Use-case driven context model selection for hybrid video coding tools

Abstract

An apparatus is described that includes an encoder. The encoder receives image or video data, encodes the received image or video data and provides a bitstream representing the image or video data. The encoder includes a CABAC encoder. The CABAC encoder receives a binary valued syntax element associated with a block of the image or video data to be encoded, and encodes, using a selected context model, the binary valued syntax element into coded bits for the bitstream. The binary valued syntax element includes a tool flag indicating whether or not certain coding tool, like a linear weighted intra prediction, LWIP, is employed when encoding a block of the image or video data. For a block of the image or video data having an aspect ratio larger than 2 and to which the certain coding tool is applicable, a first context model for encoding the tool flag is selected from a group of one or more first context models, and for a block of the image or video data having an aspect ratio less than or equal to 2 and to which the certain coding tool is applicable, a second context model for encoding the tool flag is selected from a group of one or more second context models.

Description

(19) *EP004765824A2* (11) EP 4 765 824 A2 (12) EUROPEAN PATENT APPLICATION (43) Date of publication: 24.06.2026 Bulletin 2026 / 26 (21) Application number: 26169069.7 (22) Date of filing: 04.03.2020 (51) International Patent Classification (IPC): H04N 19 / 91 (2014.01) (52) Cooperative Patent Classification (CPC): H04N 19 / 13; H04N 19 / 159; H04N 19 / 176; H04N 19 / 91 (84) Designated Contracting States: AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR (30) Priority: 05.03.2019 EP 19160705 (62) Document number(s) of the earlier application(s) in accordance with Art. 76 EPC: 25163380.6 / 4 546 773 20707126.7 / 3 935 842 (71) Applicant: Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e. V. 80686 München (DE) (72) Inventors: • PFAFF, Jonathan 10587 Berlin (DE) • HELLE, Philipp 10587 Berlin (DE) • SCHÄFER, Michael 10587 Berlin (DE) • HINZ, Tobias 10587 Berlin (DE) • STALLENBERGER, Björn 10587 Berlin (DE) • MERKLE, Philipp 10587 Berlin (DE) • SCHWARZ, Heiko 10587 Berlin (DE) • MARPE, Detlev 10587 Berlin (DE) • WIEGAND, Thomas 10587 Berlin (DE) (74) Representative: Röthinger, Rainer Wuesthoff & Wuesthoff Patentanwälte und Rechtsanwalt PartG mbB Schweigerstrasse 2 81541 München (DE) Remarks: This application was filed on 30‑03‑2026 as a divisional application to the application mentioned under INID code 62. (54) USE‑CASE DRIVEN CONTEXT MODEL SELECTION FOR HYBRID VIDEO CODING TOOLS (57) An apparatus is described that includes an en- coder. The encoder receives image or video data, en- codes the received image or video data and provides a bitstream representing the image or video data. The encoder includes a CABAC encoder. The CABAC en- coder receives a binary valued syntax element asso- ciated with a block of the image or video data to be encoded, and encodes, using a selected context model, the binary valued syntax element into coded bits for the bitstream. The binary valued syntax element includes a tool flag indicating whether or not certain coding tool, like a linear weighted intra prediction, LWIP, is employed when encoding a block of the image or video data. For a block of the image or video data having an aspect ratio larger than 2 and to which the certain coding tool is applicable, a first context model for encoding the tool flag is selected from a group of one or more first context models, and for a block of the image or video data having an aspect ratio less than or equal to 2 and to which the certain coding tool is applicable, a second context model for encoding the tool flag is selected from a group of one or more second context models. EP 4 76 5 82 4 A 2 Processed by Luminess, 75001 PARIS (FR) Description FIELD OF THE INVENTION

[0001] The present invention relates to the field of encoding / decoding pictures, images or videos, more specifically, to the encoding of one or more coding tools, like an affine linear weighted intra prediction, LWIP, or matrix-based intra prediction, MIP, of the versatile video coding, VVC, standard, using context or context modes of a context adaptive binary arithmetic coding, CABAC, engine. Embodiments concern the encoding of a flag, like the intra_mip_flag, indicating the applicability of the LWIP or MIP of the VVC standard based on context models selected dependent on an aspect ratio of a block of the image or video data to be processed. BACKGROUND

[0002] In state-of-the-art video coding standards like ITU T H.265 I MPEG H HEVC [1], pictures are divided into CodingTreeUnits of fixed square size that can further be subpartitioned into smaller blocks. The reconstructed signal of such a block is usually a superposition of a prediction signal and a residual signal. The prediction signal is obtained either by extrapolation of samples of the adjacent neighborhood into the current block (intra prediction) or by copying a filtered or unfiltered representation of samples from one or two reference pictures (inter prediction). Reference pictures are pictures that have already been reconstructed from the bitstream and that are stored for reference in a picture buffer. The residual signal is obtained by inverse transforming dequantized transform coefficients that are read from the bitstream. After the block reconstructionprocess, in-loopfiltersareapplied toenhance thesignalsof the reconstructedblocksand toobtain the reconstructed picture.

[0003] The entropy decoding process of reading symbols like transform coefficients, deltaQP, intra prediction mode, motion vector differences etc. from the bitstream is done by a parser that uses a context adaptive binary arithmetic coding (CABAC) engine to convert bits read from the bitstream into binary decisions (bins). The parser converts or combines these bins into the symbols or syntax elements. The adaptivity of the entropy coding process is realized by using CABAC Contexts (CCs). Each context represents an adaptive probability model, whichmodels the entropy of a specific symbol or sets of symbols. The term adaptive indicates a permanent update of themodel towards the current coded state. Thus, the model adapts to local statistics of the corresponding symbols. The update step is usually embedded in the arithmetic coding operation. First, the current state of the CC is used to parameterize the arithmetic coding process. Then, once the decoded symbol is derived, it is used to update the CC with a given step size towards the current decoded probability.

[0004] In the JEM-Software [2] as well as in the upcoming VVC-Standard [3], various improvements concerning the arithmetic coding process have been evaluated and have been adopted. The arithmetic coding engine has been changed and also the initialization and update process of the CCs have been improved. The dynamic range of the probability representationaswell as thebehavior of theupdateprocessofCCshavebeen refined.EachCChasan individual twolevel update step that controls the adaption-strength of the CC towards the current probability. This improvement helps to customize the CC update process according to the expected CC usage statistics.

[0005] Due to the high number of binary decisions that are required to transmit syntax elements and because of the number of the syntax elements as such, binary decisions have to be grouped using the same CC in order to reach a practical number or amount ofCCs that canbehandled byadecoder. Furthermore, the groupinghelps theupdate process to exploit local statistics and it improves the stability of the underlying probability models.

[0006] The binary decisions with statistically similar probabilities that belong to the same syntax element are usually grouped into one CC. Exceptions of this kind of grouping are made in the case when a binary decision is likely to have different probabilities that can be predicted from already decoded symbols in the adjacent neighborhood. In this case, the chosenCC is predicted fromalready decoded symbols in the adjacent neighborhood. Such a procedure is usually applied to symbols that are transmitted in the bitstream rather frequently.

[0007] Besides the context controlled arithmetic coding, there is a bypass mode with a fixed probability of 0.5. This bypass mode, incorporated in the arithmetic coder, is a low-complexity mode for high throughput. The bypass mode is extensively used for transform coding for example. SUMMARY OF THE INVENTION

[0008] The evolution of video coding methods has shown an increasing variety of block shapes and an increasing number of coding tools, leading to a considerable amount of algorithmic complexity at the encoder to find a good coded representation.Thus, it canbebeneficial at theencoder toskipevaluationof (i.e. switchoff)acoding tool in certain contexts to achieve a better complexity versus compression efficiency trade off. The usage of a coding tool on a block is usually communicated to the decoder by submitting context modeled tool-enable-flags in the bitstream. 2 EP 4 765 824 A2 5 10 15 20 25 30 35 40 45 50 55 Problem

[0009] Ideally, adecoderhasaminimumofconstraintsonwhethera tool enableflag, i.e. a flag thatdetermineswhethera tool like e.g. a coding mode or a prediction mode is applied to a certain block or not, is sent in the bitstream or not. The reason is that disabling a tool for certain cases may deteriorate its impact on the compression performance in some scenarios, even if these scenarios are rather unlikely. In fact, one of the main reasons of the efficiency of hybrid video codecs is that a very large variety of competing coding tools are always possible and only one of these tools is selected in a given case.

[0010] For instance, a constraint on allowing a tool, and thus sending the tool enable flag, only for small block sizes, would potentially decrease the coding efficiency of future applicationswith very high resolutions that usually contain only a small portion of small blocks.

[0011] On the other hand, sending a tool enable flag for all possible cases is inefficient for application scenarioswhere a fastencoder searchstrategywouldnot test the tool for somecasesbecause for thesecases the tool isunlikely tobechosen either because it is too expensive in terms of run time or the effect of using the tool in these cases on the overall coding efficiency is rather small. In such a setting, not testing the tool for certain cases would lead to a faster encoder, but for the price of a somewhat decreased coding efficiency: A tool enable flag would be sent in the bitstream for a certain case, although the tool is never used for that case in the given scenario. Thus, in this scenario, the coding efficiency would be higher if the encoder search constraint would also be represented by a constraint on sending the tool enable flag in the bitstream.

[0012] Starting from a prior art as described above, there may be a need for improvements or enhancements for encoding of one or more coding tools used for encoding / decoding pictures, images or videos BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Embodiments of the present invention are described in further detail with reference to the accompanying drawings, in which: Fig. 1 illustrates an apparatus for encoding image or video data in accordance with embodiments of the present in- vention; Fig. 2 illustrates an apparatus for decoding encoded image or video data in accordance with embodiments of the present invention; Fig. 3 illustrates an apparatus for encoding image or video data in accordance with embodiments of the present in- vention introducing an individual, additional CABAC context for sending a flag for blocks with an aspect ratio larger than 2; Fig. 4 illustrates an apparatus for decoding image or video data in accordance with embodiments of the present in- vention introducing the individual, additional CABAC context for a flag for blocks with an aspect ratio larger than 2 and encoded using the apparatus of Fig. 3; and Fig. 5 illustrates an example of a computer system on which units or modules as well as the steps of the methods described in accordance with the inventive approach may execute. DETAILED DESCRIPTION OF THE INVENTION

[0014] Embodiments of the present invention are now described in more detail with reference to the accompanying drawings, in which the same or similar elements have the same reference signs assigned.

[0015] As mentioned above, in the design on how to group binary decisions into context models, only the aspect of lowering the overall entropy vs. the number of used context models has been taken into account in previous standards. In contrast to this approach, a new aspect for the design on how to group binary decisions into context models is raised in accordancewith the present invention. This approach takes into account the increasing complexity of algorithms as such. The present invention adapts the context modeling, e.g. by inserting contexts, in order to align coding contexts with contexts that are suited to switch off certain coding tools. This enables encoders to choose among operating points with different algorithmic complexity while avoiding deterioration of compression efficiency.

[0016] The inventive approach is explained using the following example. It is assumed that a probabilitymodeled binary decision switches between a first tool represented by algorithm 1 and a second tool represented by algorithm 2. Here, the second tool is thought to be a baseline tool while the first one is thought to be a more special tool. In accordance to this 3 EP 4 765 824 A2 5 10 15 20 25 30 35 40 45 50 55 assumption, the second tool is overall more likely to be chosen than the first one. As an example, it is assumed that the overall probability that the first tool is preferred over the second tool is 0.3.

[0017] Now it is assumed that twoapplication scenariosaregiven. In thefirst applicationA, thereareNcaseswhereboth tools are tested and the best performing one is chosen. In the secondapplicationB, for some reason, only for a determined portion of all N cases both tools are tested and for the remaining cases, only the baseline tool, i.e. the second tool, is chosen. For both applications, the decisions for all N cases have to be context modeled and to be transmitted in the bitstream. It is assumed, as an example, that the number of cases where both tools are tested in application B is equal to N / 2. For the other N / 2 cases, the tool is tested in application A but not tested in application B.

[0018] If the tool flag is context codedwith a singleCC, in the first applicationA, the probability of algorithm1has astable value of 0.3, whereas in the second application B, its average probability drops to 0.15, introducing a fixed probability penalty of ±0.15 for all N cases: In the cases, where the tool is tested in application B, the true probability would be 0.3 insteadof 0.15while in thecaseswhere the tool is not tested inapplicationB, the trueprobabilitywouldbe0. Inotherwords, in applicationB, usinga singleCC, theprobability of theactual decisionswouldbemodeled inanon-optimalway leading to higher bitrate when transmitted in the bitstream.

[0019] In the inventive approach, the disadvantage of such a non-optimal modeling of the probability of the decision is overcome as follows. Instead of using one probability model for all decisions, two (or more) probability models resp. CABAC contexts are assigned to the same decision. Which of the probability models is used for each of the N cases is selected with a selection index.

[0020] When considering again the above example, the selection index is chosen in such a way that it distinguishes between the determined portion where algorithm1 and algorithm 2 are tested and the remaining portion where only algorithm2 is tested in the second application B. In otherwords, the two different case are clustered, and represented by a different value of the selection index.

[0021] Whenusing this selection indexwith two probabilitymodels in applicationA, several different CCs can still model the statistics of the tool flag, although the selection index switchesbetween twoprobabilitymodels. In bothmodels, the tool represented by algorithm1 has a probability of 0.3. This leads to an equivalentmodeling as in the original case, whichwas modeled with only one probability model.

[0022] However, using the aforementioned selection index with two probability models in the second application B causes an optimal modeling for this case too. For the determined portion of all the N / 2 cases where both algorithms are tested, the probability of algorithm 1 is 0.3 and for the remaining portion that only tests the baseline algorithm 2, the probability of algorithm 1 is 0.0. Both probabilities are captured in a well distinguished way in the two probability models, thus yielding a modeling without any modelling penalty resulting in a low bitrate when transmitted in the bitstream.

[0023] Thus, embodiments of the inventive approach is based on introducing an additional CABAC context for sending the tool enable flag. This extra context is only applied if a condition is met. Otherwise, the CABAC context is selected as usual. In accordance with an embodiment, the condition could be that the size of current block belongs to a predefined subset of block sizes that might be skipped by a fast encoder search strategy, but that might be beneficial for applications that requireahighcodingefficiency. Inaccordancewithanother embodiment, the condition couldbe that theaspect ratio of a current block is above a certain value, like 2, that might be skipped by a fast encoder search strategy, but that might be beneficial for applications that require a high coding efficiency and thatmay be controlled on a block-level rather than on a picture-level or slice-level.

[0024] On the one hand, due to the probability adaptation of a CABAC context, if in an application scenario the tool is never tested for specific casesdefinedby the condition, the signaling overhead for sending the tool flag for these caseswill be very small. Thus, the coding efficiency will be almost as good as if the tool flag would not be sent for these cases.

[0025] On the other hand and again due to the probability adaptation of CABAC contexts, if in a different application scenario the tool is testedbyanencoderalso in thecasesdeterminedby thecondition, thecodingefficiency for sending the tool flag will not decrease significantly if a separate CABAC context is used for the cases determined by the condition.

[0026] Thus, in contrast to state of the art approaches, the assignment of different CABAC contexts as proposed by the present invention is not guided by attempting to model the overall conditional probability distribution of the tool flag. Instead, as explained by the above example, the assignment of different CABAC Contexts corresponds to different application scenariosof a tool.Here, eachapplication scenario is definedasaspecific conditionunderwhich theexecution of the given tool would in principle be possible but would never be tested by the encoder in the given scenario.

[0027] There can be different reasons for excluding tools resp. algorithms under specific conditions, and below some embodiments of the cases determined by the conditions are given but the present invention is not limited to these embodiments. First, the excluded algorithm may be too complex for these cases. Second, the algorithm may not be implemented or even not be implementable for these cases for example due to hardware or resource limitations. Third, there are scenarios in which the algorithm used in these cases may only marginally improve the compression perfor- mance. Fourth, using the underlying algorithm in these cases basically always gives very limited compression benefit and is thus feasible only if maximal compression performance is targeted. Fifth, these cases do not comprise the core application field the algorithm resp. the tool was initially designed for. 4 EP 4 765 824 A2 5 10 15 20 25 30 35 40 45 50 55 More than one context split for a tool

[0028] Embodiments of present inventionalso incorporate the splitting of a context into avariety of contexts,whereeach of the additional contexts corresponds to different use case scenarios for the underlying tool. An embodiment for this specification could be as follows. An original or convention tool enable flag is modeled by a single context. In accordance with embodiments, the inventive approach uses three contexts insteadwhere the selection index is controlled for example by quantized versions of the block area. Here, the selection index might be assigned as follows:

[0029] In accordance with other embodiments, the inventive approach uses four contexts or context models for the binarization of a syntax element intra_mip_flag representing an affine linear weighted intra prediction, LWIP, or matrix- based intra prediction, MIP, of the versatile video coding, VVC, standard. The selection index is controlled by the aspect ratio (width / height or height / width) of a current block:

[0030] Fig. 1 illustrates an apparatus 100 for encoding image or video data in accordance with embodiments of the present invention. The apparatus 100 includes an encoder 102. The encoder 102 receives image or video data 104 and encodes the received image or video data 104 so as to provide a bitstream 106 representing the encoded image or video data. Theencoder102 includesaCABACencoder108.TheCABACencoder108 receivesabinary valuedsyntaxelement 110 associated with a certain data block of the image or video data to be encoded and encodes, using a selected context model, the binary valuedsyntax element into codedbits 112 for thebitstream.Thebinary valued syntaxelement includesa tool flag indicatingwhether or not a certain coding tool is employedwhen encoding the image or video data. A group of first context models for encoding the tool flag is selected for one or more first parts of the certain data block to which, independent of an application, the coding tool is alwaysapplicable. Agroupof second contextmodels for encoding the tool flag is selected for one or more second parts of the certain data block to which, dependent on the application, the coding tool is applicable or is not applicable. In accordancewith embodiments, as also described below, theCABACencoder 108 selects the first context model or the second context model for a currently processed part of the certain data block responsive to a selection index. The selection index has a first value indicating that the currently processed part of the certain data block is the first part, and the selection index has a second value indicating that the currently processed part of the certain data block is the second part.

[0031] Fig. 2 illustrates anapparatus 200 for decoding encoded imageor videodata in accordancewith embodiments of the present invention. The apparatus 200 includes a decoder 202. The decoder 202 receives a bitstream 106, like the bitstream provided by the encoder 102 in Fig. 1. The bitstream 106 includes the encoded image or video data, and the decoder 202 decodes from the received bitstream the encoded image or video data and provides the decoded image or video data 204. The decoder includes a CABAC decoder 206 which decodes from the bitstream 106, using a selected context model, the binary valued syntax element 110 associated with a certain data block of the encoded image or video data.Thebinaryvaluedsyntaxelement includesa tool flag indicatingwhetherornotacertain coding tool isemployedwhen encoding the image or videodata. A groupof first contextmodel for decoding the tool flag is selected for parts of the certain data block to which, independent of an application, the coding tool is always applicable, and a group of second context model for decoding the tool flag is selected for parts of the certain data block to which, dependent on an application, the coding tool is applicable or is not applicable. In accordance with embodiments, as also described below, the CABAC decoder 206 selects the first context model or the second context model for a currently processed part of the certain data block responsive toaselection index.Theselection indexhasafirst value indicating that thecurrentlyprocessedpart of the certain data block is the first part, and the selection index has a second value indicating that the currently processed part of the certain data block is the second part.

[0032] In accordancewith embodiments, the groupof first contextmodels includes one first contextmodel or a variety of first context models, and the group of second context models includes one second context model or a variety of second 5 EP 4 765 824 A2 5 10 15 20 25 30 35 40 45 50 55 context model. Combination with original context indexing

[0033] As mentioned before, in many cases tool enable flags are modeled by more than one context model when a binary decision is likely to have different probabilities that can be predicted from already decoded symbols in the adjacent neighborhood. The inventive approach can also be applied as a combination of such an entropy driven context selection and the inventive selection index. The motivation for such a combination is obvious since the original entropy driven context indexing might also be suitable after a separation.

[0034] An embodiment where a context model for a tool flag is selected both by an entropy driven context selection and by the approach of the present invention is as follows.

[0035] Thus, in this embodiment, a purely entropy driven context selectionwould yield three possible contextmodels for a given tool flag while its combination with the context model selection of the present invention yields six possible context models that are indexed by the index Combinedlndex. Replacement of bypass mode coded enable flag

[0036] The inventiveapproach isalsoapplicable for flags thatareoriginally codedusing thebypassmode. Insuchacase either one or two additional context models are required to realize the use-case driven implementation. If only one additional context model is used, the selection index distinguishes between using the bypass mode for coding the not affected portion of cases and the one context model for all other cases, where the tool can now be switched on and off.

[0037] Inacasewhere twocontextmodelsareused, thebypassmode isentirely replacedbycontextmodeledarithmetic coding and the selection index would distinguish between the two context models. It should be mentioned that with the improved update techniques in the upcoming VVC-standard, the one context that models the not switchable portion of cases could be updated with a small update strength to achieve a quasi stationary model.

[0038] It shouldalsobementioned that theadditionalmodel thatmodelsa switchableportionof caseswould certainly be usedwith a very strong update-strength to achieve a fast adaption of the contextmodel either towards the tool on or tool off probability. Alternative way of signaling partial tool enabling in a parameter set

[0039] The aforementioned partial tool enabling behavior can also be signaled in a parameter set that is signaled in the bitstream for each predetermined portion of one or several frames, for example for each slice. In this case, context separation would not be required because the transmitted flags from the parameter set would contain all necessary information. However, a disadvantage of this kind of signaling in comparison to a use case driven context model selection is that in the formercase, throughout thewholeportionof thevideosequence forwhich theparameter setapplies, a tool can onlybeeither completelyenabledordisabled for thecasecorresponding to theapplicationscenariowhile in the latter case, it can also be disabled within any variable portion of the video sequence that does not need to be predetermined resp. known to thedecoder. The reason is that in the latter case, if a special contextmodel for a tool flag is assigned to all cases in which for a specific application scenario it is sometimes feasible to disable the tool, then, starting anywhere in the sequential encoding, an encoder corresponding to the specific application scenario could never test the tool for the corresponding cases until a flexible point in the sequential encodingwith onlyminor signalization overhead in comparison to the situationwhere the tool would be completely forbidden for all cases corresponding to the application scenario in this specific part of the video sequence. 6 EP 4 765 824 A2 5 10 15 20 25 30 35 40 45 50 55 Application with arbitrary coding tools

[0040] The inventive context splitting can be applied to any coding tool that is controlled by an enable flag. Current candidate tools occurring in the future VVC standard that could be used with the inventive approach are listed below. However, the application of the inventive approach is not limited to these tools. Candidate tools are DMVR, OBMC, BIO, FRUC, LIC, ISP, ALF, SAO, Inter- or intraMTS, 65Angular IntraModes,MRL and also partitioning tools likeQTBT,MTTor QTBT+TT.

[0041] The inventiveapproach canalsobeapplied to a tool that hasdifferent configurationswhichare representedbyan index value. Here, the CC assignment of the present invention would be determined by the fact that in some application scenarios, only a subset of all configurations of the tool are feasible either for specific cases or in general. The CC assignment of the present invention would take these different application scenarios into account by assigning extra CCs to the configuration in which the tool is unfeasible for a scenario. An embodiment for this aspect of the present invention wouldbe the tool of applyingoneout of n transforms for the inverse transformof prediction residuals,where the indexof the transform is transmitted in the bitstream.

[0042] With the inventive approach at hand, many of the tools could be provided with some context splitting enabling tools only under specific circumstances. Here, the context splitting according to the present invention is in accordance to specific cases under which a specific tool might be unfeasible in certain application scenarios. These specific cases depend on key properties of the tool. A list of the properties that could be evaluated or combined to specific circumstance that does not limit to use cases of this invention is: block size, block shape, block aspect ratio, temporal level, QP, picture type, picture resolution, dynamic range of the picture, reference picture, leading picture of a GOP.

[0043] The specific circumstance can also be a combination of those aforementioned properties. Embodiment of an application of the inventive approach of context assignment

[0044] Affine Linear weighted intra prediction (LWIP) [4] is a new intra prediction technique. Like conventional intra prediction, LWIP consists of a set of prediction modes. Given the left and above reconstructed (reference) samples, each of the signaled prediction modes corresponds to a prediction function that generates a different prediction signal.

[0045] Withbothconventional andLWIPmodesavailable, theencoder compares the rate-distortioncost of conventional and LWIP modes and decides for the mode with lowest overall cost. The selected prediction mode is then sent to the decoder in the bitstreamand it is used to select the correspondingprediction function for predicting theblock. Signaling the prediction mode has the following syntax: First, a flag is sent that indicates whether the block is predicted with a conventional or a LWIP mode. If conventional prediction is chosen, a conventional prediction mode is read from the bitstream according to the intra prediction signaling. Otherwise, if LWIP prediction is selected, the flag is followed by sending amode index within the set of available LWIPmodes. Since both conventional and LWIPmodes are available for all block sizes that are generally supported by the codec for intra prediction, the flag has to be sent for every block.

[0046] The VVC coding standard currently supports intra prediction for luma block sizes in the rangeW× H = {4, 8,16, 32, 64} × {4,8,16,32, 64}. Obviously, the encoder search with rate-distortion optimization (RDO) can become very complex, as the rate-distortion costs of all thedifferent block sizes (resulting fromsplitting larger blocks into smaller blocks) areevaluated forall thedifferent predictionmodes.For reducing theencoder complexity, optimizations typically reduce the number of combinations that are tested, excluding cases that are statistically unlikely to lead to the lowest rate-distortion cost.

[0047] The core prediction of LWIP only supports square blocks withWLWIP = HLWIP ∈ {4, 8, 16}. For enabling LWIP prediction for all the other block sizes, the reference samples are downsampled to match the core prediction size and the output is upsampled tomatch the block size. This has the effect that the prediction quality is reduced for block shapeswith highly unequal downand upsampling ratios for the horizontal and vertical direction. This means that for blocks with an aspect ratio larger than 2, i.e. or , it is less likely that an LWIP mode leads to a lower rate-distortion cost than a conventional prediction mode.

[0048] This effect can be used to reduce the encoder complexity by restricting LWIPmodes to blocks with aspect ratios smaller than or equal to 2 and not testing them for blocks with a higher aspect ratio. However, this also results in a somewhat lower codingefficiency.Not sending theflag for blockswith anaspect ratio larger than2would reduce the loss in codingefficiency, butwouldalsomake it impossible to realizeanencoder that achievesabetter overall codingefficiencyby testing LWIP modes for blocks with an aspect ratio larger than 2, which might be required for different applications.

[0049] The solution to support both fast and highly-compression-efficient encoders is introducing an individual, additional CABAC context for sending the flag for blocks with an aspect ratio larger than 2. Now, in case LWIP modes are not tested at the encoder for those blocks, the flag is always 0 (=block not predicted with an LWIPmode) and sending the flag causes almost no overhead (only for adapting the context towards a zero probability for sending a1),meaning that the coding efficiency is very close to the solution of not sending the flag for those blocks. In case LWIP blocks are tested at 7 EP 4 765 824 A2 5 10 15 20 25 30 35 40 45 50 55 theencoder for thoseblocks, theflag is 1 (=block is predictedwithanLWIPmode)witha certainprobability andsending the flag causes almost no overhead,meaning that the coding efficiency is very close to the solution of sending the flagwith the same context for all block sizes.

[0050] Inotherwords,asmentionedabove, inaccordancewithembodiments, the inventiveapproachuses four contexts or context models for the binarization of a syntax element intra_mip_flag representing an affine linear weighted intra prediction, LWIP, ormatrix-based intra prediction,MIP, of the versatile video coding, VVC, standard. The selection index is controlled the aspect ratio (width / height or height / width) of a current block:

[0051] Fig. 3 illustrates an apparatus, e.g., an apparatus similar to the one of Fig. 1, for encoding image or video data in accordance with embodiments of the present invention introducing an individual, additional CABAC context for sending a flag for blocks with an aspect ratio larger than 2. The encoder 102 receives the image or video data 104 and encodes the received image or video data 104 so as to provide the bitstream 106 representing the encoded image or video data. The CABAC encoder 108 receives the tool flag 110 indicating whether or not certain coding tool, like the affine linear weighted intra prediction, LWIP, is employed when encoding a block of the image or video data. The affine linear weighted intra prediction, LWIP, is also referred to asmatrix-based intra prediction,MIP, in the versatile video coding, VVC, standard, and the tool flag is also referred to as intra_mip_flag, indicating the applicability of the affine LWIP orMIP of the VVC standard. For a block 300 of the image or video data having an aspect ratio larger than 2 and to which the certain coding tool is applicable, a first contextmodel for encoding the tool flag is selected froma group of one ormore first context models, and provided to theCABAC108. For a block 302 of the image or video data having an aspect ratio less than or equal to 2 and to which the certain coding tool is applicable, a secondcontextmodel for encoding the tool flag is selected fromagroupof one ormore second contextmodels, and provided to theCABAC108. For example, in case LWIPmodes are not tested for the block 300 with an aspect ratio larger than 2, the flag is always 0 and the additional CABAC context adapts towards a zero probability for sending a 1,while in case LWIPmodes are tested for the block 300with an aspect ratio larger than 2, the flag is 1with a certain probability. In accordancewith embodiments, theCABACencoder 108mayselect the first contextmodel and the second context model for a currently processed block responsive to a selection index. The selection index indicates that a currently processed block has an aspect ratio larger than 2 or has an aspect ratio less than or equal to 2.

[0052] Fig. 4 illustratesanapparatus, e.g., anapparatus 100similar to theoneof Fig. 2, for decoding imageor videodata inaccordancewithembodimentsof thepresent invention introducing the individual, additionalCABACcontext for a flag for blocks with an aspect ratio larger than 2 and encoded using the apparatus of Fig. 3. For a block 300 of the image or video datahavinganaspect ratio larger than2and towhich thecertain coding tool isapplicable, afirst contextmodel for encoding the tool flag is selected fromagroupof oneormore first contextmodels, andprovided to theCABAC206.For ablock302of the image or video data having an aspect ratio less than or equal to 2 and to which the certain coding tool is applicable, a second context model for encoding the tool flag is selected from a group of one or more second context models, and provided to the CABAC 206.

[0053] For example, the binarization of syntax element intra_mip_flag may employ a total of four context models with context index {0, 1, 2, 3} as follows: - if the aspect ratio (width / height or height / width) of the current block is larger than2, use the contextmodelwith index 3, - otherwise use one of the contextmodels 0, 1, 2, wherein the selectionmay depend on the intra_mip_flag of blocks left and above of the current block, as is, for example, known and used for several other syntax elements.

[0054] In the VVC specification the coding unit syntax may be as follows (see for example 7.3.10.5 of [5] - Coding unit syntax): ... if( sps_mip_enabled_flag ) intra_mip_flag[ x0 ][ y0 ] ae(v) if( intra_mip_flag[ x0 ][ y0 ] ) { intra_mip_transposed[ x0 ][ y0 ] ae(v) 8 EP 4 765 824 A2 5 10 15 20 25 30 35 40 45 50 55 (continued) intra_mip_mode[ x0 ][ y0 ] ae(v) } else ...

[0055] intra_mip_flag[ x0 ][ y0 ] equal to 1 specifies that the intra prediction type for luma samples is matrix-based intra prediction. intra_mip_flag[ x0 ][ y0 ] equal to 0 specifies that the intra prediction type for luma samples is not matrix-based intra prediction (see for example 7.4.11.5 of [5] - Coding unit semantics).

[0056] The binarization of syntax element intra_mip_flag may be as follows (see for example 9.3.4.2 of [5] - Derivation process for ctxTable, ctxldx and bypassFlag): Syntax element binldx 0 1 2 3 4 >= 5 ... intra_mip_flag[ ][ ] (Abs( Log2(cbWidth) - Log2(cbHeight) ) > 1) ? na na na na na 3 : ( 0,1,2 (clause 9.3.4.2.2) )

[0057] Although some aspects of the described concept have been described in the context of an apparatus, it is clear that these aspects also represent a description of the corresponding method, where a block or a device corresponds to a methodstepor a featureof amethodstep.Analogously, aspects described in the context of amethodstepalso represent a description of a corresponding block or item or feature of a corresponding apparatus.

[0058] Various elements and features of the present invention may be implemented in hardware using analog and / or digital circuits, in software, through the execution of instructions by one or more general purpose or special-purpose processors, or as a combination of hardware and software. For example, embodiments of the present invention may be implemented in the environment of a computer system or another processing system. Fig. 5 illustrates an example of a computer system400. The units ormodules aswell as the steps of themethods performed by these unitsmay execute on one or more computer systems 400. The computer system 400 includes one or more processors 402, like a special purpose or a general purpose digital signal processor. The processor 402 is connected to a communication infrastructure 404, like a bus or a network. The computer system 400 includes a main memory 406, e.g., a random access memory (RAM), and a secondary memory 408, e.g., a hard disk drive and / or a removable storage drive. The secondary memory 408may allow computer programs or other instructions to be loaded into the computer system400. The computer system 400 may further include a communications interface 410 to allow software and data to be transferred between computer system 400 and external devices. The communication may be in the from electronic, electromagnetic, optical, or other signals capable of being handled by a communications interface. The communication may use a wire or a cable, fiber optics, a phone line, a cellular phone link, an RF link and other communications channels 412.

[0059] The terms "computer programmedium" and "computer readablemedium" are used to generally refer to tangible storage media such as removable storage units or a hard disk installed in a hard disk drive. These computer program products are means for providing software to the computer system 400. The computer programs, also referred to as computer control logic, are stored in main memory 406 and / or secondary memory 408. Computer programsmay also be received via the communications interface 410. The computer program, when executed, enables the computer system 400 to implement the present invention. In particular, the computer program, when executed, enables processor 402 to implement the processes of the present invention, such as any of the methods described herein. Accordingly, such a computer program may represent a controller of the computer system 400. Where the disclosure is implemented using software, the software may be stored in a computer program product and loaded into computer system 400 using a removable storage drive, an interface, like communications interface 410.

[0060] The implementation in hardware or in software may be performed using a digital storage medium, for example cloud storage, a floppy disk, a DVD, a Blue-Ray, a CD, a ROM, a PROM, an EPROM, an EEPROM or a FLASHmemory, having electronically readable control signals stored thereon, which cooperate (or are capable of cooperating) with a programmablecomputer systemsuch that the respectivemethod isperformed.Therefore, thedigital storagemediummay be computer readable.

[0061] Some embodiments according to the invention comprise a data carrier having electronically readable control signals, which are capable of cooperatingwith a programmable computer system, such that one of themethods described herein is performed.

[0062] Generally, embodiments of the present invention may be implemented as a computer program product with a 9 EP 4 765 824 A2 5 10 15 20 25 30 35 40 45 50 55 program code, the program code being operative for performing one of themethods when the computer program product runs on a computer. The program code may for example be stored on a machine readable carrier.

[0063] Other embodiments comprise the computer program for performing oneof themethods described herein, stored on amachine readable carrier. In other words, an embodiment of the inventive method is, therefore, a computer program having a program code for performing one of the methods described herein, when the computer program runs on a computer.

[0064] A further embodiment of the inventive methods is, therefore, a data carrier (or a digital storage medium, or a computer-readable medium) comprising, recorded thereon, the computer program for performing one of the methods described herein. A further embodiment of the inventive method is, therefore, a data stream or a sequence of signals representing the computer program for performingoneof themethods describedherein. Thedata streamor the sequence of signals may for example be configured to be transferred via a data communication connection, for example via the Internet. A further embodiment comprises a processingmeans, for example a computer, or a programmable logic device, configured to or adapted to perform one of the methods described herein. A further embodiment comprises a computer having installed thereon the computer program for performing one of the methods described herein.

[0065] In someembodiments, a programmable logic device (for example a field programmable gate array)maybe used to perform some or all of the functionalities of themethods described herein. In some embodiments, a field programmable gate array may cooperate with a microprocessor in order to perform one of the methods described herein. Generally, the methods are preferably performed by any hardware apparatus.

[0066] The above described embodiments are merely illustrative for the principles of the present invention. It is understood that modifications and variations of the arrangements and the details described herein are apparent to others skilled in the art. It is the intent, therefore, to be limited only by the scope of the impending patent claims and not by the specific details presented by way of description and explanation of the embodiments herein. References

[0067] [1] ISO / IEC, ITU-T. High efficiency video coding. ITU-T Recommendation H.265 | ISO / IEC 23008 10 (HEVC), edition 1, 2013; edition 2, 2014. [2] JEM reference software, https: / / jvet.hhi.fraunhofer.de / svn / svn_HMJEMSoftware / . [3] B. Bross, J. Chen, Shan Liu, "Versatile Video Coding (Draft 4)", JVET-M1001-v5, February 2019, Marrakesh, Morocco [4] J. Pfaff, B. Stallenberger, M. Schäfer, P. Merkle, P. Helle, R. Rische, H. Schwarz, D. Marpe, T. Wiegand, "Affine Linear Weighted Intra Prediction", JVET-M0043, February 2019, Marrakesh, Morocco [5] B. Bross, J. Chen, Shan Liu, "Versatile Video Coding (Draft 8)", JVET-Q2001-vD, February 2020, Brussels, Belgium The following specific embodiments are also encompassed by the present disclosure. Specific embodiment 1. An apparatus, comprising: anencoder, theencoder to receive imageor videodata, toencode the received imageor videodataand toprovide a bitstream representing the image or video data, the encoder including a CABAC encoder, the CABAC encoder to receive a binary valued syntax element associatedwith a block of the image or video data to be encoded, and to encode, using a selected contextmodel, the binary valued syntax element into coded bits for the bitstream, wherein the binary valued syntax element includes a tool flag indicating whether or not certain coding tool, like a linear weighted intra prediction, LWIP, is employed when encoding a block of the image or video data, wherein, for a block of the image or video data having an aspect ratio larger than 2 and towhich the certain coding tool is applicable, a first context model for encoding the tool flag is selected from a group of one or more first context models, wherein, for a block of the image or video data having an aspect ratio less than or equal to 2 and to which the certain coding tool is applicable, a second contextmodel for encoding the tool flag is selected froma group of one or more second context models. 10 EP 4 765 824 A2 5 10 15 20 25 30 35 40 45 50 55 Specific embodiment 2. The apparatus of specific embodiment 1, wherein the coding tool is a linear weighted intra prediction, LWIP, and the first contextmodel includesaCABACcontext for sending the tool flag for ablockof the image or video data with an aspect ratio larger than 2, wherein • in case LWIP modes are not tested for the block with an aspect ratio larger than 2, the flag is always 0 and the additional CABAC context adapts towards a zero probability for sending a 1, and • in caseLWIPmodesare tested for theblockwithanaspect ratio larger than2, theflag is1withacertainprobability. Specific embodiment 3. Theapparatus of specific embodiment 1 or 2,wherein theCABACencoder is to select the first context model and the second context model for a currently processed block responsive to a selection index, the selection index indicating that a currently processed block has an aspect ratio larger than 2 or has an aspect ratio less than or equal to 2. Specific embodiment 4. An apparatus, comprising: a decoder, thedecoder to receiveabitstream includingencoded imageor videodata, to decode from the received bitstream the encoded image or video data and to provide the decoded image or video data, the decoder including a CABAC decoder, the CABAC decoder to decode from the bitstream, using a selected context model, a binary valued syntax element associated with a block of the encoded image or video data, wherein the binary valued syntax element includes a tool flag indicatingwhether or not a certain coding tool, like a linear weighted intra prediction, LWIP, is employed when encoding a block of the image or video data, wherein, for a block of the image or video data having an aspect ratio larger than 2 and towhich the certain coding tool is applicable, a first context model for decoding the tool flag is selected from a group of one or more first context models, wherein, for a block of the image or video data having an aspect ratio less than or equal to 2 and to which the certain coding tool is applicable, a second contextmodel for decoding the tool flag is selected froma group of one or more second context models. Specific embodiment 5. A method for encoding image or video data, the method comprising: receiving image or video data, encoding the received image or video data, and providing a bitstream representing the image or video data, wherein encoding the received image or video data comprises: receiving, using a CABAC encoder, a binary valued syntax element associated with a block of the image or video data to be encoded, and encoding, using a selected contextmodel, the binary valued syntax element into coded bits for the bitstream, wherein the binary valued syntax element includes a tool flag indicating whether or not certain coding tool, like a linear weighted intra prediction, LWIP, is employed when encoding a block of the image or video data, wherein, for a block of the image or video data having an aspect ratio larger than 2 and towhich the certain coding tool is applicable, a first context model for encoding the tool flag is selected from a group of one or more first context models, wherein, for a block of the image or video data having an aspect ratio less than or equal to 2 and to which the certain coding tool is applicable, a second contextmodel for encoding the tool flag is selected froma group of one or more second context models. 11 EP 4 765 824 A2 5 10 15 20 25 30 35 40 45 50 55 Specific embodiment 6. A method for decoding image or video data, the method comprising: receiving a bitstream including encoded image or video data, decoding from the received bitstream the encoded image or video data, and providing the decoded image or video data, wherein decoding the received image or video data comprises decoding from the bitstream, using a CABAC decoder and a selected context model, a binary valued syntax element associated with a block of the encoded image or video data, wherein the binary valued syntax element includes a tool flag indicatingwhether or not a certain coding tool, like a linear weighted intra prediction, LWIP, is employed when encoding a block of the image or video data, wherein, for a block of the image or video data having an aspect ratio larger than 2 and towhich the certain coding tool is applicable, a first context model for decoding the tool flag is selected from a group of one or more first context models, wherein, for a block of the image or video data having an aspect ratio less than or equal to 2 and to which the certain coding tool is applicable, a second contextmodel for decoding the tool flag is selected froma group of one or more second context models. Specificembodiment 7.Acomputer programproduct comprising instructionswhich,when theprogram isexecutedby a computer, cause the computer to carry out the method of any one of specific embodiments 5 to 6. Specific embodiment 8. An apparatus, comprising: anencoder, theencoder to receive imageor videodata, toencode the received imageor videodataand toprovide a bitstream representing the image or video data, the encoder including a CABAC encoder, the CABAC encoder to receive a binary valued syntax element associatedwith a certain data block of the image or videodata to be encoded, to encode, using a selected context model, the binary valued syntax element into coded bits for the bitstream, wherein the binary valued syntax element includes a tool flag indicating whether or not a certain coding tool is employed when encoding the image or video data, wherein agroupof first contextmodels for encoding the tool flag is selected for oneormore first parts of the certain data block to which, independent of an application, the coding tool is always applicable, whereinagroupof secondcontextmodels for encoding the tool flag is selected for oneormoresecondparts of the certain data block to which, dependent on the application, the coding tool is applicable or is not applicable. Specificembodiment 9.Theapparatusof specificembodiment8,wherein theCABACencoderselects thefirst context model or the second context model for a currently processed part of the certain data block responsive to a selection index, the selection indexhavinga first value indicating that the currently processedpart of the certain data block is the first part, and the selection index having a second value indicating that the currently processed part of the certain data block is the second part. Specific embodiment 10. An apparatus, comprising: a decoder, thedecoder to receiveabitstream includingencoded imageor videodata, to decode from the received bitstream the encoded image or video data and to provide the decoded image or video data, the decoder including a CABAC decoder, the CABAC decoder to decode from the bitstream, using a selected contextmodel, a binary valuedsyntax element associatedwith a certain data blockof the encoded imageor video data, 12 EP 4 765 824 A2 5 10 15 20 25 30 35 40 45 50 55 wherein the binary valued syntax element includes a tool flag indicating whether or not a certain coding tool is employed when encoding the image or video data, wherein a group of first context model for decoding the tool flag is selected for parts of the certain data block to which, independent of an application, the coding tool is always applicable, wherein a group of second contextmodel for decoding the tool flag is selected for parts of the certain data block to which, dependent on an application, the coding tool is applicable or is not applicable. Specific embodiment 11. The apparatus of specific embodiment 10, wherein the CABAC decoder selects the first context model or the second context model for a currently processed part of the certain data block responsive to a selection index, the selection index having a first value indicating that the currently processed part of the certain data block is the first part, and the selection index having a second value indicating that the currently processed part of the certain data block is the second part. Specific embodiment 12. The apparatus of any one of specific embodiments 8 to 11, wherein the group of first context models includes one first context model or a variety of first context models, and wherein the group of second context models includes one second context model or a variety of second context model. Claims 1. Anencoder (102) for encodingapicture, theencoder comprisinga context adaptivebinary arithmetic coding,CABAC, encoder (180), the encoder configured to: - receive the picture, - encode the received picture, - receive, by the CABAC encoder (108), a binary valued syntax element (110) associated with a block of the picture to be encoded, - encode, by the CABAC encoder and using a selected context model, the binary valued syntax element into coded bits (112) for a bitstream (106), -‑ wherein the binary valued syntax element includes a tool flag indicating whether or not a certain linear weighted intra prediction, LWIP, coding tool is employed when encoding a block of the picture, wherein the certain LWIP coding tool corresponds to a matrix-based intra prediction, MIP, of the versatile video coding, VVC, standard, -‑wherein, for a block of the picture having an aspect ratio larger than 2 and towhich the certain LWIP coding tool is applicable, a first context model for encoding the tool flag is selected from a group of one or more first context models, and -‑ wherein, for a block of the picture having the aspect ratio less than or equal to 2 and to which the certain LWIPcoding tool isapplicable, asecondcontextmodel, different from thefirst contextmodel, for encoding the tool flag is selected from a group of one or more second context models; and - provide the encoded bitstream representing the picture. 2. Theencoder of claim1,wherein the first contextmodel includesaCABACcontext for sending the tool flag for theblock of the picture with the aspect ratio larger than 2, wherein • in case LWIP modes are not tested for the block with the aspect ratio larger than 2, the flag is always 0 and the additional CABAC context adapts towards a zero probability for sending a 1, and • in case LWIP modes are tested for the block with the aspect ratio larger than 2, the flag is 1 with a certain probability. 3. The apparatus of claim 1 or 2, wherein the CABAC encoder is to select the first context model and the second context model for a currently processed block responsive to a selection index, the selection index indicating that a currently processed block has the aspect ratio larger than 2 or has the aspect ratio less than or equal to 2. 4. A method for encoding a picture, the method comprising: 13 EP 4 765 824 A2 5 10 15 20 25 30 35 40 45 50 55 - receiving the picture, - encoding the received picture data, wherein encoding the received picture comprises: -‑ receiving, using a context adaptive binary arithmetic coding, CABAC, encoder (108), a binary valued syntax element (110) associated with a block of the picture to be encoded, and -‑ encoding, by the CABAC encoder and using a selected context model, the binary valued syntax element into coded bits (112) for a bitstream (106), -‑ wherein the binary valued syntax element includes a tool flag indicating whether or not a certain linear weighted intra prediction, LWIP, coding tool is employed when encoding a block of the picture, wherein the certain LWIP coding tool corresponds to a matrix-based intra prediction, MIP, of the versatile video coding, VVC, standard, -‑wherein, for a block of the picture having an aspect ratio larger than 2 and towhich the certain LWIP coding tool is applicable, a first context model for encoding the tool flag is selected from a group of one or more first context models, and -‑ wherein, for a block of the picture having the aspect ratio less than or equal to 2 and to which the certain LWIPcoding tool isapplicable, asecondcontextmodel, different from thefirst contextmodel, for encoding the tool flag is selected from a group of one or more second context models, and - providing the encoded bitstream representing the picture. 5. A computer program product comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method of claim 4. 14 EP 4 765 824 A2 5 10 15 20 25 30 35 40 45 50 55 15 EP 4 765 824 A2 16 EP 4 765 824 A2 17 EP 4 765 824 A2 18 EP 4 765 824 A2 REFERENCES CITED IN THE DESCRIPTION This list of references cited by the applicant is for the reader’s convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard. Non-patent literature cited in the description • ISO / IEC, ITU-T. High efficiency video coding. ITU-T Recommendation H.265 | ISO / IEC 23008 10 (HEVC), 2014

[0067] • JEM reference software, https: / / jvet.hhi.fraunhofer. de / svn / svn_HMJEMSoftware

[0067] • B. BROSS ; J. CHEN ; SHAN LIU. Versatile Video Coding (Draft 4). JVET-M1001-v5, February 2019

[0067] • J.PFAFF ;B.STALLENBERGER ;M.SCHÄFER ;P. MERKLE ;P.HELLE ;R.RISCHE ;H.SCHWARZ ;D. MARPE ; T. WIEGAND. Affine Linear Weighted Intra Prediction. JVET-M0043, February 2019

[0067] • B. BROSS ; J. CHEN ; SHAN LIU. Versatile Video Coding (Draft 8). JVET-Q2001-vD, February 2020

[0067] 摘要 本文描述了一種包含編碼器的裝置。此編碼器接收影像或視訊數據，對接收到的影像 或視訊資料進行編碼，並提供表示該影像或視訊資料的位元流。此編碼器包含一個 CABAC 編碼器。 CABAC 編碼器接收與待編碼的影像或視訊資料區塊關聯的二進位值 語法元素，並使用選定的上下文模型將該二進位值語法元素編碼為位元流的編碼位 元。此二進位值語法元素包含一個工具標誌，用於指示在對影像或視訊資料區塊進行 編碼時是否採用特定的編碼工具，例如線性加權幀內預測 (LWIP)。對於縱橫比大於 2 且適用特定編碼工具的圖像或視訊資料區塊，從一個或多個第一上下文模型中選擇用 於編碼工具標誌的第一上下文模型；對於縱橫比小於或等於 2 且適用特定編碼工具的 圖像或視訊資料區塊，從一個或多個第二上下文模型中選擇用於編碼工具標誌的第二 上下文模型。

Claims

1. An encoder (102) for encoding a picture, the encoder comprising a context adaptive binary arithmetic coding, CABAC, encoder (180), the encoder configured to: - receive the picture, - encode the received picture, - receive, by the CABAC encoder (108), a binary valued syntax element (110) associated with a block of the picture to be encoded, - encode, by the CABAC encoder and using a selected context model, the binary valued syntax element into coded bits (112) for a bitstream (106), -- wherein the binary valued syntax element includes a tool flag indicating whether or not a certain linear weighted intra prediction, LWIP, coding tool is employed when encoding a block of the picture, wherein the certain LWIP coding tool corresponds to a matrix-based intra prediction, MIP, of the versatile video coding, VVC, standard, -- wherein, for a block of the picture having an aspect ratio larger than 2 and to which the certain LWIP coding tool is applicable, a first context model for encoding the tool flag is selected from a group of one or more first context models, and -- wherein, for a block of the picture having the aspect ratio less than or equal to 2 and to which the certain LWIP coding tool is applicable, a second context model, different from the first context model, for encoding the tool flag is selected from a group of one or more second context models; and - provide the encoded bitstream representing the picture.

2. The encoder of claim 1, wherein the first context model includes a CABAC context for sending the tool flag for the block of the picture with the aspect ratio larger than 2, wherein • in case LWIP modes are not tested for the block with the aspect ratio larger than 2, the flag is always 0 and the additional CABAC context adapts towards a zero probability for sending a 1, and • in case LWIP modes are tested for the block with the aspect ratio larger than 2, the flag is 1 with a certain probability.

3. The apparatus of claim 1 or 2, wherein the CABAC encoder is to select the first context model and the second context model for a currently processed block responsive to a selection index, the selection index indicating that a currently processed block has the aspect ratio larger than 2 or has the aspect ratio less than or equal to 2.

4. A method for encoding a picture, the method comprising: - receiving the picture, - encoding the received picture data, wherein encoding the received picture comprises: -- receiving, using a context adaptive binary arithmetic coding, CABAC, encoder (108), a binary valued syntax element (110) associated with a block of the picture to be encoded, and -- encoding, by the CABAC encoder and using a selected context model, the binary valued syntax element into coded bits (112) for a bitstream (106), -- wherein the binary valued syntax element includes a tool flag indicating whether or not a certain linear weighted intra prediction, LWIP, coding tool is employed when encoding a block of the picture, wherein the certain LWIP coding tool corresponds to a matrix-based intra prediction, MIP, of the versatile video coding, VVC, standard, -- wherein, for a block of the picture having an aspect ratio larger than 2 and to which the certain LWIP coding tool is applicable, a first context model for encoding the tool flag is selected from a group of one or more first context models, and -- wherein, for a block of the picture having the aspect ratio less than or equal to 2 and to which the certain LWIP coding tool is applicable, a second context model, different from the first context model, for encoding the tool flag is selected from a group of one or more second context models, and - providing the encoded bitstream representing the picture.

5. A computer program product comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method of claim 4.

Images