartitionsSplitTypeがISP_HOR_SPLITに等しく、candIntraPredModeAがINTRA_ANGULAR34より小さい、
－IntraSubpartitionsSplitTypeがISP_VER_SPLITに等しく、candIntraPredModeAがINTRA_ANGULAR34以上である、
－IntraLumaRefLineIdx[xCb][yCb]が0に等しくない
のうちの一つが真である場合、下記が適用される：
candModeList[3]＝2＋((candIntraPredModeA＋60)%64) (8-18)
candModeList[4]＝2＋(candIntraPredModeA%64) (8-19)
candModeList[5]＝2＋((candIntraPredModeA＋59)%64) (8-20)
－それ以外の場合は、下記が適用される：
candModeList[3]＝ispDefaultMode1 (8-21)
candModeList[4]＝ispDefaultMode2 (8-22)
candModeList[5]＝INTRA_PLANAR (8-23)
－そうではなく、candIntraPredModeBがcandIntraPredModeAに等しくなく、candIntraPredModeAまたはcandIntraPredModeBがINTRA_DCより大きい場合、下記が適用される：
－変数minABおよびmaxABは次のように導出される：
minAB＝Min(candIntraPredModeA,candIntraPredModeB) (8-24)
maxAB＝Max(candIntraPredModeA,candIntraPredModeB) (8-25)
－candIntraPredModeAとcandIntraPredModeBがいずれもINTRA_DCよりも大きい場合、x＝0…5として、candModeList[x]は次のように導出される：
candModeList[0]＝candIntraPredModeA (8-26)
candModeList[1]＝candIntraPredModeB (8-27)
－IntraLumaRefLineIdx[xCb][yCb]が0に等しく、IntraSubpartitionsSplitTypeがISP_NO_SPLITに等しい場合、下記が適用される：
candModeList[2]＝INTRA_PLANAR (8-28)
candModeList[3]＝INTRA_DC (8-29)
・maxAB－minABが2～62の範囲（両端を含む）にある場合は、下記が適用される：
candModeList[4]＝2＋((maxAB＋61)%64) (8-30)
candModeList[5]＝2＋((maxAB－1)%64) (8-31)
・それ以外の場合は、下記が適用される：
candModeList[4]＝2＋((maxAB＋60)%64) (8-32)
candModeList[5]＝2＋((maxAB)%64) (8-33)
－それ以外の場合（IntraLumaRefLineIdx[xCb][yCb]が0に等しくないか、またはIntraSubPartitionsSplitTypeがISP_NO_SPLITに等しくない場合）、下記が適用される：
－IntraSubPartitionsSplitTypeがISP_NO_SPLITに等しくなく、かつabs(candIntraPredModeB－ispDefaultMode1)がabs(candIntraPredModeA－ispDefaultMode1)より小さい場合は、下記が適用される：
candModeList[0]＝candIntraPredModeB (8-34)
candModeList[1]＝candIntraPredModeA (8-35)
－もしmaxAB－minABが1に等しい場合、下記が適用される：
candModeList[2]＝2＋((minAB＋61)%64) (8-36)
candModeList[3]＝2＋((maxAB－1)%64) (8-37)
candModeList[4]＝2＋((minAB＋60)%64) (8-38)
candModeList[5]＝2＋(maxAB%64) (8-39)
－それ以外の場合、maxAB－minABが2に等しい場合、下記が適用される：
candModeList[2]＝2＋((minAB－1)%64) (8-40)
candModeList[3]＝2＋((minAB＋61)%64) (8-41)
candModeList[4]＝2＋((maxAB－1)%64) (8-42)
candModeList[5]＝2＋((minAB＋60)%64) (8-43)
－それ以外の場合、maxAB－minABが61より大きい場合、下記が適用される：
candModeList[2]＝2＋((minAB－1)%64) (8-44)
candModeList[3]＝2＋((maxAB＋61)%64) (8-45)
candModeList[4]＝2＋(minAB%64) (8-46)
candModeList[5]＝2＋((maxAB＋60)%64) (8-47)
－それ以外の場合、下記が適用される：
candModeList[2]＝2＋((minAB＋61)%64) (8-48)
candModeList[3]＝2＋((minAB－1)%64) (8-49)
candModeList[4]＝2＋((maxAB＋61)%64) (8-50)
candModeList[5]＝2＋((maxAB－1)%64) (8-51)
－それ以外の場合（candIntraPredModeAまたはcandIntraPredModeBがINTRA_DCより大きい）、x＝0…5として、candModeList[x]は次のように導出される：
－IntraLumaRefLineIdx[xCb][yCb]が0に等しく、IntraSubpartitionsSplitTypeがISP_NO_SPLITに等しい場合、下記が適用される：
candModeList[0]＝candIntraPredModeA (8-52)
candModeList[1]＝candIntraPredModeB (8-53)
candModeList[2]＝1－minAB (8-54)
candModeList[3]＝2＋((maxAB＋61)%64) (8-55)
candModeList[4]＝2＋((maxAB－1)%64) (8-56)
candModeList[5]＝2＋((maxAB＋60)%64) (8-57)
－それ以外の場合で、IntraLumaRefLineIdx[xCb][yCb]が0に等しくない場合、下記が適用される：
candModeList[0]＝maxAB (8-58)
candModeList[1]＝2＋((maxAB＋61)%64) (8-59)
candModeList[2]＝2＋((maxAB－1)%64) (8-60)
candModeList[3]＝2＋((maxAB＋60)%64) (8-61)
candModeList[4]＝2＋(maxAB%64) (8-62)
candModeList[5]＝2＋((maxAB＋59)%64) (8-63)
－それ以外の場合（IntraSubPartitionsSplitTypeがISP_NO_SPLITと等しくない場合）、下記が適用される：
candModeList[0]＝INTRA_PLANAR (8-64)
candModeList[1]＝maxAB (8-65)
candModeList[2]＝2＋((maxAB＋61)%64) (8-66)
candModeList[3]＝2＋((maxAB－1)%64) (8-67)
candModeList[4]＝2＋((maxAB＋60)%64) (8-68)
candModeList[5]＝2＋(maxAB%64) (8-69)
－それ以外の場合は、下記が適用される：
－IntraLumaRefLineIdx[xCb][yCb]が0に等しく、IntraSubpartitionsSplitTypeがISP_NO_SPLITに等しい場合、下記が適用される：
candModeList[0]＝candIntraPredModeA (8-70)
candModeList[1]＝(candModeList[0]==INTRA_PLANAR)?INTRA_DC:INTRA_PLANAR (8-71)
candModeList[2]＝INTRA_ANGULAR50 (8-72)
candModeList[3]＝INTRA_ANGULAR18 (8-73)
candModeList[4]＝INTRA_ANGULAR46 (8-74)
candModeList[5]＝INTRA_ANGULAR54 (8-75)
－それ以外の場合で、IntraLumaRefLineIdx[xCb][yCb]が0に等しくない場合、下記が適用される：
candModeList[0]＝INTRA_ANGULAR50 (8-76)
candModeList[1]＝INTRA_ANGULAR18 (8-77)
candModeList[2]＝INTRA_ANGULAR2 (8-78)
candModeList[3]＝INTRA_ANGULAR34 (8-79)
candModeList[4]＝INTRA_ANGULAR66 (8-80)
candModeList[5]＝INTRA_ANGULAR26 (8-81)
－それ以外の場合で、IntraSubPartitionsSplitTypeがISP_HOR_SPLITに等しい場合、下記が適用される：
candModeList[0]＝INTRA_PLANAR (8-82)
candModeList[1]＝INTRA_ANGULAR18 (8-83)
candModeList[2]＝INTRA_ANGULAR25 (8-84)
candModeList[3]＝INTRA_ANGULAR10 (8-85)
candModeList[4]＝INTRA_ANGULAR65 (8-86)
candModeList[5]＝INTRA_ANGULAR50 (8-87)
－それ以外の場合で、IntraSubPartitionsSplitTypeがISP_VER_SPLITに等しい場合、下記が適用される：
candModeList[0]＝INTRA_PLANAR (8-88)
candModeList[1]＝INTRA_ANGULAR50 (8-89)
candModeList[2]＝INTRA_ANGULAR43 (8-90)
candModeList[3]＝INTRA_ANGULAR60 (8-91)
candModeList[4]＝INTRA_ANGULAR3 (8-e)
candModeList[5]＝INTRA_ANGULAR18 (8-93)
５．IntraPredModeY[xCb][yCb]は、以下の手順を適用して導出される：
－intra_luma_mpm_flag[xCb][yCb]が1に等しい場合、IntraPredModeY[xCb][yCb]はcandModeList[intra_luma_mpm_idx[xCb][yCb]]に等しく設定される。
－それ以外の場合、IntraPredModeY[xCb][yCb]は、次の順序付けられたステップを適用することによって導出される：
１．i＝0…4および各i、j＝(i＋1…5)についてcandModeList[i]がcandModeList[j]より大きい場合、両方の値は次のように交換される：
(candModeList[i],candModeList[j])＝Swap(candModeList[i],candModeList[j]) (8-94)
２．IntraPredModeY[xCb][yCb]は、以下の順序付けられたステップによって導出される：
ｉ．IntraPredModeY[xCb][yCb]は、intra_luma_mpm_remainder[xCb][yCb]に等しく設定される。
ｉｉ．0～5に等しいiについて、IntraPredModeY[xCb][yCb]がcandModeList[i]以上である場合、IntraPredModeY[xCb][yCb]の値は1だけインクリメントされる。
変数IntraPredModeY[x][y]（ここで、x＝xCb…xCb＋cbWidth－1およびy＝yCb…yCb＋cbHeight－1）は、IntraPredModeY[xCb][yCb]に等しく設定される。 Derivation process for Luma intra prediction mode The inputs to this process are as follows:
The luma position (xCb, yCb), which specifies the upper left sample of the current luma coding block for the upper left luma sample of the current picture,
Variable cbWidth, which specifies the width of the current coding block in units of luma samples,
A variable cbHeight that specifies the height of the current coding block in luma sample units.
In this process, the Luma Intra Prediction Mode IntraPredModeY [xCb] [yCb] is derived.
Table 8-1 specifies values for IntraPredModeY [xCb] [yCb] and related names.
Table 8-1-Specifying intra-prediction mode and associated names

Note: Intra prediction modes INTRA_LT_CCLM, INTRA_L_CCLM, INTRA_T_CCLM are only applicable to chroma components.
IntraPredModeY [xCb] [yCb] is derived by the following ordered steps:
1. 1. The neighborhood positions (xNbA, yNbA) and (xNbB, yNbB) are set equally to (xCb-1, yCb + cbHeight-1) and (xCb + cbWidth-1, yCb-1), respectively.
2. 2. For X replaced by either A or B, the variable candIntraPredModeX is derived as follows:
-6.4. The availability derivation process for blocks specified in Section X [Editor's Note (BB): Neighborhood block availability check process tbd] is set to be equal to (xCb, yCb) (xCurr, yCb). yCurr) and the neighborhood position (xNbY, yNbY) set to be equal to (xNbX, yNbX) are called as inputs and the output is assigned to availableX.
-CandIntraPredModeX is derived as follows:
-If one or more of the following conditions are true, canDIntraPredModeX is set equally to INTRA_PLANAR.
-The variable avalableX is equal to FALSE.
-CuPredMode [xNbX] [yNbX] is not equal to MODE_INTRA and ciip_flag [xNbX] [yNbX] is not equal to 1.
-Pcm_flag [xNbX] [yNbX] is equal to 1.
-X is equal to B and yCb-1 is less than ((yCb >> CtbLog2SizeY) << CtbLog2SizeY).
-Otherwise, candIntraPredModeX is set equal to IntraPredModeY [xNbX] [yNbX].
3. 3. The variables ispDefaultMode1 and ispDefaultMode2 are defined as:
-If IntraSubPartitionsSplitType is equal to ISP_HOR_SPLIT, ispDefaultMode1 is set equal to INTRA_ANGULAR18 and ispDefaultMode2 is set equal to INTRA_ANGULAR5.
-Otherwise, ispDefaultMode1 is set equal to INTRA_ANGULAR50 and ispDefaultMode2 is set equal to INTRA_ANGULAR63.
4. With x ＝ 0… 5, the candModeList [x] is derived as follows:
-If candIntraPredModeB is equal to candIntraPredModeA and candIntraPredModeA is greater than INTRA_DC, then with x = 0… 5, the candModeList [x] is derived:
-If IntraLumaRefLineIdx [xCb] [yCb] is equal to 0 and IntraSubPartitionsSplitType is equal to ISP_NO_SPLIT, the following applies:
candModeList [0] = candIntraPredModeA (8-9)
candModeList [1] = INTRA_PLANAR (8-10)
candModeList [2] = INTRA_DC (8-11)
candModeList [3] = 2 + ((candIntraPredModeA + 61)% 64) (8-12)
candModeList [4] = 2 + ((candIntraPredModeA-1)% 64) (8-13)
candModeList [5] = 2 + ((candIntraPredModeA + 60)% 64) (8-14)
-Otherwise (if IntraLumaRefLineIdx [xCb] [yCb] is not equal to 0 or IntraSubPartitionsSplitType is not equal to ISP_NO_SPLIT), the following applies:
candModeList [0] = candIntraPredModeA (8-15)
candModeList [1] = 2 + ((candIntraPredModeA + 61)% 64) (8-16)
candModeList [2] = 2 + ((candIntraPredModeA-1)% 64) (8-17)
-The following conditions:
-IntraSubpartitionsSplitType is equal to ISP_HOR_SPLIT and candIntraPredModeA is smaller than INTRA_ANGULAR34,
-IntraSubpartitionsSplitType is equal to ISP_VER_SPLIT and candIntraPredModeA is INTRA_ANGULAR34 or higher.
-If IntraLumaRefLineIdx [xCb] [yCb] is not equal to 0 and one of them is true, the following applies:
candModeList [3] = 2 + ((candIntraPredModeA + 60)% 64) (8-18)
candModeList [4] = 2 + (candIntraPredModeA% 64) (8-19)
candModeList [5] = 2 + ((candIntraPredModeA + 59)% 64) (8-20)
-Otherwise, the following applies:
candModeList [3] ＝ ispDefaultMode1 (8-21)
candModeList [4] ＝ ispDefaultMode2 (8-22)
candModeList [5] = INTRA_PLANAR (8-23)
-Otherwise, if candIntraPredModeB is not equal to candIntraPredModeA and candIntraPredModeA or candIntraPredModeB is greater than INTRA_DC, the following applies:
-Variables minAB and maxAB are derived as follows:
minAB ＝ Min (candIntraPredModeA, candIntraPredModeB) (8-24)
maxAB ＝ Max (candIntraPredModeA, candIntraPredModeB) (8-25)
-If both candIntraPredModeA and candIntraPredModeB are greater than INTRA_DC, then with x = 0… 5, the candModeList [x] is derived as follows:
candModeList [0] = candIntraPredModeA (8-26)
candModeList [1] = candIntraPredModeB (8-27)
-If IntraLumaRefLineIdx [xCb] [yCb] is equal to 0 and IntraSubpartitionsSplitType is equal to ISP_NO_SPLIT, the following applies:
candModeList [2] = INTRA_PLANAR (8-28)
candModeList [3] = INTRA_DC (8-29)
-If maxAB-minAB is in the range 2-62 (including both ends), the following applies:
candModeList [4] ＝ 2 ＋ ((maxAB ＋ 61)% 64) (8-30)
candModeList [5] = 2 + ((maxAB-1)% 64) (8-31)
-Otherwise, the following applies:
candModeList [4] ＝ 2 ＋ ((maxAB ＋ 60)% 64) (8-32)
candModeList [5] = 2 + ((maxAB)% 64) (8-33)
-Otherwise (if IntraLumaRefLineIdx [xCb] [yCb] is not equal to 0, or IntraSubPartitionsSplitType is not equal to ISP_NO_SPLIT), the following applies:
-If IntraSubPartitionsSplitType is not equal to ISP_NO_SPLIT and abs (candIntraPredModeB-ispDefaultMode1) is less than abs (candIntraPredModeA-ispDefaultMode1), the following applies:
candModeList [0] = candIntraPredModeB (8-34)
candModeList [1] = candIntraPredModeA (8-35)
-If maxAB-minAB is equal to 1, the following applies:
candModeList [2] ＝ 2 ＋ ((minAB ＋ 61)% 64) (8-36)
candModeList [3] = 2 + ((maxAB-1)% 64) (8-37)
candModeList [4] ＝ 2 ＋ ((minAB ＋ 60)% 64) (8-38)
candModeList [5] = 2 + (maxAB% 64) (8-39)
-Otherwise, if maxAB-minAB is equal to 2, the following applies:
candModeList [2] = 2 + ((minAB-1)% 64) (8-40)
candModeList [3] ＝ 2 ＋ ((minAB ＋ 61)% 64) (8-41)
candModeList [4] = 2 + ((maxAB-1)% 64) (8-42)
candModeList [5] ＝ 2 ＋ ((minAB ＋ 60)% 64) (8-43)
-Otherwise, if maxAB-minAB is greater than 61, the following applies:
candModeList [2] = 2 + ((minAB-1)% 64) (8-44)
candModeList [3] ＝ 2 ＋ ((maxAB ＋ 61)% 64) (8-45)
candModeList [4] = 2 + (minAB% 64) (8-46)
candModeList [5] ＝ 2 ＋ ((maxAB ＋ 60)% 64) (8-47)
-Otherwise, the following applies:
candModeList [2] ＝ 2 ＋ ((minAB ＋ 61)% 64) (8-48)
candModeList [3] = 2 + ((minAB-1)% 64) (8-49)
candModeList [4] ＝ 2 ＋ ((maxAB ＋ 61)% 64) (8-50)
candModeList [5] = 2 + ((maxAB-1)% 64) (8-51)
-Otherwise (candIntraPredModeA or candIntraPredModeB is greater than INTRA_DC), with x = 0… 5, the candModeList [x] is derived as follows:
-If IntraLumaRefLineIdx [xCb] [yCb] is equal to 0 and IntraSubpartitionsSplitType is equal to ISP_NO_SPLIT, the following applies:
candModeList [0] = candIntraPredModeA (8-52)
candModeList [1] = candIntraPredModeB (8-53)
candModeList [2] = 1－minAB (8-54)
candModeList [3] ＝ 2 ＋ ((maxAB ＋ 61)% 64) (8-55)
candModeList [4] = 2 + ((maxAB-1)% 64) (8-56)
candModeList [5] ＝ 2 ＋ ((maxAB ＋ 60)% 64) (8-57)
-Otherwise, if IntraLumaRefLineIdx [xCb] [yCb] is not equal to 0, the following applies:
candModeList [0] = maxAB (8-58)
candModeList [1] ＝ 2 ＋ ((maxAB ＋ 61)% 64) (8-59)
candModeList [2] = 2 + ((maxAB-1)% 64) (8-60)
candModeList [3] ＝ 2 ＋ ((maxAB ＋ 60)% 64) (8-61)
candModeList [4] = 2 + (maxAB% 64) (8-62)
candModeList [5] ＝ 2 ＋ ((maxAB ＋ 59)% 64) (8-63)
-Otherwise (if IntraSubPartitionsSplitType is not equal to ISP_NO_SPLIT), the following applies:
candModeList [0] = INTRA_PLANAR (8-64)
candModeList [1] = maxAB (8-65)
candModeList [2] ＝ 2 ＋ ((maxAB ＋ 61)% 64) (8-66)
candModeList [3] = 2 + ((maxAB-1)% 64) (8-67)
candModeList [4] ＝ 2 ＋ ((maxAB ＋ 60)% 64) (8-68)
candModeList [5] = 2 + (maxAB% 64) (8-69)
-Otherwise, the following applies:
-If IntraLumaRefLineIdx [xCb] [yCb] is equal to 0 and IntraSubpartitionsSplitType is equal to ISP_NO_SPLIT, the following applies:
candModeList [0] = candIntraPredModeA (8-70)
candModeList [1] = (candModeList [0] == INTRA_PLANAR)? INTRA_DC: INTRA_PLANAR (8-71)
candModeList [2] = INTRA_ANGULAR50 (8-72)
candModeList [3] ＝ INTRA_ANGULAR18 (8-73)
candModeList [4] = INTRA_ANGULAR46 (8-74)
candModeList [5] = INTRA_ANGULAR54 (8-75)
-Otherwise, if IntraLumaRefLineIdx [xCb] [yCb] is not equal to 0, the following applies:
candModeList [0] = INTRA_ANGULAR50 (8-76)
candModeList [1] = INTRA_ANGULAR18 (8-77)
candModeList [2] = INTRA_ANGULAR2 (8-78)
candModeList [3] = INTRA_ANGULAR34 (8-79)
candModeList [4] ＝ INTRA_ANGULAR66 (8-80)
candModeList [5] = INTRA_ANGULAR26 (8-81)
-Otherwise, if IntraSubPartitionsSplitType is equal to ISP_HOR_SPLIT, the following applies:
candModeList [0] = INTRA_PLANAR (8-82)
candModeList [1] = INTRA_ANGULAR18 (8-83)
candModeList [2] = INTRA_ANGULAR25 (8-84)
candModeList [3] ＝ INTRA_ANGULAR10 (8-85)
candModeList [4] ＝ INTRA_ANGULAR65 (8-86)
candModeList [5] = INTRA_ANGULAR50 (8-87)
-Otherwise, if IntraSubPartitionsSplitType is equal to ISP_VER_SPLIT, the following applies:
candModeList [0] = INTRA_PLANAR (8-88)
candModeList [1] = INTRA_ANGULAR50 (8-89)
candModeList [2] = INTRA_ANGULAR43 (8-90)
candModeList [3] = INTRA_ANGULAR60 (8-91)
candModeList [4] ＝ INTRA_ANGULAR3 (8-e)
candModeList [5] = INTRA_ANGULAR18 (8-93)
5. IntraPredModeY [xCb] [yCb] is derived by applying the following procedure:
-If intra_luma_mpm_flag [xCb] [yCb] is equal to 1, IntraPredModeY [xCb] [yCb] is set equal to candModeList [intra_luma_mpm_idx [xCb] [yCb]].
-Otherwise, IntraPredModeY [xCb] [yCb] is derived by applying the following ordered steps:
1. 1. If canandModeList [i] is greater than candModeList [i] for i ＝ 0… 4 and each i, j ＝ (i ＋ 1… 5), both values are exchanged as follows:
(candModeList [i], candModeList [j]) = Swap (candModeList [i], candModeList [j]) (8-94)
2. 2. IntraPredModeY [xCb] [yCb] is derived by the following ordered steps:
i. IntraPredModeY [xCb] [yCb] is set equal to intra_luma_mpm_remainder [xCb] [yCb].
ii. For i equal to 0-5, if IntraPredModeY [xCb] [yCb] is greater than or equal to candModeList [i], the value of IntraPredModeY [xCb] [yCb] is incremented by 1.
The variables IntraPredModeY [x] [y] (where x = xCb ... xCb + cbWidth-1 and y = yCb ... yCb + cbHeight-1) are set equally to IntraPredModeY [xCb] [yCb].

3. 3. Examples of problems solved by embodiments

In LIC, two parameters, including the scaling parameter and offset b, need to be derived by using a reconstructed sample in the neighborhood, which can cause delay problems.

The set of acceptable weighting factors used in GBI is fixed, which can be inefficient.

CIIP mode applies only to non-skip merge mode, which can be inefficient.

In the current design, the CIIP mode flag is the CIIP mode flag because the CIIP mode flag of the neighboring block is used for the intramode derivation process of the intracoding block and the intramode derivation process of the CIIP coding block . Need to be remembered.

In the current design, CIIP weight derivation can be further improved, for example, checking for two neighborhood blocks and three different weight pairs could be simplified.

In the current design, during the CIIP weight derivation process, if a neighboring block is coded in BDPCM mode, it is treated as intramode. However, since BDPCM is designed for screen content coding , it may be more rational to treat it as a non-intra mode.

4. Example of embodiment

Hereinafter, the block is used to represent a subblock in a conversion unit (TU) / prediction unit (PU) / coding unit (CU) / one TU / PU / CU or the like.

The following detailed techniques should be considered as examples to illustrate general concepts. These techniques should not be construed in a narrow sense. Moreover, these techniques can be combined in any way.

ここで、offsetは(1<<(N－1))または0に設定され、Nはlog2(wIntra＋wInter)に設定されてもよい。
１２．CIIPで使用される許容される重みペア(wIntra,wInter)の数は3から1に減らされてもよい。
ａ．一例では、1つの重みペアは、(1,3)または(2,2)または(1,7)または(2,6)または(3,5)または(4,4)として定義される。
ｂ．一例では、(wIntra,wInter)が(2,2)に等しく設定される。
ｃ．一例では、(wIntra,wInter)が(1,3)に等しく設定される。
ｄ．上記の例については、CIIP予測は次のように形成される：

ここで、offsetは(1<<(N－1))または0に設定され、Nはlog2(wIntra＋wInter)に設定されてもよい。
１３．(wIntr,wInter)の重みペアは、一つまたは複数の近傍ブロックがCIIPモードでコーディングされるかどうかに基づいて導出されてもよい。
ａ．現在ブロックについて使用される重みペアは、前にコーディングされたブロックについて使用された重みペアから導出されてもよい。
ｂ．一例では、重みペアは、1つの近傍ブロックAのみに基づいて決定される。
ｉ．一例では、1つの近傍ブロックAがCIIPモードでコーディングされる場合、現在ブロックの(wIntra,wInter)は近傍ブロックAの(wIntra,wInter)に等しく設定される；そうでない場合、近傍ブロックAが利用可能ありで、かつイントラコーディングされている場合、(wIntra,wInter)は(a,b)に等しく設定される；そうでない場合は、(wIntra,wInter)は(c,d)に等しく設定される。
ｉｉ．一例では、1つの近傍ブロックAがCIIPモードでコーディングされる場合、現在ブロックの(wIntra,wInter)は近傍の(wIntra,wInter)に等しく設定され；そうでない場合は、(wIntra,wInter)は(a,b)に等しく設定される。
ｉｉｉ．上記の例について、下記が適用されてもよい：
１．一例では、(a,b)が(2,2)または(3,1)に等しく設定される。
２．一例では、(c,d)が(1,3)に等しく設定される。
３．一例では、2つの条件の少なくとも一方が真である：aはcに等しくない；bはdに等しくない。
ｃ．一例では、重みペアは、2つ以上の近傍ブロックに基づいて決定され、前記近傍ブロックのうちの少なくとも1つについて、ある種の条件が満たされる。
ｉ．複数の隣接ブロックは、CIIPモードの使用を識別するために、所与のチェック順序でチェックされてもよい（たとえば、まず左ブロックをチェックし、次に上のブロックをチェックする）。
ｉｉ．一例では、近傍ブロックAおよびBの少なくとも一方がCIIPモードでコーディングされる場合、現在ブロックの(wIntra,wInter)は、所与のチェック順序において最初のCIIPコーディングされた近傍ブロックの(wIntra,wInter)に等しく設定される；そうでない場合、AおよびBのうち少なくとも1つの近傍ブロックが利用可能であり、かつイントラコーディングされる場合は、(wIntra,wInter)は(a,b)に等しく設定され；そうでない場合は(wIntra,wInter)は(c,d)に等しく設定される。
ｉｉｉ．上記の諸例について、下記が適用されてもよい：
１．一例では、(a,b)が(2,2)または(3,1)に等しく設定される。
２．一例では、(c,d)が(1,3)に等しく設定される。
３．一例では、2つの条件の少なくとも1つが真である：aはcに等しくない；bはdに等しくない。
ｉｖ．一例では、近傍ブロックAおよびBの少なくとも一方がCIIPモードでコーディングされる場合、現在ブロックの(wIntra,wInter)は、所与のチェック順序における最初のCIIPコーディングされた近傍ブロックの(wIntra,wInter)に等しく設定される；そうでない場合、近傍ブロックAおよびBがいずれも利用可能でありかつイントラコーディングされている場合は、(wIntra,wInter)は(a,b)に等しく設定される；そうでない場合は、(wIntra,wInter)は(c,d)に等しく設定される。
ｖ．一例では、近傍ブロックAおよびBの少なくとも一方がCIIPモードでコーディングされている場合、現在ブロックの(wIntra,wInter)は、所与のチェック順序における最初のCIIPコーディングされた近傍ブロックの(wIntra,wInter)に等しく設定され；そうでない場合は、(wIntra,wInter)は(c,d)に等しく設定される。
ｖｉ．上記の諸例について、下記が適用されてもよい：
１．一例では、(c,d)は(2,2)または(1,3)に等しく設定される。
ｄ．一例では、上述の近傍ブロックAまたはBは、隣接するまたは隣接しない空間的な近傍ブロック、または時間的な隣接ブロックである。
ｉ．一例では、近傍ブロックAまたはBは、左（または上）の近傍ブロックである。
ｉｉ．一例では、近傍ブロックAおよびBは、それぞれ左および上の近傍ブロックである。
ｉｉｉ．一例では、左の近傍ブロックは位置(xCb－1,yCb＋cbHeight－1)をカバーする。
ｉｖ．一例では、上の近傍ブロックは位置(xCb＋cbWidth－1,yCb－1)をカバーする。
ｖ．一例では、左の近傍ブロックは位置(xCb－1,yCb)をカバーする。
ｖｉ．一例では、上の近傍ブロックは位置(xCb,yCb－1)をカバーする。
ｅ．上記の諸例については、CIIP予測は次のように形成される：

ここで、offsetは(1<<(N－1))または0に設定され、Nはlog2(wIntra＋wInter)に設定されてもよい。
１４．CIIPにおいて使用される近傍ブロックの選択は、ブロック寸法／ブロック形状／低遅延チェック・フラグ／参照ピクチャー情報／現在ブロックおよび近傍ブロックの動き情報／近傍ブロックのイントラ予測モードのようなコーディングされた情報に依存しうる。
１５．CIIPにおいて使用される重みペアは、ブロック寸法／ブロック形状／低遅延チェック・フラグ／参照ピクチャー情報／現在ブロックおよび近傍ブロックの動き情報／近傍ブロックのイントラ予測モードのようなコーディングされた情報に依存しうる。
ａ．一例では、重み付け対の集合は、さらに信号伝達される。
ｂ．一例では、重みペアの選択は、近傍ブロックがイントラコーディングでなくインターコーディングされるかどうかに依存しうる。
１６．CIIP情報を決定するときは（たとえば、前述の以前の設計およびブレットのためのイントラ予測信号またはインター予測信号に適用される重みの決定においては）、近傍ブロックがMODE_INTRA（すなわち、イントラモード）でコーディングされるのでない場合、それはイントラコーディングされたものとして取り扱われてもよい。
ａ．一例では、近傍ブロックがIBCコーディングされている場合。
ｂ．一例では、近傍ブロックがCIIPコーディングされている場合。
ｃ．一例では、近傍ブロックがTPMコーディングされている場合。
ｄ．一例では、近傍ブロックがパレットコーディングされている場合。
ｅ．一例では、近傍ブロックがRDPCMコーディングされている場合。
ｆ．一例では、近傍ブロックが変換（たとえば、変換スキップ・モード）および／または量子化（たとえば、変換バイパス量子化モード）の適用なしでコーディングされている場合。
ｇ．あるいはまた、近傍ブロックが上記の諸サブブレットで言及されたモード（たとえば、IBC、RDPCM、パレット）でコーディングされている場合、近傍ブロックは、非イントラコーディングされている（たとえば、インターコーディングされている）として扱われてもよい。
ｈ．一例では、近傍ブロックがイントラコーディングされているが、ある種の予測モード（たとえば、平面）でコーディングされているのではない場合、近傍ブロックは、非イントラコーディングされている（たとえば、インターコーディングされている）として扱われてもよい。
ｉ．あるいはまた、近傍ブロックが、行列ベースのイントラ予測方法および／またはK（たとえば、K＝0）に等しくない参照線インデックスを有する複数参照線イントラ予測方法および／またはBDPCMによってコーディングされている場合、近傍ブロックは、非イントラコーディングされている（たとえば、インターコーディングされている）として扱われてもよい。
ｊ．提案される方法は、近傍ブロックがイントラモードでコーディングされるか否かに依拠する他のコーディングツールに適用されてもよい。
ｋ．提案される方法は、近傍ブロックがインター・モードでコーディングされるか否かに依拠する他のコーディングツールに適用されてもよい。
１７．上記の諸方法を有効にするか無効にするかは、SPS／PPSS／VPS／シーケンスヘッダ／ピクチャーヘッダ／スライスヘッダ／タイルグループヘッダ／タイル／CTUのグループ等において信号伝達されてもよい。
ａ．あるいはまた、どの方法が使用されるかが、SPS／PPSS／VPS／シーケンスヘッダ／ピクチャーヘッダ／スライスヘッダ／タイルグループヘッダ／タイル／CTUのグループ等において信号伝達されてもよい。
ｂ．あるいはまた、上記の諸方法を有効にするか無効にするか、および／またはどの方法が適用されるかは、ブロック寸法、ビデオ処理データ単位（video processing data unit、VPDU）、ピクチャータイプ、低遅延チェック・フラグ、現在ブロックのコーディング情報（たとえば参照ピクチャー、単予測または双予測）または以前にコーディングされたブロックに依存してもよい。 If (xCb, yCb) is the position of the sample on the upper left of the current block with respect to the sample on the upper left of the current picture, cbWidth and cbHeight specify the width and height of the current block.
1. 1. The CIIP flag and / and the LIC flag and / and the diffuse filtering flag and / and the bilateral filtering flag and / and the transformation domain filtering flag, and / and the enable flag of other types of post-reconstruction filters are other. Depending on the coding mode of the blocks (eg adjacent or non-adjacent neighboring blocks and / or reference blocks), they may be constrained to be false (and CIIP or / and LIC and / and diffusion filters or / and. Bilateral filters and / and translation domain filters may be implicitly disabled).
a. If one flag is derived as false, the corresponding method may not apply.
b. If one flag is derived as false, signaling of the use of such a mode is skipped.
c. If one flag is derived to be false, the instructions to use such a mode can still be signaled, but the compliant bitstream is constrained to be false and such a mode does not apply.
d. In one example, whether or not to apply one or more proposed constraints (such as 1.a, 1.b and 1.c) is the entire row or column of adjacent and / or non-adjacent neighborhoods. Or it may depend on the coding mode of some.
i. Alternatively, whether or not to apply such a constraint may depend on at least N samples (N ≧ 1) in adjacent or non-adjacent neighboring rows or columns that are not coded in certain modes.
e. In one example, adjacent and / or non-adjacent neighborhood rows may include the above row and / or the upper right row.
f. In one example, adjacent and / or non-adjacent neighborhood columns in the neighborhood may include the left column and / or the lower left and / or the upper left corner.
g. In one example, certain modes of blocks that are not currently blocks may include intramode and / or CIIP mode and / or CPR mode.
h. In one example, any of the neighboring / non-neighboring blocks in a nearby and / or non-neighboring row or column are coded in certain modes (eg, intra and / or CIIP mode and / or CPR mode). , One or more constraints (such as 1.a, 1.b and 1.c) apply.
i. In one example, if all neighboring / non-neighboring blocks in adjacent and / or non-adjacent neighborhood rows or columns are coded in certain modes (eg, intra and / or CIIP and / or CPR mode). , One or more proposed constraints apply.
j. In one example, if at least N neighboring / non-neighboring blocks in nearby or non-neighboring rows or columns are not coded in certain modes (eg, intra and / or CIIP mode and / or CPR mode). One or more proposed constraints (such as 1.a, 1.b and 1.c) do not apply.
k. In one example, whether or not to apply one or more proposed constraints (such as 1.a, 1.b and 1.c) may depend on the current block position.
i. In one example, if the current block is at the top of the current CTU (the current block and the neighboring blocks above it belong to different CTUs), the proposed constraint does not apply.
ii. In one example, if the current block is to the left of the current CTU (the current block and its adjacent blocks to the left belong to different CTUs), the proposed constraint does not apply.
2. 2. The CIIP flag and / or intramode of CIIP mode may be stored with motion information in a history-based motion vector prediction (HMVP) table.
a. In one example, when comparing the movement information of two candidates (such as the pruning process), the CIIP flag and / and intramode in CIIP mode are not considered in the comparison.
b. In one example, when comparing the motion information of two candidates, the CIIP flag and / and intramode in CIIP mode are taken into account in the comparison.
c. In one example, if the merge candidate is from an entry in the HMVP table, the CIIP flag for that entry is also copied to the merge candidate.
d. In one example, if the merge candidate is from an entry in the HMVP table, the CIIP flag and intramode for that entry will also be copied to the merge candidate.
3. 3. CIIP may be performed for AMVP mode (AMVP mode and / and AMVP mode with SMVD).
a. The CIIP flag and / or intramode in the CIIP mode of nearby / non-adjacent blocks is also known as the current block merge mode and / and UMVE (merge with motion vector difference), also known as MMVD for short. It may be inherited in the mode.
b. The CIIP flag and / or intramode in CIIP mode may be signaled for non-skip merge mode and / and non-skip UMVE mode, and the CIIP flag and / or intramode in CIIP mode for nearby / non-adjacent blocks. May be inherited in skip merge mode and / and skip UMVE mode.
c. The CIIP flag and / or intramode in CIIP mode may be signaled for skip merge mode and / and skip UMVE mode, and the CIIP flag and / or intramode in CIIP mode for nearby / non-adjacent blocks may be signaled. It may be inherited in non-skip merge mode and / and non-skip UMVE mode.
d. In one example, the CIIP flag and intramode may be signaled for AMVP mode.
e. In one example, the CIIP flag may be inherited in merge mode and / and UMVE mode, and if the CIIP flag is true, then intramode further signals if more than one intramode is allowed in CIIP. It may be transmitted.
f. In one example, the CIIP flags and intramodes of neighboring / non-adjacent blocks in merge mode and / and UMVE mode may be inherited by the current block.
g. In one example, CIIP may be disabled for skip mode.
h. In one example, when comparing two merge candidates, the CIIP flag and / and intramode in CIIP need not be considered in the comparison.
i. In one example, when comparing two merge candidates, the CIIP flag and / and intramode in CIIP may be considered in the comparison.
4. CIIP may be disabled for pair-to-pair predictions, compound bi-predictions, or other types of virtual / artificial candidates (eg, zero motion vector candidates).
a. Alternatively, if one of the two candidates involved in pair-by-pair or compound-bi-merge prediction adopts CIIP prediction, even if CIIP is enabled for the combined-bi-merge candidate. good.
ii. In one example, the intra mode of the CIIP mode candidate may be inherited.
iii. In one example, the intramode may be explicitly signaled.
b. Alternatively, CIIP may be enabled for paired or compound dual merge candidates if both candidates involved in paired or compound dual prediction adopt CIIP prediction.
iv. In one example, the intramode of one of the two candidates may be inherited.
v. In one example, the intramode may be derived from two candidate intramodes and is used for pair-by-pair or compound twin merge candidates.
vi. In one example, the intramode may be explicitly signaled.
5. Neighboring and / or non-neighboring spatially reconstructed samples required for CIIP mode and / and diffuse filters and / and bilateral filters and / and transformation domain filters and / and other types of post-reconstruction filters. May be replaced by a corresponding sample in the reference picture, such as a spatial sample near and / or not adjacent to the reference block, as shown in FIG.
a. In one example, if the current block is bi-predicted, the neighborhood samples of the two reference blocks may be averaged to produce the final neighborhood sample.
b. In one example, if the current block is bi-predicted, the neighborhood sample of one of the two reference blocks may be used as the final neighborhood sample.
c. In one example, if the current block is bi-predicted, if an unequal weighted GBI or weighted prediction or LIC is applied to the current block, the neighborhood samples of the two reference blocks are weighted and averaged to final. Peripheral samples may be generated.
d. In one example, the corresponding sample in the reference picture (eg, a sample near the reference block) may be identified by the motion information of the current block.
e. In one example, the corresponding sample in the reference picture (eg, a sample near the reference block) may be identified by the modified motion information of the current block. For example, motion vectors may be rounded to integer precision before being used to identify neighboring samples.
f. In one example, the corresponding sample in the reference picture (eg, the sample near the reference block) is identified via the motion vector with integer pixel accuracy.
i. In one example, the MV that references the reference block is first rounded to integer pixels. The rounded MV is used to identify nearby samples of the reference block.
ii. In one example, the proposed substitution is currently applied when the block is coded with integer pixel motion information. Therefore, no rounding is required.
g. In one example, the proposed substitution may only be applied for certain color components, such as the luma component only.
i. Alternatively, the proposed substitution may be applied for all color components.
6. It is proposed to prohibit the use of previously coded CIIP flags to code subsequent blocks.
a. In one example, the CIIP flag checking process from a previously coded block is skipped to improve throughput.
b. In one example, the process of checking CIIP flags from neighboring blocks in the intramode derivation process is skipped.
i. In one example, if for one neighborhood block it is coded in CIIP mode, the associated intra mode may be set to a given mode (eg, planar mode). Alternatively, in addition, the associated intramode may be used in the MPM list derivation process.
ii. In one example, if for one neighborhood block it is coded in CIIP mode or normal intermode, the associated intramode may be set to a given mode (eg, planar mode). Alternatively, in addition, the associated intramode may be used in the MPM list derivation process.
iii. In one example, the intramode of the CIIP coding block may not be remembered and the CIIP coding block may be considered unavailable in the decoding process. Alternatively, the CIIP coding block may be treated like a normal intermode.
c. Alternatively, it is further proposed to remove the CIIP flag from memory in order to save the memory size required to store the mode information.
7. It is suggested that the CIIP mode flag be coded by bypass coding .
a. Alternatively, the CIIP mode flag may be context coded , but does not refer to the CIIP mode flag of the neighboring block.
8. It is proposed that the weighted sum approach in CIIP may depend on the color components.
a. For example, the weighting values in CIIP differ between the main color component (eg, G component) and the other color components (eg, B and R components).
b. For example, the weighting values in CIIP are different for the luma component and the chroma component.
c. For example, the weighted values for inter-prediction and intra-prediction are equal for the chroma component.
9. If the weighting factor selection follows a neighborhood block, it is suggested that if a neighborhood block is coded as CIIP mode, it may be treated as an intercoding block.
a. Alternatively, if the weighting factor selection follows a neighborhood block, it may be treated as an intracoding block if it is coded as CIIP mode for a neighborhood block.
10. The proposed method described above may be applied under certain conditions such as block size, slice / picture / tile type, or motion information.
a. In one example, if the block size contains samples smaller than M × H, such as 16 or 32 or 64 luma samples, the proposed method is not allowed.
b. Alternatively, if the minimum width and / and height of the block is less than or equal to X, the proposed method is not allowed. In one example, X is set to 8.
c. Alternatively, if the minimum width and / and height of the block is greater than or equal to X, the proposed method is not allowed. In one example, X is set to 8.
d. Alternatively, if block width> th1 or ≧ th1 and / or block height> th2 or ≧ th2, the proposed method is not allowed. In one example, th1 and / or th2 is set to 8.
e. Alternatively, if the block width <th1 or ≤th1 and / or the block height <th2 or ≤th2, the proposed method is not allowed. In one example, th1 and / or th2 is set to 8.
11. The number of allowed weight pairs (wIntra, wInter) used in the CIIP shown as (a, b) and (c, d) may be reduced from 3 to 2.
a. In one example, the two pairs are {(1,3) and (3,1)}; or {(1,3) and (2,2)}; or {(3,1) and (2,2). }; Or {(3,5) and (4,4)} or {(5,3) and (4,4)} or {(1,7) and (4,4)} or {(7,1) ) And (4,4)}.
b. In one example, the weight pair is determined based on only one neighborhood block A.
i. In one example, if one neighborhood block A is available and is intracoded, (wIntra, wInter ) is set to be equal to (a, b); otherwise, (wIntra, wInter) is (wIntra, wInter). Set to be equal to c, d).
ii. In one example, block A is the neighborhood block on the left.
iii. In one example, block A is the upper neighborhood block.
c. In one example, the weight pair is determined based on two or more neighborhood blocks, and certain conditions are met for at least one of the neighborhood blocks.
i. In one example, (wIntra, wInter) is set equal to (a, b) if at least one of the neighborhood blocks A and B is available and is intracoded , otherwise (wIntra, wInter). Is set equal to (c, d).
1. 1. In one example, block A is the left neighborhood block and B is the top neighborhood block.
d. In one example, the weight pair is determined based on two or more neighborhood blocks, and the same conditions are met for each of the neighborhood blocks.
i. In one example, if neighborhood blocks A and B are both available and intracoded , (wIntra, wInter) is set equal to (a, b), otherwise (wIntra, wInter) is. Set equal to (c, d).
1. 1. In one example, block A is the left neighborhood block and B is the top neighborhood block.
e. In one example, the settings for (a, b) and (c, d) may be as follows:
1. 1. In one example, (a, b) is set to (2,2) or (3,1).
2. 2. In one example, (c, d) is set equal to (1,3).
3. 3. In one example, at least one of the two conditions is true: a is not equal to c, or b is not equal to d.
4. (a, b) is not equal to (c, d).
f. In one example, the neighborhood block described above (eg, A or B) is a spatial or temporal neighborhood block that is adjacent or non-adjacent.
i. In one example, the neighborhood block A or B is the left (or top) neighborhood block.
ii. In one example, neighborhood blocks A and B are neighborhood blocks on the left and above, respectively.
iii. In one example, the left neighborhood block covers the position (xCb-1, yCb + cbHeight-1).
iv. In one example, the upper neighborhood block covers the position (xCb + cbWidth-1, yCb-1).
v. In one example, the left neighborhood block covers the position (xCb-1, yCb).
vi. In one example, the upper neighborhood block covers the position (xCb, yCb-1).
g. For the above examples, the CIIP prediction is formed as follows:

Here, offset may be set to (1 << (N-1)) or 0, and N may be set to log2 (wIntra + wInter).
12. The number of allowable weight pairs (wIntra, wInter) used in CIIP may be reduced from 3 to 1.
a. In one example, one weight pair is defined as (1,3) or (2,2) or (1,7) or (2,6) or (3,5) or (4,4).
b. In one example, (wIntra, wInter) is set equal to (2,2).
c. In one example, (wIntra, wInter) is set equal to (1,3).
d. For the above example, the CIIP prediction is formed as follows:

Here, offset may be set to (1 << (N-1)) or 0, and N may be set to log2 (wIntra + wInter).
13. The weight pair (wIntr, wInter) may be derived based on whether one or more neighboring blocks are coded in CIIP mode.
a. The weight pair currently used for the block may be derived from the weight pair used for the previously coded block.
b. In one example, the weight pair is determined based on only one neighborhood block A.
i. In one example, if one neighborhood block A is coded in CIIP mode, the current block's (wIntra, wInter) is set equal to the neighborhood block A's (wIntra, wInter); otherwise, the neighborhood block A is utilized. If possible and intracoded , (wIntra, wInter) is set equal to (a, b); otherwise, (wIntra, wInter) is set equal to (c, d). ..
ii. In one example, if one neighborhood block A is coded in CIIP mode, the current block's (wIntra, wInter) is set equal to the neighborhood's (wIntra, wInter); otherwise, (wIntra, wInter) is (wIntra, wInter). It is set equally to a and b).
iii. For the above example, the following may apply:
1. 1. In one example, (a, b) is set equal to (2,2) or (3,1).
2. 2. In one example, (c, d) is set equal to (1,3).
3. 3. In one example, at least one of the two conditions is true: a is not equal to c; b is not equal to d.
c. In one example, the weight pair is determined based on two or more neighborhood blocks, and certain conditions are met for at least one of the neighborhood blocks.
i. Multiple adjacent blocks may be checked in a given check order to identify the use of CIIP mode (eg, check the left block first, then the block above).
ii. In one example, if at least one of the neighborhood blocks A and B is coded in CIIP mode, the current block (wIntra, wInter) is the first CIIP- coded neighborhood block (wIntra, wInter) in a given check order. If not, ( wIntra , wInter) is set equal to (a, b) if at least one neighborhood block of A and B is available and is intracoded; Otherwise (wIntra, wInter) is set equal to (c, d).
iii. For the above examples, the following may apply:
1. 1. In one example, (a, b) is set equal to (2,2) or (3,1).
2. 2. In one example, (c, d) is set equal to (1,3).
3. 3. In one example, at least one of the two conditions is true: a is not equal to c; b is not equal to d.
iv. In one example, if at least one of the neighborhood blocks A and B is coded in CIIP mode, the current block (wIntra, wInter) is the first CIIP- coded neighborhood block (wIntra, wInter) in a given check order. If not, (wIntra, wInter) is set equal to (a, b) if both neighborhood blocks A and B are available and intracoded ; If so, (wIntra, wInter) is set equal to (c, d).
v. In one example, if at least one of the neighborhood blocks A and B is coded in CIIP mode, the current block (wIntra, wInter) is the first CIIP- coded neighborhood block (wIntra, wInter) in a given check order. ) Equally; otherwise, (wIntra, wInter) is set equal to (c, d).
vi. For the above examples, the following may apply:
1. 1. In one example, (c, d) is set equal to (2,2) or (1,3).
d. In one example, the neighborhood blocks A or B described above are spatially adjacent or non-adjacent spatial neighborhood blocks, or temporally adjacent blocks.
i. In one example, the neighborhood block A or B is the left (or top) neighborhood block.
ii. In one example, neighborhood blocks A and B are neighborhood blocks on the left and above, respectively.
iii. In one example, the left neighborhood block covers the position (xCb-1, yCb + cbHeight-1).
iv. In one example, the upper neighborhood block covers the position (xCb + cbWidth-1, yCb-1).
v. In one example, the left neighborhood block covers the position (xCb-1, yCb).
vi. In one example, the upper neighborhood block covers the position (xCb, yCb-1).
e. For the above examples, the CIIP prediction is formed as follows:

Here, offset may be set to (1 << (N-1)) or 0, and N may be set to log2 (wIntra + wInter).
14. The selection of neighborhood blocks used in CIIP is coded information such as block dimensions / block shape / low delay check flag / reference picture information / current block and neighborhood block motion information / neighborhood block intra-prediction mode. Can depend on it.
15. The weight pairs used in CIIP depend on coded information such as block dimensions / block shape / low delay check flag / reference picture information / current block and neighborhood block motion information / neighborhood block intra-prediction mode. sell.
a. In one example, the set of weighted pairs is further signaled.
b. In one example, the choice of weight pairs may depend on whether neighboring blocks are intercoded rather than intracoded .
16. When determining CIIP information (for example, in determining the weights applied to the intra-predictive or inter-predictive signals for the previous design and bullets described above), the neighborhood block is coded in MODE_INTRA (ie, intra mode). If not, it may be treated as intracoded .
a. In one example, the neighborhood block is IBC coded .
b. In one example, the neighborhood block is CIIP coded .
c. In one example, the neighborhood block is TPM coded .
d. In one example, the neighborhood block is palette coded .
e. In one example, the neighborhood block is RDPCM coded .
f. In one example, the neighborhood block is coded without the application of transformation (eg, transformation skip mode) and / or quantization (eg, transformation bypass quantization mode).
g. Alternatively, if the neighborhood block is coded in the mode mentioned in the subblets above (eg, IBC, RDPCM, palette), the neighborhood block is non- intracoded (eg, intercoded ). ) May be treated.
h. In one example, if the neighborhood block is intracoded , but not in some predictive mode (eg, plane), the neighborhood block is non- intracoded (eg, interconnected ) . May be treated as).
i. Alternatively, if the neighborhood block is coded by a matrix-based intra-prediction method and / or a multi-reference line intra-prediction method with a reference line index that is not equal to K (eg, K = 0) and / or BDPCM, the neighborhood. Blocks may be treated as non- intracoded (eg, intercoded ).
j. The proposed method may be applied to other coding tools that rely on whether neighboring blocks are coded in intramode.
k. The proposed method may be applied to other coding tools that depend on whether the neighboring blocks are coded in intermode.
17. Whether to enable or disable the above methods may be signaled in a group of SPS / PPSS / VPS / sequence header / picture header / slice header / tile group header / tile / CTU.
a. Alternatively, which method is used may be signaled in a group such as SPS / PPSS / VPS / sequence header / picture header / slice header / tile group header / tile / CTU.
b. Alternatively, whether to enable or disable the above methods and / or which method is applied depends on the block size, video processing data unit (VPDU), picture type, and low latency. It may depend on the check flag, the coding information of the current block (eg reference picture, mono-predictive or bi-predictive) or the previously coded block.

5. Embodiment

5.1 Embodiment 1

注記：イントラ予測モードINTRA_LT_CCLM、INTRA_L_CCLM、INTRA_T_CCLMはクロマ成分にのみ適用できる。
IntraPredModeY[xCb][yCb]は、以下の順序付けられたステップによって導出される：
１．近傍位置(xNbA,yNbA)および(xNbB,yNbB)は、それぞれ(xCb－1,yCb＋cbHeight－1)および(xCb＋cbWidth－1,yCb－1)に等しく設定される。
２．AまたはBのいずれかで置き換えられるXについて、変数candIntraPredModeXは次のように導出される：
－6.4.X節に規定されるブロックについての利用可能性導出プロセス［編注（BB）：近傍ブロック利用可能性検査プロセスtbd］は、(xCb,yCb)に等しいと設定された位置(xCurr,yCurr)と、(xNbX,yNbX)に等しいと設定された近傍位置(xNbY,yNbY)とを入力として呼び出され、出力はavailableXに割り当てられる。
－候補イントラ予測モードcandIntraPredModeXは以下のように導出される：
－下記の条件の一つまたは複数が真の場合、candIntraPredModeXはINTRA_PLANARに等しく設定される。
－変数avalableXはFALSEに等しい。
－CuPredMode[xNbX][yNbX]はMODE_INTRAと等しくない。
－pcm_flag[xNbX][yNbX]は1に等しい。
－XはBに等しく、yCb－1は((yCb>>CtbLog2SizeY)<<CtbLog2SizeY)より小さい。
－それ以外の場合、candIntraPredModeXはIntraPredModeY[xNbX][yNbX]に等しく設定される。
３．変数ispDefaultMode1およびispDefaultMode2は次のように定義される：
－IntraSubPartitionsSplitTypeがISP_HOR_SPLITに等しい場合、ispDefaultMode1はINTRA_ANGULAR18に等しく設定され、ispDefaultMode2はINTRA_ANGULAR5に等しく設定される。
－そうでない場合、ispDefaultMode1はINTRA_ANGULAR50に等しく設定され、ispDefaultMode2はINTRA_ANGULAR63に等しく設定される。
４．x＝0…5として、candModeList[x]は次のように導出される：
… An example of skipping the CIIP flag check for a previously coded block is given below.
8.4.2 Derivation process for Luma Intra prediction mode
Derivation process for Luma intra prediction mode
The inputs to this process are:
-Ruma position (xCb, yCb), which specifies the upper left sample of the current luma coding block for the upper left luma sample of the current picture,
Variable cbWidth, which specifies the width of the current coding block in units of luma samples,
A variable cbHeight that specifies the height of the current coding block in luma sample units.
In this process, the Luma Intra Prediction Mode IntraPredModeY [xCb] [yCb] is derived.
Table 8-1 specifies values for IntraPredModeY [xCb] [yCb] and related names.
Table 8-1-Specifying intra-prediction mode and associated names

Note: Intra prediction modes INTRA_LT_CCLM, INTRA_L_CCLM, INTRA_T_CCLM are only applicable to chroma components.
IntraPredModeY [xCb] [yCb] is derived by the following ordered steps:
1. 1. The neighborhood positions (xNbA, yNbA) and (xNbB, yNbB) are set equally to (xCb-1, yCb + cbHeight-1) and (xCb + cbWidth-1, yCb-1), respectively.
2. 2. For X replaced by either A or B, the variable candIntraPredModeX is derived as follows:
-6.4. The availability derivation process for blocks specified in Section X [Editor's Note (BB): Neighborhood block availability check process tbd] is set to be equal to (xCb, yCb) (xCurr, yCb). yCurr) and the neighborhood position (xNbY, yNbY) set to be equal to (xNbX, yNbX) are called as inputs and the output is assigned to availableX.
-CandIntraPredModeX is derived as follows:
-If one or more of the following conditions are true, canDIntraPredModeX is set equally to INTRA_PLANAR.
-The variable avalableX is equal to FALSE.
-CuPredMode [xNbX] [yNbX] is not equal to MODE_INTRA.
-Pcm_flag [xNbX] [yNbX] is equal to 1.
-X is equal to B and yCb-1 is less than ((yCb >> CtbLog2SizeY) << CtbLog2SizeY).
-Otherwise, candIntraPredModeX is set equal to IntraPredModeY [xNbX] [yNbX].
3. 3. The variables ispDefaultMode1 and ispDefaultMode2 are defined as:
-If IntraSubPartitionsSplitType is equal to ISP_HOR_SPLIT, ispDefaultMode1 is set equal to INTRA_ANGULAR18 and ispDefaultMode2 is set equal to INTRA_ANGULAR5.
-Otherwise, ispDefaultMode1 is set equal to INTRA_ANGULAR50 and ispDefaultMode2 is set equal to INTRA_ANGULAR63.
4. With x ＝ 0… 5, the candModeList [x] is derived as follows:
…

Embodiment # 2
If the neighborhood on the left is available and intracoded , (wIntra, wInter) is set equal to (2,2); otherwise, (wIntra, wInter) is set equal to (1,3). Will be done.
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 3
If the neighborhood on the left is available and intracoded , (wIntra, wInter) is set equal to (3,1); otherwise, (wIntra, wInter) is set equal to (1,3). Will be done.
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 4
If the above neighborhood is available and intracoded , (wIntra, wInter) is set equal to (2,2); otherwise, (wIntra, wInter) is set equal to (1,3). Will be done.
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 5
If the above neighborhood is available and intracoded , (wIntra, wInter) is set equal to (3,1); otherwise, (wIntra, wInter) is set equal to (1,3). Will be done.
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 6
If the left and upper neighborhoods are available and intracoded , (wIntra, wInter) is set equal to (2,2); otherwise, (wIntra, wInter) is (1,3). ) Is set equally.
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 7
If the left and upper neighborhoods are available and intracoded , (wIntra, wInter) is set equal to (3,1); otherwise, (wIntra, wInter) is (1,3). ) Is set equally.
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 8
If the left or top neighborhood is available and intracoded , (wIntra, wInter) is set equal to (2,2); otherwise, (wIntra, wInter) is set to (1,3). Set equally.
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.
Embodiment # 9
If the left or top neighborhood is available and intracoded , (wIntra, wInter) is set equal to (3,1); otherwise, (wIntra, wInter) is set to (1,3). Set equally.
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 10
(wIntra, wInter) is set equal to (2,2).
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 11
(wIntra, wInter) is set equal to (1,3).
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 12
If the left neighborhood is CIIP coded , the current block's (wIntra, wInter) is set equal to the left neighborhood (wIntra, wInter); otherwise, the left neighborhood is available and intracoded . If yes, (wIntra, wInter) is set equal to (3,1); otherwise, (wIntra, wInter) is set equal to (1,3).
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 13
If the left neighborhood is CIIP coded , the current block's (wIntra, wInter) is set equal to the left neighborhood (wIntra, wInter); otherwise, the left neighborhood is available and intracoded . If yes, (wIntra, wInter) is set equal to (2,2); otherwise, (wIntra, wInter) is set equal to (1,3).
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 14
If the upper neighborhood is CIIP coded , the current block's (wIntra, wInter) is set equal to the upper neighborhood (wIntra, wInter); otherwise, the upper neighborhood is available and intracoded . If yes, (wIntra, wInter) is set equal to (3,1); otherwise, (wIntra, wInter) is set equal to (1,3).
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 15
If the upper neighborhood is CIIP coded , the current block's (wIntra, wInter) is set equal to the upper neighborhood (wIntra, wInter); otherwise, the upper neighborhood is available and intracoded . If yes, (wIntra, wInter) is set equal to (2,2); otherwise, (wIntra, wInter) is set equal to (1,3).
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 16
If the left or / and upper neighborhood is CIIP- coded , the (wIntra, wInter) of the current block is set equal to the (wIntra, wInter) of the first CIIP- coded neighborhood (left → top); If the left or / and upper neighborhood are available and intracoded , (wIntra, wInter) is set equal to (3,1); otherwise, (wIntra, wInter) is (1). , 3) are set equally.
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 17
If the left or / and upper neighborhood is CIIP- coded , the (wIntra, wInter) of the current block is set equal to the (wIntra, wInter) of the first CIIP- coded neighborhood (left → top); If the left or / and upper neighborhood is available and intracoded , (wIntra, wInter) is set equal to (2,2); otherwise, (wIntra, wInter) is (1). , 3) are set equally.
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 18
If the left neighborhood is CIIP coded , the current block's (wIntra, wInter) is set equal to the left neighborhood (wIntra, wInter); otherwise, (wIntra, wInter) is set to (1,3). Set equally.
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 19
If the left neighborhood is CIIP coded , the current block's (wIntra, wInter) is set equal to the left neighborhood (wIntra, wInter); otherwise, (wIntra, wInter) is set to (2,2). Set equally.
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 20
If the upper neighborhood is CIIP coded , the current block's (wIntra, wInter) is set equal to the upper neighborhood's (wIntra, wInter); otherwise, (wIntra, wInter) is set to (1,3). Set equally.
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 21
If the upper neighborhood is CIIP coded , the current block's (wIntra, wInter) is set equal to the upper neighborhood's (wIntra, wInter); otherwise, (wIntra, wInter) is set to (2,2). Set equally.
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 22
If the left or / and upper neighborhood is CIIP coded , the (wIntra, wInter) of the current block is set equal to the (wIntra, wInter) of the first CIIP coded neighborhood (left → top); If so, (wIntra, wInter) is set equal to (1,3).
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 23
If the left or / and upper neighborhood is CIIP coded , the (wIntra, wInter) of the current block is set equal to the (wIntra, wInter) of the first CIIP coded neighborhood (left → top); If so, (wIntra, wInter) is set equal to (2,2).
CIIP forecast
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ 2) >> 2
It is formed like.

Embodiment # 24
8.5.6 Decoding process for interblocks
8.5.6.1 General This process is called when decoding a coding unit in interpredictive mode.
If ciip_flag [xCb] [yCb] is equal to 1, the array of predicted samples predSamples is modified as follows:
-If cIdx is equal to 0, the following applies:
-The general intra-sample prediction process specified in Section 8.4.5.2.5 is the conversion block width nTbW set equal to the position (xTbCmp, yTbCmp) set equal to (xCb, yCb), cbWidth and cbHeight. And the coding block width nCbW and height nCbH set equal to the height nTbH, cbWidth and cbHeight, and the variable cIdx are called as inputs, and the output is an array predSamplesIntra _L of (cbWidth) × (cbHeight).
-The weighted sample prediction process for combined merge and intra predictions specified in Section 8.5.6.7 is the position (xTbCmp, yTbCmp) set equal to (xCb, yCb), coding block width cbWidth, coding . Called and output with block height cbHeight, predSamplesInter and PredSamplesIntra set to be equal to predSamples and predSamplesIntra _L , respectively, intraprediction mode predModeIntra set to be equal to IntraPredModeY [xCb] [yCb], and color component index cIdx. Is assigned to the array predSamples of (cbWidth) × (cbHeight).
...
8.5.6.7 Weighted sample prediction process for combined merge and intra prediction
The inputs to this process are:
-Ruma position (xCb, yCb), which specifies the upper left sample of the current luma coding block for the upper left luma sample of the current picture,
-Current coding block width cbWidth,
-Current coding block height cbHeight,
-Two (cbWidth) x (cbHeight) arrays, preSamplesInter and preSamplesIntra,
-Variable cIdx that specifies the color component index.
The output of this process is a (cbWidth) x (cbHeight) array of predicted sample values, preSamplesComb.
The variable bitDepth is derived as follows:
-If cIdx is equal to 0, bitDepth is set equal to BitDepth _Y.
-Otherwise, bitDepth is set equal to BitDepth _C.
The variable scallFact is derived as follows:
scallFact ＝ (cIdx == 0)? 0: 1 (8-838)
Neighborhood luma positions (xNbA, yNbA) and (xNbB, yNbB) are set equal to (xCb-1, yCb-1 + (cbHeight << scallFact)) and (xCb-1 + (cbWidth << scallFact), yCb-1), respectively. Will be done.
For X replaced by either A or B, the variables availableX and isIntraCodedNeighbourX are derived as follows:
-The neighborhood location (xCurr, yCurr), (xNbY, yNbY) set equal to (xCb, yCb) and the neighborhood position (xNbY, yNbY) set equally by the derivation process for neighborhood block availability specified in Section 6.4.4 (xCb, yCb). xNbY, yNbY), checkPredModeY set equal to FALSE, and cIdx set equal to 0 are called as inputs and the output is assigned to availableX.
-The variable isIntraCodedNeighbourX is derived as follows:
-If avallableX is equal to TRUE, CuPredMode [0] [xNbX] [yNbX] is equal to MODE_INTRA, and BdpcmFlag [xNbX] [yNbX] is equal to 0, isIntraCodedNeighbourX is set equal to TRUE.
-Otherwise, isIntraCodedNeighbourX is set equal to FALSE.
The weight w is derived as follows:
-If isIntraCodedNeighbourA and isIntraCodedNeighbourB are both equal to TRUE, w is set equal to 3.
-Otherwise, if isIntraCodedNeighbourA and isIntraCodedNeighbourB are both equal to FALSE, w is set to 1.
-Otherwise, w is set equal to 2.

If cIdx is equal to 0 and slice_lmcs_enabled_flag is equal to 1, predSamplesInter [x] [y] (x = 0 ... cbWidth-1 and y = 0 ... cbHeight-1) is modified as follows.

FIG. 18 is a block diagram of the video processing device 1800. Device 1800 may be used to implement one or more of the methods described herein. The device 1800 may be embodied in a smartphone, tablet, computer, Internet of Things (IoT) receiver, or the like. Equipment 1800 may include one or more processors 1802, one or more memories 1804, and video processing hardware 1806. Processor 1802 may be configured to implement one or more of the methods described herein. Memory (s) 1804 may be used to store data and codes used to implement the methods and techniques described herein. Video processing hardware 1806 may be used in hardware circuits to implement some of the techniques described in this article.

FIG. 20 is a flow chart for Method 2000 to process video. Method 2000 constrains one or more flags to a certain operating state based on the determination of the coding mode of the first video block (2005) and the determination of the coding mode of the first video block. The first video block comprises the second step of performing further processing of the second video block according to the operating state of the one or more flags (2110). A video block in the vicinity or a reference video block with respect to the video block in.

FIG. 21 is a flow chart for method 2100 to process video. Method 2100 is based on the step (2105) of determining that pair-to-pair or compound bi-prediction is used for the first video block and the determination that pair-to-pair or compound bi-prediction is used. , A step (2110) of determining the operational state of the composite inter-intra prediction, the step in which the operational state is enabled or disabled, and the further of the first video block, depending on the operational state of the CIIP. Includes a step (2115) to perform the process.

FIG. 23 is a flowchart of the method 2300 for processing video. Method 2300 determines the coding mode of one or more of the second blocks during the conversion between the first block and the bitstream representation of the first block in the video data. A step (2305) and a step (2310) of determining the coding mode constraint of the first block based on the one or more coding modes of the one or more second blocks, and at least the first. The one or more second blocks include adjacent blocks, non-adjacent blocks, and adjacent blocks of the first block, including the step (2315) of performing the transformation by applying the coding mode constraint of the block. Contains at least one of the reference blocks.

FIG. 24 is a flowchart for the method 2400 for processing video. Method 2400 transforms the conversion between the current block and the bitstream representation of the current block in the video data into compound inter-intra-prediction (CIIP), diffusion filtering, bilateral filtering, transformation domain filtering, or diffusion filtering, by It involves performing by using at least one of lateral filtering, and another type of post-reconstruction filtering that is different from translation domain filtering (2405), where compound inter-intra prediction (CIIP),. The reconstructed neighborhood sample of the current block used in at least one of diffusion filtering, bilateral filtering, transformation domain filtering, or another type of post-reconstruction filtering is in one or more reference pictures. , Replaced by an approximate sample generated from the corresponding sample of the reconstructed neighborhood sample.

FIG. 25 is a flow chart for Method 2500 to process video. Method 2500 stores the CIIP mode's composite inter-intra-prediction (CIIP) flag and / or intra-mode, along with motion information, in a history-based motion vector prediction (HMVP) table. 2505) and; including performing a conversion between the current block in the video data and the bitstream representation of the current block, at least based on the HMVP table.

FIG. 26 is a flowchart for the method 2600 for processing video. Method 2600 is a step (2605) of determining the prediction mode for the current block during the conversion between the current block and the bitstream representation of the current block in the video data; the current block is an advanced motion vector prediction (2605). Steps to determine the applicability of compound inter-intra-prediction (CIIP) mode, which indicates that CIIP mode is currently enabled for the block, in response to the determination that it is coded in AMVP) mode or merge mode (1610). And; including the step (2615) of performing the conversion based on the applicability of the CIIP mode.

FIG. 27 is a flow chart for the method 2700 to process video. Method 2700 is a step (2705) of determining the type of selected merge candidate for the current block during the conversion between the current block in the video data and the bitstream representation of the current block; The current block is coded in merge mode, including step (2710) to determine the applicability of compound inter-intra prediction (CIIP) for the current block depending on the type of merge candidate made.

FIG. 28 is a flow chart for the method 2800 to process video. Method 2800 is a CIIP flag for one or more neighboring video blocks to the current video block during conversion between the current video block in the video data and the bitstream representation of the current video block. Coding the Composite Inter-Intra Prediction (CIIP) flag for the current video block (2805) by coding based on the context model without reference to; at least the composite inter of the current video block. Includes step (2810) of performing the transformation by applying the Intra Prediction (CIIP) flag.

FIG. 29 is a flow chart for Method 2900 to process video. Method 2900 is a composite inter-intra prediction (CIPP) for the current video block by bypass coding during conversion between the current video block in the video data and the bitstream representation of the current video block. Includes a step of coding the flag (2905) and at least performing the transformation by applying the compound inter-intra prediction (CIIP) flag (2910).

FIG. 30 is a flowchart for the method 3000 for processing video. Method 3000 is a step of determining the intra-prediction mode of the first video block of video according to a rule: the first video during the process of deriving the intra-prediction mode of the first video block. With step (3005), including skipping checking the compound inter-intra-prediction (CIIP) flag for one or more neighboring video blocks to the block; at least based on the determined intra-prediction mode. Includes a step (3010) of performing a conversion between the first video block and the bitstream representation of the first video block.

FIG. 31 is a flow chart for method 3100 to process video. Method 3100 is a video block during conversion between the current video block coded in compound intra and inter-predictive (CIIP) mode of the video and the bitstream representation of the current video block. A weight pair containing a first weight for the first prediction result of the current video block and a second weight for the second prediction result of the current video block for one or more neighboring video blocks to the current video block. The first prediction result is generated by the intra prediction mode, and the second prediction result is generated by the inter prediction mode, which is a step of determining based on the block (3105); It includes a step (3110) of determining the prediction result of the current block based on the weighted sum of the prediction result and the second prediction result.

With reference to Methods 2000 and 2100, some examples of compound inter-intra prediction and its use for encoding are given in Section 4 of this article. For example, as described in Section 4, video blocks can be used and processed according to compound intra-inter predictions.

Referring to methods 2000 and 2100, the video block may be encoded in a bitstream, where bitefficiency is achieved by using the bitstream generation rules associated with compound intra-interprediction. sell.

In these methods, the one or more flags are compound inter-intra prediction (CIIP) flag, local lighting compensation (LIC) flag, diffuse filtering flag, bilateral filtering flag, conversion domain filter flag, or It can include, including another type of post-reconstruction filter flag.

In these methods, the one or more flags include the first flag associated with the first action, the first flag is derived to be false, and the first action does not apply. Can be included.

These methods can include signaling that the use of said first operation is skipped.

These methods can include signaling such that the use of the first operation is constrained to be false in the conforming bitstream.

These methods can include the first video block being in a nearby row or column with respect to the second video block.

These methods can include a nearby row containing an upper row or an upper right row.

These methods can include neighboring columns including a left column, a lower left column, or an upper left corner column.

These methods can include coding modes including intra mode, CIIP mode, or CPR mode.

These methods can include determining the position of the second video block, and constraining the one or more flags to a false state is responsible for determining the position of the second video block. Also based on.

These methods can include that the location is at the top of the current coding tree unit (CTU) and the neighboring blocks above the second video block are in different CTUs.

These methods can include that the location is at the left edge of the current coding tree unit (CTU) and the left neighborhood block of the second video block is in a different CTU.

In these methods, the one or more flags include the intramode of the CIIP flag or the CIIP mode flag, and the data related to the one or more flags are in the history-based motion vector prediction (HMVP) table. It can include being stored with motion information.

These methods can include comparing the motion information of two candidates, the CIIP flag or the intramode CIIP mode flag is not used in the motion information comparison.

These methods can include comparing the motion information of two candidates, the CIIP flag or the intramode CIIP mode flag being used in the comparison of motion information.

These methods can include determining that the merge candidate is from an entry in the HMVP table and copying the CIIP flag of the entry to the merge candidate.

These methods can include determining that the merge candidate is from an entry in the HMVP table and copying the CIIP flag and intramode of the entry to the merge candidate.

These methods can include that one of the flags is related to CIIP and that CIIP is performed for Advanced Motion Vector Prediction (AMVP).

These methods can include signaling the CIIP flag and the intramode flag for AMVP mode.

These methods include the operating state in which the CIIP flag is inherited in merge mode or UMVE mode and the CIIP flag is true, multiple intramodes are allowed in CIIP, and the multiple intramodes allowed in CIIP. Intramode can include signal transduction based on.

These methods can include the CIIP flags and intramodes of nearby or non-adjacent video blocks in merge mode or UMVE mode being inherited by the second video block.

These methods can include disabling CIIP for skip mode.

These methods can include comparing information related to the two candidates, and the CIIP flag or intramode CIIP mode flag is not used in the comparison of information.

These methods can include comparing information related to the two candidates, and the CIIP flag or the intramode CIIP mode flag is used in the comparison of information.

These methods can include one or more candidates involved in pair-by-pair or compound bi-prediction adopting CIIP prediction and enabling said operational state.

These methods can include inheriting the intramode of one of the two candidates.

These methods can include signal transduction of intramode.

These methods can include two candidates involved in pair-by-pair or compound bi-prediction.

These methods include replacing the reconstructed sample used in CIIP mode, diffusion filter, bilateral filter, transformation domain filter, or other type of post-reconstruction filter with the sample in the reference picture. be able to.

These methods can include bipredicting the second video block and averaging the neighborhood samples of the two reference blocks to produce the final neighborhood sample.

These methods can include bipredicting the second video block and using the neighborhood sample of the two reference blocks as the final neighborhood sample.

In these methods, the second video block is bi-predicted, and the neighborhood samples of the two reference blocks are weighted and averaged based on the generalized bi-prediction (GBI) of unequal weights to achieve the final neighborhood. It can include generating a sample or applying a weighted predictive LIC to a second video block.

These methods can include identifying the sample in the reference picture by the motion information of the second video block.

These methods can include identifying the sample in the reference picture by the motion information of the second video block.

These methods can include that the motion vector is rounded to integer precision and used to identify the sample.

These methods can include the sample being in a reference picture identified by a motion vector with integer pixel precision.

These methods can include determining the characteristics of a second video block, said characteristics being one or more of block size, slice type, picture type, tile type, or motion information. Including, and constraining the operating state is based on the determination of the features.

These methods are within a group of sequence parameter sets (SPS), view parameter sets (VPS), picture parameter sets (PPS), sequence headers, picture headers, slice headers, tile group headers, tiles, or CTUs. Signal transmission can include.

The following list of solutions provides further embodiments and variants for the items listed in the previous section (eg, items 11-16).

40. A method of video processing comprising performing a conversion between a video block of a video and a coded representation of the video block using composite intra and interpredictive modes, said coding representation. Corresponds to the weighted average of the intra-prediction and inter-prediction results of the video block, using weight pairs from a set of weight pairs containing less than three pairs.

41. The method of solution 40, wherein the set of weight pairs corresponds to two pairs.

42. The method of any of solutions 40 to 41, wherein the weight pair is determined based on a single neighborhood block.

43. The method of any of solutions 40 to 41, wherein the weight pair is determined based on a plurality of neighboring blocks.

44. The method of solution 40, wherein the set of weight pairs comprises exactly one pair.

45. The solution, wherein the pair is one of (1,3) or (2,2) or (1,7) or (2,6) or (3,5) or (4,4). 44.

46. A method of video processing comprising performing a conversion between a video block of a video and a coded representation of the video block using composite intra and interpredictive modes, said coding representation. Is a method corresponding to a weighted average of an intra-prediction result and an inter-prediction result of the video block using weight pairs from a set of weight pairs determined by coding information of one or more neighboring blocks.

47. The method of solution 46, wherein the set of weight pairs is determined by just one neighborhood block.

48. 46. The method of solution 46, wherein the set of weight pairs is determined by two or more neighboring blocks.

49. The method of any of solutions 40-48, wherein the conversion comprises encoding the video to produce the coded representation.

50. The method of any of solutions 40-48, wherein the conversion comprises decoding the coded representation to produce the video.

Various other embodiments and additional features of these methods can be described using the clause-based description below.

1. 1. A way to process video:
With the step of determining one or more coding modes of one or more second blocks during the conversion between the first block in the video data and the bitstream representation of the first block;
With the step of determining the coding mode constraint of the first block based on the coding mode of the one or more second blocks;
Including the step of performing the transformation by applying the coding mode constraint of at least the first block.
The one or more second blocks include at least one of adjacent blocks, non-adjacent blocks, and reference blocks of the first block.
Method.

2. 2. Determining the coding mode constraint for the first block:
Determining one or more flags corresponding to one or more coding modes of the first block, respectively, based on the one or more coding modes of the one or more second blocks. including,
The method described in Clause 1.

3. 3. The one or more coding modes mentioned above:
Different from Composite Inter-Intra Prediction (CIIP) mode, Local Lighting Compensation (LIC) mode, Diffuse Filtering Mode, Bilateral Filtering Mode, Transformed Domain Filtering Mode, or Diffuse Filtering Mode, Bilingual Filtering Mode, and Transformed Domain Filtering Mode Including post-reconstruction filtering mode,
The method described in Clause 2.

4. The coding mode constraint of the first block is:
The first coding mode of the one or more coding modes of the first block is invalidated if the first flag corresponding to the first coding mode is derived to be false. Including that
The method described in any one of Articles 2 to 3.

5. The coding mode constraint of the first block is:
The information indicating whether or not the second coding mode of the one or more coding modes in the first block is enabled is false in the second flag corresponding to the second coding mode. Including that no signal is transmitted if it is derived to be
The method described in any one of Articles 2 to 4.

6. The coding mode constraint of the first block is:
Information indicating whether or not the third coding mode of the one or more coding modes of the first block is enabled is false in the third flag corresponding to the third coding mode. When derived to be, it is signaled and constrained to be false in the conformance bitstream.
The method described in any one of Articles 2 to 5.

7. Further including a step of determining whether to apply the coding mode constraint of the first block.
The method described in any one of Articles 1 to 6.

8. Determining whether to apply the coding mode constraint of the first block:
Including determining whether to apply the coding mode constraint of the first block according to the row or column coding mode in the vicinity of the first block.
The method described in Clause 7.

9. Determining whether to apply the coding mode constraint of the first block:
Determines whether to apply the coding mode constraint of the first block based on at least N samples not coded in the fourth mode in the rows or columns in the vicinity of the first block. Including doing,
The method described in Clause 7.

10. The method of clause 8 or 9, wherein the neighborhood row includes a row above the first block and a row in the upper right corner.

11. The method of any one of clauses 8-10, wherein the neighborhood column comprises a left column, a lower left column, and an upper left corner column of the first block.

12. Determining whether to apply the coding mode constraint of the first block:
If any non-adjacent block in the vicinity is coded in the fourth mode, it involves determining that the coding mode constraint of the first block applies.
The method according to any one of Articles 7 to 11.

13. Determining whether to apply the coding mode constraint of the first block:
If all of the non-adjacent blocks in the neighborhood are coded in the fourth mode, it involves determining that the coding mode constraint of the first block applies.
The method according to any one of Articles 7 to 11.

14. Determining whether to apply the coding mode constraint of the first block:
If at least M non-adjacent blocks in the vicinity are not coded in the fourth mode and M is a predefined first threshold, it is determined that the coding mode constraint of the first block does not apply. Including doing,
The method according to any one of Articles 7 to 11.

15. Of clauses 9-14, the fourth coding mode comprises at least one of an intra-prediction mode, a combined inter-intra-intra-prediction (CIIP) mode, and a current picture referencing (CPR) mode. The method described in any one of the items.

16. Determining whether to apply the coding mode constraint of the first block:
Including determining whether to apply the coding mode constraint of the first block according to the position of the first block.
The method described in any one of Articles 7 to 14.

17. Determining whether to apply the coding mode constraint of the first block according to the position of the first block:
If the location is at the top of the current coding tree unit (CTU) and the neighboring block above and the first block belong to different CTUs, then the coding mode constraint of the first block does not apply. It is determined,
The method described in clause 16.

18. Determining whether to apply the coding mode constraint of the first block according to the position of the first block is:
If the position is to the left of the current coding tree unit (CTU) and the left neighborhood block and the first block belong to different CTUs, it is determined that the coding mode constraint of the first block does not apply. including,
The method described in clause 16.

19.
Currently determining the characteristics of the block;
Further comprising determining to apply the coding mode constraint of the current block if the characteristics of the current block meet the predefined conditions.
The method described in any one of Articles 1 to 18.

20.
Determine the characteristics of the current block and / or the previously coded block;
Further comprising determining whether to apply the coding mode constraint of the current block depending on the characteristics of the current block and / or the previously coded block.
The method described in any one of Articles 1 to 18.

21. A way to process video:
The conversion between the current block and the bitstream representation of the current block in the video data is compound inter-intra-prediction (CIIP), diffusion filtering, bilateral filtering, transformation domain filtering, or diffusion filtering, bilateral filtering, and transformation. Including performing using at least one of the different types of post-reconstruction filtering that is different from domain filtering,
The reconstructed neighborhood sample of the current block used in at least one of compound inter-intra-prediction (CIIP), diffusion filtering, bilateral filtering, transformation domain filtering, or the other type of post-reconstruction filtering is , Replaced by an approximate sample generated from the corresponding sample of the reconstructed neighborhood sample in one or more reference pictures.
Method.

22. 21. The method of clause 21, wherein when the current block is bi-predicted, the neighboring samples of the two reference blocks of the current block are averaged to produce the approximate sample.

23. 21. The method of clause 21, wherein when the current block is bi-predicted or uni-predicted, a neighbor sample of one reference block of the current block is used as the approximation sample.

24. Two references to the current block if the current block is bi-predicted and at least one of unequal weight generalization bi-prediction (GBI), weighted prediction and local lighting compensation (LIC) is applied to the current block. The method of clause 21, wherein the block neighborhood samples are weighted and averaged.

25. The method according to any one of clauses 22 to 24, wherein the neighborhood sample of the reference block of the current block is identified by the motion information of the current block.

26. The method of any one of clauses 22-24, wherein the neighborhood sample of the reference block of the current block is identified by the modified motion information of the current block.

27. The method of clause 24, wherein the modified motion information of the current block is a modified motion vector rounded to integer precision.

28. 21. The method of clause 21, wherein the neighborhood sample of the reference block of the current block is identified by a motion vector with integer pixel accuracy.

29. 28. The method of clause 28, wherein the motion vector of the reconstructed neighborhood sample is rounded to integer precision to identify the corresponding sample.

30. The method of any one of clauses 21-29, wherein the reconstructed neighborhood sample of the current block is replaced by the approximate sample only if the current block is coded with integer pixel precision.

31. The method of any one of clauses 21-30, wherein the reconstructed neighborhood sample of the current block is replaced by the approximation sample for only the luma or chroma component.

32.
Currently determining the characteristics of the block;
Further comprising deciding to replace the reconstructed neighborhood sample with the approximate sample if the feature of the current block meets a predefined condition.
The method according to any one of Articles 19 to 31.

33. The method of clause 19 or 32, wherein said feature of the current block comprises at least one of block size, slice type, picture type, tile type, and motion information.

34. The method of clause 33, wherein the predefined condition is that the current block contains a sample that is not less than a predefined second threshold.

35. The method of clause 33, wherein the predefined condition is that the width and height of the current block is greater than the predefined third threshold.

36. 33. The method of clause 33, wherein the predefined condition is that the width and height of the current block is less than a predefined fourth threshold.

37. The predefined condition is that the width of the current block is less than the predefined fifth threshold and / or the height of the current block is less than the predefined sixth threshold. The method described in clause 33.

38. The predefined condition is in clause 33 that the width of the current block is greater than the predefined seventh threshold and / or the height of the current block is greater than the predefined eighth threshold. The method described.

39. Apply the coding mode constraint for the current block in a group of sequence parameter sets (SPS), view parameter sets (VPS), picture parameter sets (PPS), sequence headers, picture headers, slice headers, tile group headers, tiles, or CTUs. Information indicating whether to do so and / or whether to replace the reconstructed neighbor sample with an approximate sample is signaled.
The method described in clause 38.

40.
Determine the characteristics of the current block and / or the previously coded block;
Further including deciding whether to replace the reconstructed neighborhood sample with an approximate sample, depending on the characteristics of the current block and / or the previously coded block.
The method according to any one of Articles 21 to 39.

41. The features of the current block and / or the previously coded block are the block dimensions of the current block and / or the previously coded block, video processing data unit (VPDU), picture type, low latency check flag, coding information. The method according to clause 20 or 40, comprising at least one of.

42. 31. The method of clause 41, wherein the coding information of the current block and / or the previously coded block indicates a reference picture, a simple prediction, or a bi-prediction.

43. A video processing device comprising a processor configured to implement the method according to any one of clauses 1-42.

44. The device according to clause 43, wherein the device is a video encoder.

45. The device according to clause 43, wherein the device is a video decoder.

46. A computer-readable recording medium on which a program containing the code is recorded, wherein the program is for the processor to perform the method according to any one of clauses 1-42. ..

Various other embodiments and additional features of these methods can be described using the clause-based description below.
[Clause 1]
A way to process video:
With the step of storing the composite inter-intra prediction (CIIP) flag of CIIP mode and / or the intra mode with motion information in a history-based motion vector prediction (HMVP) table;
It comprises performing a conversion between the current block in the video data and the bitstream representation of the current block, at least based on the HMVP table.
Method.
[Clause 2]
It also includes a step to compare the movement information of the two candidates.
The CIIP flag and / or the intramode of the CIIP mode is not used in the comparison of the motion information of the two candidates.
The method described in Clause 1.
[Clause 3]
It also includes a step to compare the movement information of the two candidates.
The CIIP flag and / or the intramode of the CIIP mode is used in the comparison of the motion information of the two candidates.
The method described in Clause 1.
[Clause 4]
The steps to be performed are:
If the merge candidate is from an entry in the HMVP table, this includes copying the CIIP flag of the entry to the merge candidate.
The method described in any one of Articles 1 to 3.
[Clause 5]
The steps to be performed are:
If the merge candidate is from an entry in the HMVP table, this includes copying the CIIP flag and the intramode of the entry to the merge candidate.
The method described in any one of Articles 1 to 3.
[Clause 6]
A way to process video:
With the step of determining the prediction mode of the current block during the conversion between the current block in the video data and the bitstream representation of the current block;
Combined Inter-Intra Prediction (CIIP) mode, which indicates that CIIP mode is enabled for the current block in response to the determination that the current block is coded in Advanced Motion Vector Prediction (AMVP) mode or Merge mode. Steps to determine the applicability of;
Including the step of performing the transformation based on the applicability of the CIIP mode.
Method.
[Clause 7]
The method of clause 6, wherein the CIIP flag and / or intramode of the CIIP mode is signaled when the block is currently coded in said AMVP mode.
[Clause 8]
In the current block merge mode and / or in merge (MMVD) mode with motion vector differences, the CIIP flag and / or intramode of the CIIP mode of adjacent and / or non-adjacent neighboring blocks is inherited. The method according to 6 or 7.
[Clause 9]
The CIIP flag and / or intramode of CIIP mode for non-skip merge mode and / or non-skip MMVD mode is signaled, and adjacent neighborhood blocks and / or in skip merge mode and / or skip MMVD mode. The method of any one of clauses 6-8, wherein the CIIP flag and / or intramode of the CIIP mode of non-adjacent neighboring blocks is inherited.
[Clause 10]
The CIIP flag and / or intramode of CIIP mode for skip merge mode and / or skip MMVD mode is signaled, and adjacent neighborhood blocks and / or in nonskip merge mode and / or nonskip MMVD mode. The method of any one of clauses 6-8, wherein the CIIP flag and / or intramode of the CIIP mode of non-adjacent neighboring blocks is inherited.
[Clause 11]
The method of any one of clauses 6-10, wherein the CIIP flag is inherited in merge mode and / or skip MMVD mode.
[Clause 12]
12. The method of clause 11, wherein if the CIIP flag is true and more than one intramode is allowed in the CIIP mode, the intramode is signaled.
[Article 13]
Any of clauses 6-12, where the CIIP flag and / or intramode of the CIIP mode of adjacent and / or non-adjacent neighboring blocks coded in merge mode and / or MMVD mode is inherited by the current block. The method described in item 1.
[Clause 14]
Further including disabling CIIP mode for skip mode,
The method described in any one of Articles 6 to 13.
[Clause 15]
It also includes a step to compare the information of the two merge candidates.
The CIIP flag and / or the intramode of the CIIP mode is not used in the comparison of the information of the two merge candidates.
The method described in any one of Articles 6 to 13.
[Clause 16]
It also includes a step to compare the information of the two merge candidates.
The CIIP flag and / or the intramode of the CIIP mode is used in the comparison of the information of the two merge candidates.
The method described in any one of Articles 6 to 13.
[Article 17]
A way to process video:
With the step of determining the type of selected merge candidate for the current block during the conversion between the current block in the video data and the bitstream representation of the current block;
Including the step of determining the applicability of compound inter-intra prediction (CIIP) for the current block, depending on the type of merge candidate selected above.
Currently the block is coded in merge mode,
Method.
[Article 18]
The CIIP is invalid if the selected merge candidate for the current block is at least one of a per-pair merge candidate, a compound dual merge candidate, a zero motion merge candidate, or a virtual or artificial merge candidate. The method described in clause 17.
[Clause 19]
If one of the candidates involved in the per-pair merge candidate or the compound dual merge candidate adopts the CIIP mode, the CIIP mode is enabled for the per-pair merge candidate or the compound dual merge candidate, clause 17 or 18. The method according to 18.
[Clause 20]
The method according to clause 19, wherein the intramode of the CIIP mode candidate involved is inherited.
[Clause 21]
The method of clause 19, wherein the intramode of the CIIP mode involved is signaled for the pair-by-pair merge candidate or the compound dual merge candidate.
[Clause 22]
If both candidates involved in a per-pair merge candidate or a compound dual merge candidate adopt the CIIP mode, the CIIP mode is enabled for the per-pair merge candidate or the compound dual merge candidate, clause 17 or. 18. The method according to 18.
[Clause 23]
The method of clause 22, wherein the intramode of one of the two involved candidates is inherited.
[Clause 24]
28. The method of clause 22, wherein the intramode of the CIIP mode involved is derived from the intramodes of the two related candidates and is used for the pair-by-pair merge candidate or the compound dual merge candidate.
[Clause 25]
22. The method of clause 22, wherein the intramode of the CIIP mode involved is signaled for the pair-by-pair merge candidate or the compound dual merge candidate.
[Clause 26]
A video processing device comprising a processor configured to implement the method according to any one of clauses 1-25.
[Clause 27]
The device according to clause 26, wherein the device is a video encoder.
[Clause 28]
The device according to clause 26, wherein the device is a video decoder.
[Clause 29]
A computer-readable recording medium on which a program containing code is recorded, wherein the program is for the processor to perform the method according to any one of clauses 1-25.

Various other embodiments and additional features of these methods can be described using the clause-based description below.
[Clause 1]
A way to process video:
Referencing the CIIP flag of one or more neighboring video blocks for the current video block during conversion between the current video block in the video data and the bitstream representation of the current video block. With the step of coding the Composite Inter-Intra Prediction (CIIP) flag for the current video block, without coding based on the context model,
At a minimum, including the step of performing the conversion by applying the compound inter-intra-prediction (CIIP) flag of the current video block.
Method.
[Clause 2]
The method of clause 1, wherein a fixed context is used in the coding of the current video block based on the context model of the CIIP flag.
[Clause 3]
A way to process video:
Coding the Composite Inter-Intra Prediction (CIIP) flag for the current video block by bypass coding during the conversion between the current video block in the video data and the bitstream representation of the current video block. Steps to do and
Including the step of performing the transformation, at least by applying the Composite Inter-Intra Prediction (CIIP) flag.
Method.
[Clause 4]
Coding a compound inter-intra-prediction (CIIP) flag for the current video block by bypass coding involves coding the CIIP flag with a probability equal to 0 and a probability equal to 1. The method described in.
[Clause 5]
Further including setting the intra-prediction mode associated with the current video block to a given intra-prediction mode in response to the current video block being coded in CIIP mode.
The method described in any one of Articles 1 to 4.
[Clause 6]
The method of clause 5, wherein the given intra-prediction mode is a planar mode.
[Clause 7]
The method of clause 5 or 6, wherein the given intra-prediction mode is used for the intra-prediction mode determination process of a subsequent coding video block.
[Clause 8]
If the current video block is a video block in the vicinity of the second video block during the conversion of the second video block, which is one of the subsequent coding video blocks of the current video block. Further includes that the given intra-prediction mode is added to the intra-prediction mode candidate list of the second video block.
The method described in Clause 7.
[Clause 9]
The intra-prediction mode candidate list further includes including the most probable mode candidate list.
The method described in Clause 8.
[Clause 10]
The method of any one of clauses 1-9, wherein the intra-predictive mode information of the current video block is not stored in response to the current video block being coded in CIIP mode.
[Clause 11]
The method of clause 10, wherein the current video block is considered unusable in the decoding process of other blocks.
[Clause 12]
The method of clause 10, wherein the current video block is considered to be a video block coded in interpredictive mode.
[Article 13]
The method of any one of clauses 1-12, wherein the CIIP flag of the current video block is not remembered.
[Clause 14]
A way to process video:
The step of determining the intra-prediction mode of the first video block of a video according to a rule, said rule:
Skipping checking the compound inter-intra-intra-prediction (CIIP) flag of one or more neighboring video blocks for the first video block during the process of deriving the intra-prediction mode of the first video block. Including steps and;
A step of performing a conversion between a first video block and a bitstream representation of the first video block, at least based on a determined intra-prediction mode.
Method.
[Clause 15]
Determining the intra prediction mode involves determining the most probable mode,
The method described in clause 14.
[Clause 16]
For the second video block of the video, the default intra-prediction mode is set to the second video block in response to the intercoding or CIIP coding of the second video block. The method according to clause 14 or 15, wherein the second video block is a nearby video block relative to the first video block.
[Article 17]
Determining the intra prediction mode is based on the default intra prediction mode of the second video block.
The default intra prediction mode is a planar intra prediction mode.
The method described in clause 16.
[Clause 18]
The method of any one of clauses 14-17, wherein the CIIP flag of one or more neighboring video blocks is not stored.
[Article 19]
A video processing device having a processor configured to implement the method according to any one of clauses 1-18.
[Clause 20]
The device according to clause 19, wherein the device is a video encoder.
[Clause 21]
19. The device according to clause 19, wherein the device is a video decoder.
[Clause 22]
A computer-readable recording medium on which a program containing a code is recorded, wherein the program is for the processor to perform the method according to any one of clauses 1 to 21.

Various other embodiments and additional features of these methods can be described using the clause-based description below.
[Clause 1]
A way to process video:
One or more for the current video block during conversion between the current video block coded in compound intra and inter-prediction (CIIP) mode of the video and the bitstream representation of the current video block. A weight pair containing a first weight for the first prediction result of the current video block and a second weight for the second prediction result of the current video block, based on the neighboring video blocks of. It is a step to decide
The first prediction result is generated by the intra prediction mode, and the second prediction result is generated by the inter prediction mode.
Including a step of determining the prediction result of the current block based on the weighted sum of the first prediction result and the second prediction result.
Method.
[Clause 2]
Determining the weight pair is:
Includes determining the weight pair according to the prediction mode of said one or more neighboring video blocks of the current video block.
The method described in Clause 1.
[Clause 3]
The method of clause 1 or 2, wherein for a neighborhood video block coded in CIIP mode, the neighborhood video block is treated as a block coded in interpredictive mode.
[Clause 4]
The method of clause 1 or 2, wherein for a neighborhood video block coded in CIIP mode, the neighborhood video block is treated as a block coded in intra prediction mode.
[Clause 5]
The method according to any one of clauses 1 to 4, wherein the neighborhood video block is at least one of adjacent blocks, non-adjacent blocks, and temporal neighborhood blocks of the current video block. ..
[Clause 6]
The method of any one of clauses 1-5, wherein one of the neighborhood video blocks is the neighborhood video block on the left or the neighborhood video block above.
[Clause 7]
The method of any one of clauses 1-5, wherein two of the neighborhood video blocks are the left neighborhood video block and the top neighborhood video block, respectively.
[Clause 8]
The left neighborhood video block covers the position (xCb-1, yCb + cbHeight-1), the top neighborhood video block covers the position (xCb + cbWidth-1, yCb-1), and (xCb, yCb) is. The method of clause 6 or 7, wherein the sample position in the upper left corner of the current video block, where cbWidth and cbHeight are the width and height of the current video block, respectively.
[Clause 9]
The left neighborhood video block covers the position (xCb-1, yCb), the top neighborhood video block covers the position (xCb, yCb-1), and (xCb, yCb) is the current video. The method according to clause 6 or 7, which is the location of the sample in the upper left corner of the block.
[Clause 10]
The CIIP mode result is obtained by applying the weight pair to the intra prediction result and the inter prediction result.
P _CIIP = (wInter * P _inter ＋ wIntra * P _intra ＋ offset) >> N
Obtained as
Here, P _CIIP is the CIIP mode result, P _inter is the inter prediction result, P _intra is the intra prediction result, (wInter, wIntra) is the weight pair, and offset is (1 <. <(N-1)) or set to 0, N is set to log2 (wIntra + wInter),
The method described in any one of Articles 1 to 9.
[Clause 11]
The method according to any one of Clause 1 to Clause 10, wherein the neighborhood video block is selected based on the coding information of the current block and / or the neighborhood block.
[Clause 12]
The method according to any one of clauses 1 to 11, wherein the weight pair is determined based on the coding information of the neighborhood block and / or the neighborhood block.
[Article 13]
The coding information is: block dimension information, block shape information, low delay check flag, reference picture information, motion information, or intra-prediction of the neighborhood video block of the current video block and / or the neighborhood video block. The method of clause 11 or 12, comprising at least one of the modes.
[Clause 14]
The method of clause 1, wherein the weight pair depends on the color component of the current video block.
[Clause 15]
The method of clause 14, wherein the weight pair of principal color components is different from weight pairs of other color components.
[Clause 16]
The method of clause 14, wherein the weight pair for the luma component is different from the weight pair for the chroma component.
[Article 17]
The method according to any one of Articles 14 to 16, wherein the first weight and the second weight are equal for the chroma component.
[Article 18]
The method according to any one of clauses 1 to 17, wherein the weight pair is one of two candidate weight pairs including a first candidate weight pair and a second candidate weight pair.
[Clause 19]
The two candidate weight pairs are {(1,3) and (3,1)}, {(1,3) and (2,2)}, {(3,1) and (2,2)}, {(3,5) and (4,4)}, {(5,3) and (4,4)}, {(1,7) and (4,4)} and {(7,1) and ( 4,4)}, one of the methods described in Clause 18.
[Clause 20]
19. The method of clause 18 or 19, wherein the weight pair is determined based on a single neighborhood video block of the current video block.
[Clause 21]
If the single neighborhood video block is coded in the intra prediction mode, the weight pair is the first candidate weight pair and the single neighborhood video block is coded in the inter prediction mode. If the weight pair is the second candidate weight pair, the method of clause 20.
[Clause 22]
28. The method of clause 20 or 21, wherein the single neighborhood video block is the left neighborhood video block or the top neighborhood video block.
[Clause 23]
The weight pair is determined based on two or more neighboring video blocks of the current video block, and at least one of the two or more neighboring video blocks satisfies a predetermined condition, clause 18. Or the method according to 19.
[Clause 24]
If at least one of the two or more neighborhood video blocks is coded in the intra prediction mode, the weight pair is the first candidate weight pair and of the two or more neighborhood video blocks. 23. The method of clause 23, wherein the weight pair is the second candidate weight pair when at least one of the above is coded in the interpredictive mode.
[Clause 25]
The weight pair is determined based on two or more neighboring video blocks of the current video block, each of the two or more neighboring video blocks satisfying a predetermined condition, as described in clause 18 or 19. the method of.
[Clause 26]
If all of the two or more neighborhood video blocks are coded in the intra prediction mode, the weight pair is the first candidate weight pair and all of the two or more neighborhood video blocks are the inter. 25. The method of clause 25, wherein the weight pair is the second candidate weight pair when coded in predictive mode.
[Clause 27]
Clause 23, where one of the two or more neighborhood video blocks is the left neighborhood video block and the other one of the two or more neighborhood video blocks is the top neighborhood video block. The method according to any one of 26.
[Clause 28]
One of clauses 23 to 27, wherein the first candidate weight pair is either (2,2) or (3,1) and the second candidate weight pair is (1,3). The method described in the section.
[Clause 29]
The method according to any one of clauses 23 to 28, wherein at least one weight value in the first candidate weight pair is different from at least one weight value in the second candidate weight pair.
[Clause 30]
The method according to any one of clauses 23 to 29, wherein the first candidate weight pair is different from the second candidate weight pair.
[Article 31]
The method according to any one of clauses 1 to 17, wherein the weight pair is exactly one candidate weight pair.
[Article 32]
A clause in which the one candidate weight pair is one of (1,3), (2,2), (1,7), (2,6), (3,5) and (4,4). 31.
[Clause 33]
The method according to any one of clauses 1 to 32, wherein the weight pair is determined based on whether one or more neighboring video blocks are coded in said CIIP mode.
[Clause 34]
35. The method of clause 33, wherein the weight pair is determined from the weight pair for a previously coded video block.
[Clause 35]
The method of clause 33 or 34, wherein the weight pair for the current video block is determined based on a single neighboring video block of the current video block.
[Clause 36]
When the single neighborhood video block is coded in CIIP mode, the weight pair for the current video block is set to the weight pair for the single neighborhood video block.
When the single neighborhood video block is coded in intra-predictive mode, the weight pair is the first of two candidate weight pairs, including a first candidate weight pair and a second candidate weight pair. Candidate weight pair,
When the single neighborhood video block is coded in interpredicted mode, the weight pair is the first of two candidate weight pairs, including the first candidate weight pair and the second candidate weight pair. 2 candidate weight pairs,
The method described in clause 35.
[Clause 37]
When the single neighborhood video block is coded in CIIP mode, the weight pair for the current video block is set to the weight pair for the single neighborhood video block.
When the single neighborhood video block is coded in intra-predictive mode or inter-predictive mode, the weight pair is one of two candidate weight pairs, including a first candidate weight pair and a second candidate weight pair. The first candidate weight pair,
The method described in clause 35.
[Clause 38]
The weight pair is determined based on two or more neighboring video blocks of the current video block, and at least one of the two or more neighboring video blocks satisfies a predetermined condition, clause 33. Or the method according to 34.
[Clause 39]
28. The method of clause 38, wherein the two or more neighborhood video blocks are checked in a predetermined check order to identify whether the two or more neighborhood video blocks are coded in CIIP mode. ..
[Clause 40]
When at least one of the two or more neighboring video blocks is coded in CIIP mode, the weight pair for the current video block shall be coded in CIIP mode in the given check order. Is set in the weight pair for the first identified neighborhood video block,
If neither of the two or more neighboring video blocks is coded in CIIP mode and at least one of the two or more neighboring video blocks is coded in intra-predictive mode, the weight pair is first. Is the first candidate weight pair of the two candidate weight pairs, including the candidate weight pair and the second candidate weight pair.
If neither of the two or more neighboring video blocks is coded in CIIP mode and at least one of the two or more neighboring video blocks is coded in interpredictive mode, the weight pair is said to be the first. It is the second candidate weight pair of the two candidate weight pairs including the one candidate weight pair and the second candidate weight pair.
The method described in clause 39.
[Clause 41]
If at least one of the two or more neighboring video blocks is coded in CIIP mode, the weight pair for the current video block may be coded in CIIP mode in the given check order. Set to a weight pair for the first identified neighborhood video block,
If none of the two or more neighboring video blocks are coded in CIIP mode, the weight pair is the second of two candidate weight pairs, including a first candidate weight pair and a second candidate weight pair. Candidate weight pair,
The method described in clause 39.
[Clause 42]
Any of clauses 36 to 41, wherein the first candidate weight pair is one of (2,2) and (3,1) and the second candidate weight pair is (1,3). The method described in Crab.
[Clause 43]
The method according to any one of clauses 36 to 42, wherein at least one weight value in the first candidate weight pair is different from at least one weight value in the second candidate weight pair.
[Clause 44]
The method according to any one of clauses 36 to 41, wherein the second candidate weight pair is any of (2,2) and (1,3).
[Clause 45]
The method according to any one of clauses 1 to 44, wherein the weight pair is signaled.
[Clause 46]
The method of any one of clauses 1 to 45, wherein the weight pair is determined based on whether the neighboring video block is coded in interpredictive mode.
[Clause 47]
The method of any one of clauses 1 to 46, wherein if a neighborhood video block is not coded in intra-predictive mode, the neighborhood video block is treated as a block coded in intra-predictive mode.
[Clause 48]
The neighborhood video block is not coded in at least one of intra block copy (IBC) mode, CIIP mode, triangular prediction mode (TPM), palette mode, RDPCM mode, clause 47. The method described in.
[Clause 49]
28. The method of clause 48, wherein the neighborhood video block is coded without applying transformation and / or quantization.
[Clause 50]
If the neighborhood video block is coded in at least one of in-block copy (IBC) mode, CIIP mode, triangle prediction mode (TPM), palette mode, and RDPCM mode, the neighborhood video block is non-blocking. The method according to any one of clauses 1 to 46, which is treated as a block coded in the intra prediction mode.
[Clause 51]
If the neighborhood video block is coded in intra-prediction mode but not in a given prediction mode, then the neighborhood video block is treated as a block coded in non-intra-prediction mode, of clauses 1-46 . The method described in any one of the items.
[Clause 52]
If the neighborhood video block is coded in a matrix-based intra-prediction mode and / or a multi-reference line intra-prediction mode and / or BDPCM mode with a reference line index that is not equal to 0, the neighborhood video block is non-existent. The method according to any one of clauses 1 to 46, which is treated as a block coded in intra-predictive mode.
[Article 53]
A video processing device comprising a processor configured to implement the method according to any one of clauses 1-52.
[Article 54]
The device according to clause 53, wherein the device is a video encoder.
[Clause 55]
The device according to clause 53, wherein the device is a video decoder.
[Article 56]
A computer-readable recording medium on which a program containing the code is recorded, wherein the program is for the processor to perform the method according to any one of clauses 1 to 52. ..

It will be appreciated that the disclosed techniques can be embodied in video encoders or decoders to improve compression efficiency using hash-based motion estimation.

The disclosed and other solutions, examples, embodiments, modules, and functional operations described in this article are in digital electronic circuits or in computers including the structures and structural equivalents thereof disclosed in this article. It can be implemented in software, firmware, or hardware, or in one or more combinations thereof. The disclosed and other embodiments are encoded on a computer-readable medium for execution by one or more computer program products, ie, data processing equipment, or to control the operation of the data processing equipment. It can be implemented as one or more modules of computer program instructions. The computer-readable medium can be a machine-readable storage device, a machine-readable storage board, a memory device, a composition of substances that realize a machine-readable propagating signal, or a combination thereof. The term "data processing device" includes any device, device, and machine for processing data, including, for example, a programmable processor, a computer, or a plurality of processors or computers. The device, in addition to the hardware, is the code that creates the execution environment for the computer program in question, such as the processor firmware, protocol stack, database management system, operating system, or a combination of one or more of them. Can be included. The propagated signal is an artificially generated signal, such as a machine-generated electrical signal, optical or electromagnetic signal, which is generated to code information for transmission to a suitable receiver device.

Computer programs (also known as programs, software, software applications, scripts, or code) can be written in any form of programming language, including languages that are compiled or interpreted, which are stand-alone programs. It can be deployed in any form, including as a module, component, subroutine, or other unit suitable for use in a computing environment. Computer programs do not necessarily correspond to files in the file system. The program has been tailored to a single file dedicated to the program in question, or to a portion of a file that holds other programs or data (eg, one or more scripts stored in a markup language document). It can be stored in a file (for example, a file that stores one or more modules, subprograms, or code parts). Computer programs can be deployed to run on one computer, or on multiple computers located at one site or distributed across multiple sites and interconnected by communication networks. ..

The processes and logical flows described in this article are run by one or more programmable processors that run one or more computer programs to act on the input data and perform functions by producing outputs. Can be done. Processes and logic flows can also be run by special purpose logic circuits such as FPGAs (Field Programmable Gate Arrays) or ASICs (Application Specific Integrated Circuits), where the device is implemented as such special purpose logic circuits. You can also.

Suitable processors for running computer programs include, for example, both general purpose and dedicated microprocessors, and any one or more processors of any type of digital computer. Generally, the processor receives instructions and data from read-only memory and / or random access memory. The essential elements of a computer are a processor for executing instructions and one or more memory devices for storing instructions and data. In general, a computer may also contain data on one or more mass storage devices for storing data, such as magnetic disks, optomagnetic disks, or optical disks, or from such mass storage devices. It may be operationalally coupled to receive the data and transfer the data to such mass storage, or both. However, the computer does not need to have such a device. Computer-readable media suitable for storing computer program instructions and data are semiconductor memory devices such as EPROMs, EEPROMs, and flash memory devices; magnetic disks such as internal hard disks or removable disks; optomagnetic disks; Also includes all forms of non-volatile memory, media and memory devices, including CD-ROM and DVD-ROM discs. Processors and memory can be supplemented by special purpose logic circuits or incorporated into special purpose logic circuits.

Although this patent document contains many details, they should not be construed as a limitation to the scope of any subject matter or claims, but rather specific embodiments of the particular technique. It should be interpreted as an explanation of features that may be unique to. Certain features described in the context of separate embodiments in this patent document can also be implemented in combination in a single embodiment. Conversely, the various features described in the context of a single embodiment can also be implemented separately in multiple embodiments or in any suitable subcombination. Further, the features are described above as acting in certain combinations, and may be claimed as such initially, but one or more from the combinations described in the claims. In some cases, the characteristics of the above may be cut out from such a combination, and the combination described in the claims may be directed to a sub-combination or a modification of the sub-combination.

Similarly, the drawings show the actions in a particular order, which means that such actions are performed in that particular order or sequentially in order to achieve the desired result, or show. It should not be understood as requiring that all the actions taken be performed. Moreover, the separation of the various system components in the embodiments described in this patent document should not be understood as requiring such separation in all embodiments.

Only a few implementations and examples have been described, and other implementations, improvements and modifications are possible based on what is described and described in this patent document.

Claims (20)

A way to process video:
Based on one or more neighboring video blocks to the current video block during the conversion of the video data between the current video block and the bitstream coded in composite intra and inter-predictive modes, A step in determining a weight pair that includes a first weight for the first prediction result of the current video block and a second weight for the second prediction result of the current video block.
The first prediction result is generated by the intra prediction mode, and the second prediction result is generated by the inter prediction mode.
With the step of determining the prediction result of the current block based on the weighted sum of the first prediction result and the second prediction result;
Including a step of performing the transformation based on the prediction result.
Method. The weight pair is:
Determined according to the prediction mode of said one or more neighboring video blocks of the current video block.
The method according to claim 1. 2. The prediction mode of the one or more neighborhood video blocks indicates whether the one or more neighborhood video blocks are coded in the intra prediction mode or the non-intra prediction mode. The method described in. If the neighborhood video block is coded in at least one of in-block copy mode, compound intra and interprediction mode, geometric division mode or palette mode, the neighborhood video block is non-intra. The method of any one of claims 1 to 3, which is treated as a block coded in predictive mode. 24 . Method. The method of any one of claims 1-5, wherein the weight pair is further determined according to the color component of the current video block. 6. The method of claim 6, wherein the location of the one or more neighboring video blocks is determined based on the color component of the current block. The method of any one of claims 1-7, wherein the one or more neighborhood video blocks include at least one of the left neighborhood video block or the top neighborhood video block. The weight pair is determined according to the availability of the one or more neighborhood video blocks, and the variable checkPredModeY, which specifies whether the availability of the neighborhood blocks depends on the prediction mode, is set to false. The method according to any one of claims 1 to 8. The video block in which the one or more neighboring video blocks cover the position (xCb-1, yCb-1 + (cbHeight << a)) and the position (xCb-1 + (cbWidth << a), yCb- Including the video block covering 1), (xCb, yCb) is the sample position in the upper left of the current video block, cbWidth and cbHeight are the width and height of the current video block, respectively. The method according to any one of claims 1 to 9, wherein a is determined using cIdx of the current block. 10. The method of claim 10, wherein a is set to 0 in response to cIdx of the current block being equal to 0. 10. The method of claim 10, wherein a is set to 1 in response to cIdx of the current block not equal to 0. The weight pair is determined based on two or more neighboring video blocks.
If all of the two or more neighboring video blocks are coded in intra-predictive mode, the weight pair is the first candidate weight pair.
If all of the two or more neighboring video blocks are coded in non-intra-predictive mode, the weight pair is a second candidate weight pair that is different from the first candidate weight pair.
Otherwise, the weight pair is a third candidate weight pair that is different from the first candidate weight pair and the second candidate weight pair.
The method according to any one of claims 1 to 12. The first candidate weight pair is (3,1), the second candidate weight is (1,3), the third candidate weight pair is (2,2), and (x, 13. The method of claim 13, wherein x is the first weight and y is the second weight for y). The prediction result is the weight pair for the intra prediction result and the inter prediction result.
P ＝ (wInter * P _inter ＋ wIntra * P _intra ＋ offset) >> N
Obtained by applying as
Here, P is the prediction result, P _inter is the first prediction result, P _intra is the second prediction result, (wInter, wIntra) is the weight pair, and offset is 2. Set to,
The method according to any one of claims 1 to 14. The method of any one of claims 1-15, wherein the conversion comprises encoding the current video block into the bitstream. The method of any one of claims 1-15, wherein the conversion comprises decoding the current video block from the bitstream. A device for processing video, having a processor and non-temporary memory with instructions, the instructions to the processor when executed by the processor:
Based on one or more neighboring video blocks to the current video block during the conversion of the video data between the current video block and the bitstream coded in composite intra and inter-predictive modes, A step in determining a weight pair that includes a first weight for the first prediction result of the current video block and a second weight for the second prediction result of the current video block.
The first prediction result is generated by the intra prediction mode, and the second prediction result is generated by the inter prediction mode.
With the step of determining the prediction result of the current block based on the weighted sum of the first prediction result and the second prediction result;
The step of executing the conversion is executed based on the prediction result.
Device. A non-temporary computer-readable storage medium that stores instructions that, when executed by the processor, tells the processor:
Based on one or more neighboring video blocks to the current video block during the conversion of the video data between the current video block and the bitstream coded in composite intra and inter-predictive modes, A step in determining a weight pair that includes a first weight for the first prediction result of the current video block and a second weight for the second prediction result of the current video block.
The first prediction result is generated by the intra prediction mode, and the second prediction result is generated by the inter prediction mode.
With the step of determining the prediction result of the current block based on the weighted sum of the first prediction result and the second prediction result;
The step of executing the conversion is executed based on the prediction result.
Storage medium. A non-temporary computer-readable recording medium that stores a bitstream of video data generated by a method performed by a video processor, wherein the method is:
For the current video block coded in compound intra and inter-prediction modes, the first prediction result for the current video block based on one or more neighboring video blocks for the current video block. A step in determining a weight pair that includes a weight of 1 and a second weight for the second prediction result of the current video block.
The first prediction result is generated by the intra prediction mode, and the second prediction result is generated by the inter prediction mode.
With the step of determining the prediction result of the current block based on the weighted sum of the first prediction result and the second prediction result;
Including a step of generating the bitstream based on the prediction result.
recoding media.