Intra random access point for picture coding

Abstract

To provide a method and device for video processing, such as video encoding and decoding, which can avoid redundant signaling of information and improves error resilience.SOLUTION: In one example, a video processing method comprises performing a conversion between a video having one or more video layers including one or more video pictures and a bitstream of the video according to a format rule. The format rule specifies that a first video picture, which is an associated intra random access point picture of a second picture, and the second picture are constrained to belong to the same video layer.SELECTED DRAWING: Figure 7A

Description

Technical Field

[0001] This application is a divisional application of Japanese Patent Application 2022-556152, based on International Patent Application PCT / US2021 / 022572, filed on 16 March 2021, claiming timely priority and interest of U.S. Patent Application US62 / 992,046, filed on 19 March 2020. The entire disclosure of the above application is incorporated by reference as part of the disclosure herein.

[0002] This patent specification relates to the encoding and decoding of images and videos.

Background Art

[0003] Digital videos occupy the largest bandwidth usage in the Internet and other digital communication networks. As the number of connected user devices capable of receiving and displaying videos increases, the bandwidth demand for the use of digital videos is expected to continue to grow.

Summary of the Invention

[0004] This specification discloses a technique that can be used by a video encoder and a decoder for processing a coded representation of a video using a bitstream syntax that provides improved performance. The disclosed method may be used by an apparatus for performing video processing such as video encoding, or video decoding, or video codec conversion, etc.

[0005] In one exemplary aspect, a video processing method is disclosed. The method includes performing a conversion between a video having one or more video layers including one or more video pictures and a coded representation of the video, wherein the coded representation is compiled according to a rule that defines that a first video picture, which is an intra-random access point picture of a second picture, and the second picture are constrained to belong to the same video layer.

[0006] ​​​​​​​​​​​In another exemplary embodiment, another image processing method is disclosed. This method is 1 or multiple Video and video coding having one or more video layers containing several video pictures. This includes performing conversions between the coded expression and the coded expression, intrarun The coded representation follows the first type of picture, which is a dam access point. Trailing pictures also include a second type, which includes progressively decoded refresh pictures. It complies with formatting rules that stipulate that it is permitted to be associated with a picture.

[0007] In another exemplary embodiment, another image processing method is disclosed. This method is 1 or multiple Video and video coding having one or more video layers containing several video pictures. This includes performing conversions between the coded and coded expressions, and the coded expressions are entered in the decoding order. The constraint on the output order of pictures preceding a random access point is that the output order is the same. Complies with formatting rules that specify that the rules are applicable only to pictures within the same video layer. do.

[0008] In another exemplary embodiment, another image processing method is disclosed. This method is 1 or multiple Video having one or more video layers containing several video pictures, and video coding This includes performing a conversion between the coded expression and the (1) The Irling picture, in output order, is associated with IRAP (Intra Rand (Access Point) Picture or GDR (Gradual Decode) (er Refresh) must be followed by a picture, or (2) GDR picture Pictures with the same layer ID as the one in the output order are GDR picture and G The constraint that it must precede all associated pictures of the DR picture It complies with the prescribed formatting rules.

[0009] In another exemplary embodiment, another image processing method is disclosed. This method is 1 or multiple Video having one or more video layers containing several video pictures, and video coding This involves performing conversions between the converted and the resulting representations, and the conversions have an order constraint, picture, IRA Only when P-pictures and non-reading pictures are on the same layer, the picture, IRAP (Intra Random Access Point) picture, and In accordance with the rule that applies to non-reading pictures, the rule (a) feel The first rule that defines the sequence values ​​and decoding order, and (b) the reading of the layer It is either a bi / or non-reading picture in order.

[0010] In another exemplary embodiment, another image processing method is disclosed. This method is 1 or multiple Video having one or more video layers containing several video pictures, and video coding This includes performing a conversion between the converted and the resulting expression, and the conversion is performed using Reading Picture, RAD L (Random Access Decodable Leading) Picture, And associated with GDR (Gradual Decoding Refresh) pictures RASL (Random Access Skipped Leading) The order of the characters will be determined by the rules.

[0011] In another exemplary aspect, another video processing method is disclosed. This method includes performing a conversion between a video having one or more video layers including one or more video pictures and a coded representation of the video, the conversion being compliant with rules that specify that the reference picture list constraints for clean random access pictures are restricted to one layer. including performing a conversion between a video having one or more video layers including one or more video pictures and a coded representation of the video, where the conversion is compliant with rules that specify that the reference picture list constraints for clean random access pictures are restricted to one layer.

[0012] In another exemplary aspect, another video processing method is disclosed. This method includes performing a conversion between a video having one or more video layers including one or more video pictures and a coded representation of the video, where the conversion is compliant with rules that permit referring to entries in a reference picture list generated by a decoding process for generating reference pictures that are not available for the current picture. including performing a conversion between a video having one or more video layers including one or more video pictures and a coded representation of the video, where the conversion is compliant with rules that permit referring to entries in a reference picture list generated by a decoding process for generating reference pictures that are not available for the current picture.

[0013] In another exemplary aspect, another video processing method is disclosed. This method includes performing a conversion between a video having one or more video layers including one or more video pictures and a coded representation of the video, where the conversion is compliant with rules regarding the order between the current picture and a reference picture list corresponding to the current picture. including performing a conversion between a video having one or more video layers including one or more video pictures and a coded representation of the video, where the conversion is compliant with rules regarding the order between the current picture and a reference picture list corresponding to the current picture.

[0014] In another exemplary aspect, another video processing method is disclosed. This method includes performing a conversion between a video having one or more video layers including one or more video pictures and a video bitstream in accordance with formatting rules, where the formatting rules specify that the second picture is an associated intra random access point picture. including performing a conversion between a video having one or more video layers including one or more video pictures and a video bitstream in accordance with formatting rules, where the formatting rules specify that the second picture is an associated intra random access point picture. ​​​The first and second video pictures are constrained to belong to the same video layer. This defines...

[0015] In another exemplary embodiment, another image processing method is disclosed. This method is format In accordance with the rules, having one or more video layers containing one or more video pictures. This includes performing conversions between video and the video bitstream, and formatting rules. This involves progressively decoding the trailing picture in the bitstream to refresh the picture. This stipulates that it is permitted to associate it with ya.

[0016] In another exemplary embodiment, another image processing method is disclosed. This method is format In accordance with the rules, having one or more video layers containing one or more video pictures. This includes performing conversions between video and the video bitstream, and formatting rules. This is relative to the output order of the pictures preceding the intra-random access point in the decoding order. This specifies that the constraints apply to pictures within the same video layer.

[0017] In another exemplary embodiment, another image processing method is disclosed. This method is format In accordance with the rules, having one or more video layers containing one or more video pictures. This includes performing conversions between video and the video bitstream, and formatting rules. (1) The trailing picture is associated with the intra random access point Track the picture or progressive decoder refresh picture in output order, or (2 ) Gradual decoder refresh picture same NAL (Network Abstract) The picture having the ion Layer) unit header layer identifier is output in gradual order. All decoder refresh pictures and progressive decoder refresh pictures This imposes the constraint that it must precede the associated picture.

[0018] In another exemplary embodiment, another image processing method is disclosed. This method is in accordance with the rules. And, a video having one or more video layers containing one or more video pictures, and The rules include performing conversions between the bitstream and the picture, intra When a non-reading access point picture and a non-reading picture are on the same layer Only in the case of pictures and non-reading associated with intra-random access points. This specifies that constraints should be applied to the decoding order of the gpictures.

[0019] In another exemplary embodiment, another image processing method is disclosed. This method is in accordance with the rules. And, a video having one or more video layers containing one or more video pictures, and The rules include performing a conversion between the bitstream and the reading picture. , random access decryptable reading picture, and progressive decryption refresh Reading pictures associated with random access have been skipped. It defines the order.

[0020] In another exemplary embodiment, another image processing method is disclosed. This method is in accordance with the rules. And, a video having one or more video layers containing one or more video pictures, and The rules include performing conversions between bitstreams, and clean random actions. Restricting the constraints on the reference picture list for slicing a picture to layers This defines the following.

[0021] In another exemplary embodiment, another image processing method is disclosed. This method is in accordance with the rules. And, a video having one or more video layers containing one or more video pictures, and The rules include performing a conversion between the current picture and the bitstream, and the current picture The active entry in the slice's reference picture list refers to an unavailable reference. The decoding process used to generate the illuminated picture did not produce any pictures. Define the conditions.

[0022] In another exemplary embodiment, another image processing method is disclosed. This method is in accordance with the rules. And, a video having one or more video layers containing one or more video pictures, and The rules include performing a conversion between the current picture and the bitstream, and the current picture Unavailable reference pictures referenced by entries in the reference picture list of the slice This specifies the condition that no pictures are generated by the decoding process used to create them. do.

[0023] In another exemplary embodiment, another image processing method is disclosed. This method is in accordance with the rules. And, a video having one or more video layers, including the current picture which includes the current slice. This includes performing a conversion between the video bitstream and the current slide. The reference picture list is associated with the current picture via intra-random access. Active entries pointing to the picture preceding the point picture in decoding order or output order. It stipulates the condition that having it is not permitted.

[0024] In another exemplary embodiment, another image processing method is disclosed. This method is in accordance with the rules. And, a video having one or more video layers, including the current picture which includes the current slice. This includes performing a conversion between the video bitstream and the current slide. The reference picture list is associated with the current picture via intra-random access. A point picture has an entry that points to a preceding picture in the decoding order or output order. This specifies the conditions under which it is not permitted.

[0025] In yet another exemplary embodiment, a video encoder device is disclosed. It comprises a processor configured to implement the method described above.

[0026] In yet another exemplary embodiment, a video decoder device is disclosed. The video decoder is It includes a processor configured to implement the method described above.

[0027] In yet another exemplary embodiment, a computer-readable medium on which code is stored is disclosed. This code describes one of the methods described herein in the form of code that can be executed by a processor. implement.

[0028] These and other features will be explained throughout this document. [Brief explanation of the drawing]

[0029] [Figure 1] Figure 1 is a block diagram showing an example of an image processing system. [Figure 2] Figure 2 is a block diagram of the video processing device. [Figure 3]Figure 3 is a flowchart illustrating an example of an image processing method. [Figure 4] Figure 4 is a block diagram showing a video coding system according to several embodiments of the present disclosure. [Figure 5] Figure 5 is a block diagram showing encoders according to several embodiments of the present invention. [Figure 6] Figure 6 is a block diagram showing decoders according to several embodiments of the present invention. [Figure 7A] Figure 7A shows a flowchart illustrating an exemplary method of image processing based on several implementations of the disclosed technology. [Figure 7B] Figure 7B shows a flowchart illustrating an exemplary method of image processing based on several implementations of the disclosed technology. [Figure 7C] Figure 7C shows a flowchart illustrating an exemplary method of image processing based on several implementations of the disclosed technology. [Figure 7D] Figure 7D shows a flowchart illustrating an exemplary method of image processing based on several implementations of the disclosed technology. [Figure 7E] Figure 7E shows a flowchart illustrating an exemplary method of image processing based on several implementations of the disclosed technology. [Figure 7F] Figure 7F shows flowcharts illustrating exemplary methods of image processing based on several implementations of the disclosed technology. [Figure 7G] Figure 7G shows a flowchart illustrating an exemplary method of image processing based on several implementations of the disclosed technology. [Figure 8A] Figure 8A shows a flowchart illustrating an exemplary method of image processing based on several implementations of the disclosed technology. [Figure 8B] Figure 8B shows a flowchart illustrating an exemplary method of image processing based on several implementations of the disclosed technology. [Figure 9A] Figure 9A shows a flowchart illustrating an exemplary method of image processing based on several implementations of the disclosed technology. [Figure 9B] Figure 9B shows flowcharts illustrating exemplary methods of image processing based on several implementations of the disclosed technology. [Modes for carrying out the invention]

[0030] This specification uses chapter headings to facilitate understanding of the technology and each The applicability of the embodiments described in a chapter is not limited to that chapter alone. Furthermore, H The term .266 is used in some explanations solely for the purpose of facilitating understanding and disclosure. This is not intended to limit the scope of the technology being described. The revealed technology is applicable to other video codec protocols and designs.

[0031] 1. Outline of the Invention This specification relates to video coding technology. Specifically, different types of pictures Random access, including definition, decoding order, output order, and their relationships in prediction relationships. Various forms of support for sublayer switching and scalability This relates to multi-layer video coding, either individually or in various combinations. For example, VVC (Versatile Video Coding), which is currently under development. This may apply to any video coding convention or non-standard video codec that supports it. stomach.

[0032] 2. Abbreviation APS Adaptation Parameter Set AU Access Unit AUD Access Unit Delimiter AVC Advanced Video Coding CLVS Coded Layer Video Sequence CPB Coded Picture Buffer CRA Clean Random Access CTU Coding Tree Unit CVS Coded Video Sequence DCI Decoding Capability Information DPB Decoded Picture Buffer EOB End Of Bitstream EOS End Of Sequence GDR Gradual Decoding Refresh HEVC High Efficiency Video Coding HRD Hypothetical Reference Decoder IDR Instantaneous Decoding Refresh JEM Joint Exploration Model MCTS Motion-Constrained Tile Sets NAL Network Abstraction Layer OLS Output Layer Set PH Picture Header PPS Picture Parameter Set PTL Profile,Tier and Level PU Picture Unit RADL Random Access Decodable Leading(P icture) RAP Random Access Point RASL Random Access Skipped Leading(Pic ture) RBSP Raw Byte Sequence Payload RPL Reference Picture List SEI Supplemental Enhancement Information n SPS Sequence Parameter Set STSA Step-wise Temporal Sublayer Access SVC Scalable Video Coding VCL VideoCoding Layer VPS Video Parameter Set VTM VVC Test Model VUI Video Usability Information VVC Versatile Video Coding

[0033] 3. Initial consultations Video coding standards are primarily developed based on well-known ITU-T and ISO / IEC standards. This is how it has developed. ITU-T created H.261 and H.263, and ISO / IEC developed MP EG-1 and MPEG-4 Visual were created, and the two organizations collaborated on H.262 / MPEG-2 V ideo and H.264 / MPEG-4 AVC (Advanced Video Cod They jointly created the H.265 / HEVC standard with (ing). Since H.262, video coding The standard is a hybrid video coding structure that utilizes time prediction and transformation coding. Based on this, in 2015, in order to explore future video coding technologies beyond HEVC, This is a joint project between VCEG and MPEG called JVET (Joint Video Explorer). The ion Team was established. Since then, many new methods have been adopted by JVET. This is a reference software called JEM (Joint Exploration Model). It has been incorporated into the software. JVET meets quarterly and new coding standards It aims for a 50% reduction in bitrate compared to HEVC. (JVE, April 2018) At the T conference, the new video coding standard was introduced as "VVC (Versatile Vid It was officially named "eo Coding" and at that time, the first version of VTM (VVC Test) was released. We have released the Model. Efforts to contribute to the standardization of VVC are continuing, so At all JVET meetings, new coding techniques are being adopted for the VVC standard. After each meeting, the VVC working draft and test model VTM are updated. Kuto is currently aiming for Technical Progression (FDIS) at the meeting in July 2020.

[0034] 3.1 General and SVC (Scalable Video Coding) in VVC g) SVC (Scalable Video Coding, sometimes used for video coding) Scalability in BL (Base Layer) (Sometimes called RL (Reference Layer)) and 1 or multiple scalable enhancement layers (EL: Enhancement L Refer to the video coding in which ayer is used. In SVC, the base layer is It can carry video data at a basic quality level. One or more enhancements Traya, for example, offers higher spatial, temporal, and / or SNR (Signal- It can carry additional video data to support the (to-Noise) level. The enhancement layer may be defined relative to the previous, encoded layer. For example, the bottom layer can function as a BL (Blue Line) and the top layer can function as an EL (Earth Line). Yes, it's possible. The intermediate layer can function as either an EL (Earth Layer) or an RL (Light Layer), or both. This is possible. For example, an intermediate layer (for example, a layer that is neither the bottom layer nor the top layer) , the layer below the intermediate layer, for example, the base layer or any intervening enhancement It may also be an EL for a layer, and at the same time, one or more enhancers on top of the intermediate layer. It serves as the RL for the comment layer. Similarly, it is a multiview for the HEVC standard. - Or, in 3D extensions, multiple views may exist, and the information from one view may be used. Information from another view can be coded (e.g., encoded or decoded) (example) For example, motion estimation, motion vector prediction, and / or other redundancies.

[0035] In SVC, the parameters used by the encoder or decoder utilize them. Coding levels that can be set (e.g., video level, sequence level, picture level) The parameters are grouped into sets based on factors such as the slice level and slice level. For example, a coded video sequence in different layers of a bitstream The available parameters are included in the VPS (Video Parameter Set). It may also be done by one or more pictures in the coded video sequence The parameters used are SPS (Sequence Parameter Set). It may also be included in the picture. Similarly, the para may be used in one or more slices in the picture. Meters may be included in the PPS (Picture Parameter Set). Other parameters specific to a single slice may be included in the slice header. Similarly, The instruction of which parameter set a particular layer is using at a given time is given by various methods. It may be provided at the development level.

[0036] RPR (Reference Picture Resampling) in VVC Thanks to the support of ), upsampling is required to support spatial scalability. This only requires using an RPR upsampling filter, so additional signal processing level Without requiring coding tools, multiple layers, for example, SD in VVC and Designed to support a bitstream containing two layers of HD resolution. This is possible. Nevertheless, for scalability support, a high level of syntax is required. Changes are needed (compared to not supporting scalability). Support for AVC and HEVC is specified in VVC version 1. Unlike scalability support in any previous video coding standard, this includes... The scalability design of VVCs is as suitable as possible for single-layer decoder designs. It has been made into something. The decoding ability of multilayer bitstreams It is defined as if there were only one layer. For example, decoding such as DPB size The capability is defined by a method that does not depend on the number of layers in the bitstream being decoded. Essentially, decoders designed for single-layer bitstreams are multi-layer Many changes are needed to enable the decryption of the bitstream. Compared to the multi-layer extension designs of AVC and HEVC, the HLS configuration is somewhat flexible. It has been greatly simplified at the expense of flexibility. For example, IRAP AU exists in CVS Each layer must contain a picture.

[0037] 3.2 Random Access and its Support in HEVC and VVC Random access refers to accessing any picture that is not the first picture in the bitstream in the decryption order. This refers to initiating access to and decoding of the bitstream. Broadcast / M Tuning and channel splitting in multi-person video conferences and luticas. Replacement, local playback and exploration in streaming, and in streaming To support stream adaptation, bitstreams are subject to frequent random access points. It is necessary to include an integer, and generally, it is an intracoded picture. However, it may also be an intercoded picture (for example, a progressively decoded reflection) (In the case of a shoe).

[0038] HEVC uses IRAP( Intra Random Access Point) Signaling a picture Includes three types of IRAP pictures, namely IDR (Instantaneous Decoder Refresh), CRA(Clean Random Access) ), and BLA (Broken Link Access) pictures are supported. The IDR picture has an interpicture prediction structure that is the current GOP (Group-Of- It restricts the reference of any picture prior to the previous picture, and traditionally It is called a Closed GOP Random Access Point. CRA Pictures are The restriction is lifted by allowing the picture to reference the picture before the current GOP. If relaxed and in the case of random access, these will all be discarded. CRA pictures are, In the future, it will be called an Open GOP Random Access Point. BLA pictures are usually For example, when switching streams, the CRA picture has two bitstreams. It is generated by splicing a part of it. A better system for IRAP pictures To enable the use of the system, a total of six different NAL units are used for the IRAP picture. These units are defined to signal properties, and these units use HTTP (DASH) The above is used for random access support in dynamic adaptive streaming, ISO As defined in BMFF (ISO Base Media File Format) It can be used to make it more suitable for the type of stream access point.

[0039] VVC includes three types of IRAP pictures and two types of IDR pictures (associated). One type of RADL picture having been cut and associated RADL picture It supports other types that do not have and one type of CRA picture. It is basically the same as HEVC. There are mainly two types of BLA pictures in HEVC. For the reasons stated above, it is not included in VVC. i) The basic functionality of BLA picture is to send the sequence end NAL unit to CRA picture. This can be achieved by adding this, and the presence of this sequence termination NAL unit means that the subsequent P This demonstrates that Kucha initiates a new CVS in a single-layer bitstream. ii) In VVC development, the NAL unit type field of the NAL unit header. As shown by using 5 bits instead of 6 bits, it is less than HEVC. It would have been desirable to define the NAL unit type that does not exist.

[0040] Another important difference in random access support between VVC and HEVC is that VVC The goal is to support GDR in a more prescriptive way in C. In GDR, Decryption of the stream can be started from the encoded picture. Initially, it is not possible to correctly decode the entire picture area, but after multiple pictures... This allows for the correct decoding of the entire picture area. AVC and HEV C also provides signaling for GDR random access points and recovery points. Support GDR using recovery point SEI messages in VVC. A new NAL unit type is specified to indicate a GDR picture, and the picture header Recovery points are indicated in the syntactic structure. CVS and bitstreams are G It can be started with DR picture. This means that the entire bitstream is one Intercoded pictures only, without intracoded pictures. This means that it can include. The main advantage of defining GDR support in this way is that G The goal is to provide operation that conforms to DR. GDR is an encoder that captures the entire picture. Instead of intracoding, multiple pictures are intracoded. The bitrate of the bitstream is smoothed by distributing chairs or blocks. This makes it possible to create wireless displays, online games, and more. Ultra-low latency applications, such as those based on 2chan, are becoming more common. Therefore, today, a significant reduction in end-to-end latency, which is considered more important than ever before, is possible. Make it Noh.

[0041] Another GDR-related feature in VVC is virtual boundary signaling. The refreshed area (i.e., the area between the camera and its recovery point) The boundary between the correctly decoded region and the unrefreshed region is a virtual boundary. Signal notification is permitted, and if a signal notification is received, in-loop filtering across boundaries will be applied. This is no longer applicable, and therefore, decoding inconsistencies occur in some samples near the boundary. This will no longer occur. This is because the application will not correctly decode the region during GDR processing. This could be useful if you decide to show it.

[0042] IRAP pictures and GDR pictures are combined into RAP (Random Accelerator) (ss point) This can be called a picture.

[0043] 3.3 Reference Picture Management and RPL (Reference Picture List) t) Reference picture management is required for any video coding scheme that uses interpretation. It is a functional feature. It is used with DPB (Decoded Picture Buffer). It manages the storage of reference pictures and the removal of reference pictures from there, and reference pictures Place them in the correct order within the RPL.

[0044] HEVC's reference picture management involves marking reference pictures and DPB (Decode). Removal from the ed Picture Buffer, and RPLC (Reference) This includes (e Picture List Construction), and is different from AVC. They are different. In AVC, MMCO (Memory M) adapts to the sliding window. References based on the addition of management control operations. Instead of a picture marking mechanism, HEVC uses what is known as RPS (Reference It defines a reference picture management and marking mechanism based on a Picture Set. As a result, RPLC is based on the RPS mechanism. RPS is a reference associated with a picture. It consists of a set of pictures, and all references preceding the associated pictures in the order they were decrypted. A collection of pictures, each associated with the corresponding picture in the order of decryption. It may be used for interpretation of any picture that follows. The reference picture set is... It consists of five lists of the picture. The first three lists are the interface of the current picture. It may be used in prediction, and in the decoding order, it follows the current picture by one or more. This includes all reference pictures that may be used in the interpretation of number pictures. The two lists are not used in the current picture interpretation, but in the decoding order. It can also be used in predicting the next one or more pictures following the current picture. It consists of all reference pictures. RPS is mainly used to improve fault tolerance in AVC. Instead of "intercoded" signal notifications like those in the case of DPB status, It provides "intra-coded" signal notifications. RPLC processing in HEVC is By signaling the index to the RPS subset of each reference index, It is based on RPS, and this process is simpler than RPLC processing in AVC.

[0045] Reference picture management in VVC is more similar to HEVC than AVC, but somewhat It is simple and robust. As in these standards, two RPLs, list0 And list1 is derived, which are the reference picture sets used in HEVC. This is not based on the concept of automatic sliding window processing used in AVC. , and is signaled more directly. The reference picture is the active entry for RPL. Listed as either an active or inactive entry, only active entries are currently It may also be used as a reference index in the interpretation of the current picture's CTU. Inactive entries refer to other pictures that arrive later in the bitstream. To that end, show other pictures that should be kept in the DPB.

[0046] 3.4 Parameter Set AVC, HEVC, and VVC specify parameter sets. Parameter set type This includes SPS, PPS, APS, and VPS. SPS and PPS support AVC and HEVC. It is supported in all VVC versions. VPS was introduced with HEVC. It is included in both HEVC and VVC. APS is included in either AVC or HEVC. It wasn't there before, but it's included in the text of the recent VVC draft.

[0047] SPS is designed to transmit sequence-level header information, while PPS is frequently Designed to carry picture-level header information that does not change. SPS and P Using Photoshop, it is necessary to repeat information that changes frequently for each sequence or picture. Therefore, redundant signal notifications of this information can be avoided. Furthermore, SPS and P Using PS enables out-of-band transmission of critical header information, thereby eliminating redundancy. This not only avoids the need for transmission but also improves error tolerance.

[0048] VPS is a sequence common to all layers of a multilayer bitstream. It was introduced to carry level header information.

[0049] APS requires a considerable amount of coding, and is done with multiple pictures. It is shared, and there can be a great many different variations in the sequence. It was introduced to carry picture-level or slice-level information.

[0050] 3.5 Defining Relationships in VVC The relevant definition in the recent VVC text (JVET-Q2001-vE / v15) is The following applies: Associated IRA picture (of a specific picture): Previous IRA in decoding order P-Picture (if it exists) has the same value nuh_layer_id as the specific picture. do. CRA (Clean Random Access) PU: Coated picture ya is a CRA picture PU. CRA (Clean Random Access) Picture: Each VCL NAL Unit An IRAP picture whose nal_unit_type is CRA_NUT. CVS (Coded Video Sequence): In the decoding order, all CVSS including subsequent AUs (but not subsequent AUs that are CVSS AUs) A sequence of AUs consisting of CVSS AUs, where zero or more non-AUs follow each other. CVSS(Coded Video Sequence Start)AU:CVS Each layer has a PU, and the coded picture of each PU is a CLVSS picture. ruAU. GDR (Gradual Decoding Refresh) AU: Each of these PUs An AU where the downloaded picture is a GDR picture. GDR (Gradual Decoding Refresh) PU: Coding The picture is a GDR picture. GDR (Gradual Decoding Refresh) Picture: NAL Uni A picture whose nal_unit_type is GDR_NUT. IDR (Instantaneous Decoding Refresh) PU: The printed picture is an IDR picture. IDR(Instantaneous Decoding Refreshu)Picchi Y: The nal_unit_type of each VCL NAL unit is IDR_W_RADL This is an IRAP picture that is IDR_N_LP. IRAP (Intra Random Access Point) AU:CVS each A PU exists in the layer, and the coded picture of each PU is an IRAP picture. AU. IRAP (Intra Random Access Point) PU: Code The picture that has been edited is an IRAP picture. IRAP (Intra Random Access Point) Picture: IDR_ Within the range from W_RADL to CRA_NUT, all VCL NAL units Coded pictures with the same _unit_type value. Reading picture: Located on the same layer as the associated IRAP picture, and associated The picture that precedes the IRAP picture in the output order. RADL(Random Access Decodable Leading)PU :PU where the coded picture is a RADL picture. RADL (Random Access Decodable Leading) pic Cha: The nal_unit_type of each VCL NAL unit is RADL_NUT. A picture. RASL(Random Access Skipped Leading)PU:Co The printed picture is a RASL picture (PU). RASL (Random Access Skipped Leading) Picture : The nal_unit_type of each VCL NAL unit is RASL_NUT Slurp. STSA(Step-wise Temporal Sublayer Access )PU: A PU where the coded picture is an STSA picture. STSA(Step-wise Temporal Sublayer Access )Picture: The nal_unit_type of each VCL NAL unit is STSA_NU A picture that is T. Note - STSA pictures have the same Temp for interpredictive references as STSA pictures. Do not use pictures with oralId. Temporal: Same as STSA picture. The subsequent picture of the STSA picture in the decoding order, which has an ID, is used for interpretation reference. STSA picture with the same TemporalId, in the decryption order of STSA picture Do not use the previous picture. STSA picture is STSA picture, STSA Enable the up switch from the sublayer directly below the picture for sublayers containing pictures. The TemporalId of an STSA picture must be greater than 0. Trailing picture: A non-IRAP picture following the associated IRAP picture in the output order. It's a Kucha, not an STSA picture. Note - Trailing pictures associated with IRAP pictures are also decrypted in the same order. It follows the IRAP picture. Following the associated IRAP picture in the output order, it decodes... Pictures that precede an associated IRAP picture in the order of creation are not permitted.

[0051] 3.6. NAL Unit Header Syntax and Semantics in VVC In recent VVC texts (JVET-Q2001-vE / v15), NAL Uni The header syntax and semantics are as follows:

[0052] 7.3.1.2 NAL Unit Header Syntax

[0053] [Table 1]

[0054] 7.4.2.2. Semantics of NAL Unit Headers The forbidden_zero_bit is assumed to be equal to 0. nuh_reserved_zero_bit is assumed to be equal to 0. A value of 1 for eserved_zero_bit will be determined by ITU-T|ISO / IEC in the future. It may be specified. The decoder is defined as nuh_reserved_zero_bit equal to 1. Ignore the NAL unit (i.e., remove it from the bitstream and discard it). nuh_layer_id is the identifier of the layer to which the VCL NAL unit belongs, This specifies the identifier of the layer to which the non-VCL NAL unit applies. nuh_laye The value of r_id should be within the range of 0 to 55. nuh_layer_id Other values ​​are reserved for future use by ITU-T|ISO / IEC. The value of nuh_layer_id is the VC of a single coded picture. It shall be the same for L NAL units. Coated picture or The nuh_layer_id value of the PU is the coded picture or the VC of the PU. This is the value of nuh_layer_id for the L NAL unit. The values ​​of nuh_layer_id for AUD, PH, EOS, and FD NAL units are as follows: The following constraints apply. -If nal_unit_type is equal to AUD_NUT, nuh_layer_ The ID is assumed to be equal to vps_layer_id[0]. - Alternatively, nal_unit_type is PH_NUT, EOS_NUT, FD_N If equal to UT, nuh_layer_id is the associated VCL NAL unit. Assume that it is equal to nuh_layer_id. Note 1 - nuh_layer_id of DCI, VPS, and EOB NAL units The value is not constrained.

[0055] The value of nal_unit_type is the same for all pictures in CVSS AU. It shall be considered as such. nal_unit_type is the NAL unit type as defined in Table 5. This defines the type of RBSP data structure included in the NAL unit. It falls within the scope of UNSPEC_28..UNSPEC_31, and its semantics are not defined. NAL units having nal_unit_type are decoded as defined herein. It should not affect the processing. Note 2 - NAL units within the range of UNSPEC_28...UNSPEC_31 The type may be used as determined by the application. In this specification, n The decryption process for these values ​​of al_unit_type is not specified. The application may use these NAL unit types for different purposes, Design of an encoder that generates a NAL unit with a nal_unit_type value. And the contents of NAL units having these nal_unit_type values Particular attention must be paid to the design of the decoder that interprets these values. This specification will not discuss these values. Management of these nal_unit_type values ​​is not defined. n (that is, the content of NAL units for the same nal_unit_type value) (Different definitions of the meaning of "tsu") A situation where something is not important, or is not possible, or is controlled, for example For example, in a control application or transport specification, or in a bitstream Use in a context defined or managed by controlling the distributed environment It may only be suitable in certain cases.

[0056] (As specified in Annex C) Determine the number of data in the bitstream's DU. For purposes other than this, the decoder will not use the reserved value of nal_unit_type. Ignore the content of all NAL units used (remove from bitstream) (These shall be disposed of.) Note 3 - This requirement allows for the future definition of extensions that conform to this specification.

[0057] Table 5 - NAL Unit Type Codes and NAL Unit Type Classes

[0058] [Table 2]

[0059] [Table 3]

[0060] Note 4 - CRA (Clean Random Access) pictures are bitstream The system may have associated RASL or RADL pictures. Note 5 - IDR(Ins) has a nal_unit_type equal to IDR_N_LP (Tantaneous Decoding Refresh) The picture is bitstream It does not have an associated reading picture in the frame. IDR_W_RADL An IDR picture with a nal_unit_type equal to the bitstream exists in the bitstream. Although there is no associated RASL picture, it is associated with the bitstream. It may have a RADL picture.

[0061] The value of nal_unit_type is all VCL NAL units in the subpicture. The same applies to the subpicture. The subpicture's VCL NAL unit is the same as the subpicture's. It is considered to have the NAL unit type. For any specific picture's VCL NAL units, the following applies: -If mixed_nalu_types_in_pic_flag is equal to 0, n The value of al_unit_type is the same for all VCL NAL units in the picture. It should be the same, and the picture or PU is coded as the picture or PU It is considered to have the same NAL unit type as the slice NAL unit. -Otherwise (mixed_nalu_types_in_pic_flag is 1 (equal to), the picture has at least two subpictures and the picture's VCL N The AL unit has exactly two different nal_unit_type values, as follows: It should be done. The VCL NAL unit of at least one subpicture of the picture is STSA_NUT, RADL_NUT, RASL_NUT, IDR_W_RADL, ID A specific value of nal_unit_type that is equal to R_N_LP or CRA_NUT While all of them are present, the VCL NAL units of other sub-pictures within the picture are TRA nal_unit_ equal to IL_NUT, RADL_NUT, or RASL_NUT It should have all different values ​​of type.

[0062] For single-layer bitstreams, the following constraints apply: -Each picture, in the order of decoding, is decoded, except for the first picture in the bitstream. It is thought to be associated with the previous IRAP picture in terms of order of transformation. -If the picture is the leading picture of the IRAP picture, then RADL or R Use ASL picture. -If the picture is a trailing picture of an IRAP picture, then RADL or R It should not be an ASL picture. -RASL pictures associated with IDR pictures are included in the bitstream. Assume they are not present. - IDR picture with nal_unit_type equal to IDR_N_LP The linked RADL pictures are not included in the bitstream. Note 6 - When each parameter set is referenced, (in the bitstream, or IRAP PU, insofar as it is available (by external means not specified herein) By discarding all preceding PUs (and all subsequent non-IRAP pictures and non-IRAP pictures) By correctly decoding the RASL picture in the decoding order, the IRAP PU location is run Dam access is possible. -In the decoding order, pictures preceding the IRAP picture are output in the order of IRAP picture Prior to the chat, in the output order, the RADL picture associated with the IRAP picture is It shall take precedence. -RASL pictures associated with CRA pictures are output in the order of CRA pictures. It shall precede the linked RADL picture. -RASL pictures associated with CRA pictures are decoded in the same order as CRA pictures. It is assumed that the IRAP picture preceding the chat will follow in the output order. -field_seq_flag is equal to 0, and the current picture is an IRAP picture. If it is equal to the associated reading picture, in the decoding order, the same IRAP P It shall precede all non-reading pictures associated with Kucha. If not, picA and picB are associated with the IRAP picture, respectively. In the decoding order, the first and last reading pictures are considered as follows: There is at most one non-reading picture preceding picA, and in the decoding order Assume there are no non-reading pictures between picA and picB.

[0063] nuh_temporal_id_plus1-1 is the temporal identifier of the NAL unit. This defines... The value of nuh_temporal_id_plus1 is not equal to 0. The variable TemporalId is derived as follows: TemporalId=nuh_temporal_id_plus1-1 (3 6) nal_unit_type is in the range of IDR_W_RADL~RSV_IRAP_12 If it is within the range, TemporalId is equal to 0. nal_unit_type is equal to STSA_NUT, vps_independ ent_layer_flag[GeneralLayerIdx[nuh_layer If _id]] is equal to 1, then TemporalId is not equal to 0. The value of TemporalId is the same for all VCL NAL units in the AU. Assume there is a coded picture, PU, ​​or AU with Temporal I The value of d is the VCL NAL unit of the coded picture, PU, ​​or AU. This is the value of TemporalId. The value of TemporalId in the sublayer representation is sub The maximum value of the TemporalId for all VCL NAL units in the layer representation. be.

[0064] The TemporalId value for non-VCL NAL units is constrained as follows: -nal_unit_type is DCI_NUT, VPS_NUT, VPS_NUT, Or, if it is equal to SPS_NUT, TemporalId is equal to 0, and NAL unit The TemporalId of an AU containing the element is assumed to be equal to 0. -Otherwise, if nal_unit_type is equal to PH_NUT, Tem The poralId is assumed to be the TemporalId of the PU, including the NAL unit. - Otherwise, nal_unit_type is EOS_NUT or EOB_NU If T is equal, then TemporalId is equal to 0. -Otherwise, nal_unit_type is AUD_NUT, FD_NUT, P If equal to REFIX_SEI_NUT or SUFFIX_SEI_NUT, Te The temporalId is assumed to be the temporalId of the AU containing the NAL unit. . -Otherwise, nal_unit_type is PPS_NUT, PREFIX_A If it is equal to PS_NUT or SUFFIX_APS_NUT, TemporalI d shall be greater than or equal to the TemporalId of the PU including the NAL unit. Note 7 - If the NAL unit is a non-VCL NAL unit, TemporalId The value is the TemporalId value of all AUs to which non-VCL NAL units apply. It is equal to the minimum value of nal_unit_type is PPS_NUT, PREFIX_AP If it is equal to S_NUT or SUFFIX_APS_NUT, then TemporalId This may be greater than or equal to the TemporalId of the encompassing AU, and all PPS and APS may be included at the start of the bitstream (for example, if they are transported out of band) If present, the receiver places them at the beginning of the bitstream. The selected picture has a TemporalId equal to 0.

[0065] 3.7. Syntax and Semantics of Picture Header Structure in VVC In recent VVC texts (JVET-Q2001-vE / v15), the present invention is most... The syntax and semantics of the related picture header structure are as follows:

[0066] 7.3.2.7 Picture Header Structure Syntax

[0067] [Table 4]

[0068] 7.4.3.7 Picture Header Structural Semantics A PH syntactic structure is all the coded pictures associated with the PH syntactic structure. The slices contain common information. If gdr_or_irap_pic_flag is equal to 1, the current picture is GDR. This specifies that it is an IRAP picture. gdr_or_irap_pi equals 0 c_flag can be used regardless of whether the current picture is a GDR or IRAP picture. It stipulates that... If gdr_pic_flag is equal to 1, the picture associated with PH is a GDR pic. Defines that it is a chat. gdr_pic_flag equal to 0 is associated with PH. Specifies that the selected picture is not a GDR picture. If it does not exist, gdr_pic The value of _flag is presumed to be equal to 0. gdr_enabled_flag is equal to 0 If not, the value of gdr_pic_flag will be equal to 0. Note 1 - gdr_or_irap_pic_flag is equal to 1, gdr_pic_f If lag is equal to 0, the picture associated with PH is an IRAP picture. ...

[0069] ph_pic_order_cnt_lsb is the picture order for the current picture. Defines the dacount modulo MaxPicOrderCntLsb. ph_pic_o The length of the rder_cnt_lsb syntax elements is log2_max_pic_order_ cnt_lsb_minus4+4 bits. ph_pic_order_cnt_ The value of LSB should be within the range of 0 to MaxPicOrderCntLsb-1. no_output_of_prior_pics_flag is specified in Annex C. Therefore, after decoding the CLVSS picture that is not the first picture in the bitstream, This affects the output of the previously decoded picture in DPB. recovery_poc_cnt is the recovery of the decrypted picture in the output order. Defines the vari point. The current picture is a GDR picture associated with PH. , the current GDR picture's PicOrderCntVal is recovery_poc_ In a CLVS that has PicOrderCntVal, which is the value of cnt added to the CLVS If, in the decoding order, there is a picture picA that follows the current GDR picture, pic Picture A is called the recovery point picture. Otherwise, the current picture P The value obtained by adding the value of recovery_poc_cnt to icOrderCntVal is The first picture in the output order, which also has a large PicOrderCntVal, This is called a recovery point picture. The recovery point picture is currently in the order of decryption. It should not precede the GDR picture. The value of recovery_poc_cnt is The range is from 0 to MaxPicOrderCntLsb-1.

[0070] If the current picture is a GDR picture, then the variable RpPicOrderCntVal This is derived as follows: RpPicOrderCntVal=PicOrderCntVal+recove ry_poc_cnt (81) Note 2 - gdr_enabled_flag is equal to 1, and the current picture's PicOr RpPicOrderCntVal of GDR picture associated with derCntVal If l is greater than or equal to l, the current and subsequent decoded pictures in the output order are in the decoded order. The IRAP picture preceding the associated GDR picture (if any) ) perfectly matches the corresponding picture generated by starting the decryption process from ) . ...

[0071] 3.8. Constraints of RPL in VVC In recent VVC textbooks (JVET-Q2001-vE / v15), VVC... The constraints of the RPL are as follows (see VVC clause 8.3.2 picturelis (As part of the decryption process for constructing the data.)

[0072] 8.3.2 Decryption process for constructing the reference picture list ... If each i is equal to 0 or 1, the first NumR in RefPicList[i] The efIdxActive[i] entry is an active entry in RefPicList[i] These are called IV entries, and other entries in RefPicList[i] are RefP These are referred to as inactive entries in icList[i]. Note 2 - A specific picture is an entry in RefPicList[0] and Ref It may be referenced by both entries in PicList[1]. A particular picture can be found through multiple entries in RefPicList[0], or It may also be referenced by multiple entries in RefPicList[1]. Note 3 - Active entries in RefPicList[0] and RefPicList [1] The active entry is the current picture and the current picture in the decoding order. All reference pictures that may be used for interpretation of one or more pictures that follow Referencing all entries together. Inactive entries and Re in RefPicList[0] Inactive entries in fPicList[1] are for predicting the current picture's interface. It is not used for decoding, but one or more pictures that follow the current picture in the decoding order. This references all the reference pictures that may be used in the interpretation for the purpose of [the prediction]. Note 4 - RefPicList[0] or RefPicList[1] contains the corresponding Since the picture does not exist in the DPB, there is one or more entries that are equivalent to "No reference picture". There may be multiple. For RefPicList[0] equal to "no reference picture" or each non-active entry in RefPicList[0] should be ignored For each active entry in RefPicList[0] equal to "no reference picture" or RefPicLis t[1], an unintentional picture loss should be estimated. Yes.

[0073] The requirements for bitstream compliance are that the following constraints apply. - When each i is equal to 0 or 1, num_ref_entries[i][Rpls Idx[i]] must not be less than NumRefIdxActive[i]. Each active entry in RefPicList[0] or RefPicList[1] The picture referenced by the entry is included in the DPB and is less than or equal to the Tempor alId of the current picture. - Each entry in RefPicList[0] or RefPicList[1] The picture referenced by the entry is not the current picture and has a non_referen ce_picture_flag equal to 0. - The STRP entry in RefPicList[0] or RefPicList[1] of a picture slice and the LTRP entry in RefPicList[0] or RefPicList[1] of the same slice or a different slice of the same picture shall not refer to the same picture. - The difference between the PicOrderCntVal of the current picture and the PicOrderCntVal of the picture referenced by the entry is 2 or more for RefPicList picture 24 ​In 0] or RefPicList[1], assume there is no LTRP entry. - setOfRefPics is the set of all entries in RefPicList[0] that have the same nuh_layer_id as the current picture and all entries in RefPicList[1 that have the same nuh_layer_id as the current picture, plus the set of unique pictures referenced thereby. The number of pictures in setOfRefPics must be no more than MaxDpbSize - 1, where MaxDpbSize is , as defined in Clause A.4.2, and setOfRefPics is the same for all slices of the picture.

[0074] - When the current slice has a nal_unit_type equal to STSA_NUT , assume that there is no active entry in RefPicList[0] or RefPicL ist[1] whose TemporalId is equal to that of the current picture and whose nuh_layer_ id is equal to that of the current picture. - If the current picture is a picture that follows an STSA picture in the decoding order and whose TemporalI d is equal to that of the current picture and whose nuh_layer_id is equal to that of the current picture, assume that there is no picture that precedes the STSA picture in the decoding order and whose TemporalId is equal to that of the current picture and whose n uh_layer_id is equal to that of the current picture and that is included as an active entry in RefPicList[0] or Ref PicList[1]. - If the current picture is a CRA picture, RefPicList[0] or Ref PicList[1] The pictures referenced by the entries in PicList[1], in the order of decoding In the preceding IRAP picture (if any), in the output order or decoding order, Assume there are no pictures.

[0075] -If the current picture is a trailing picture, then RefPicList[0] or the picture referenced by the active entry in RefPicList[1] Therefore, the reference picture is unavailable because it is an IRAP picture associated with the current picture. Assume that no pictures are generated by the decoding process used to produce the result. - The current picture is related to the same IRAP picture in both the decoding order and the output order. If it is a trailing picture following one or more attached leading pictures , referenced by entries in RefPicList[0] or RefPicList[1] A picture that is used for the IRAP picture associated with the current picture. There are no pictures generated by the decoding process that creates unusable reference pictures. Let's assume that. - Is the current picture a recovery point picture, or is it a recovery point picture in the output order? If it is a picture following a repoint picture, the GDR of the recovery point picture Generated by a decoding process to create a reference picture that is unavailable for the picture. The contents of RefPicList[0] or RefPicList[1] that contain the picture There will be no birds.

[0076] -If the current picture is a trailing picture, in output order or decoding order , RefPicList[0] or Ref preceding the associated IRAP picture Assume that there is no picture referenced by the active entry of PicList[1]. . - If the current picture is a trailing picture following one or more leading pictures associated with the same IRAP picture in both the decoding order and the output order then assume that there is no picture referenced by an entry in Re fPicList[0] or RefPicList[1] that precedes the associated IRAP picture in the output order or the decoding order. - Assume that there is no active entry in RefPicList[0] or RefPicList[1] that corresponds to any of the following when the current picture is a RADL picture. - If the current picture is a RADL picture, then assume that there is no active entry in RefPicList[0] or RefPicList[1] that corresponds to any of the following. ○ RASL picture ○ A picture generated by the decoding process to produce an unavailable reference picture picture ○ A picture that precedes the associated IRAP picture in the decoding order

[0077] - Assume that each picture referenced by an ILRP entry in RefPicList[0] or RefPicList[1] of the slice of the current picture is within the same AU as the current picture. - Assume that each picture referenced by an ILRP entry in RefPicList[0] or RefPicList[1] of the slice of the current picture is present in the DPB and has a nuh_layer_id smaller than that of the current picture. - Assume that each ILRP entry in RefPicList[0] or RefPicList[1] of the slice is an active entry. ... ​​​​

[0078] 4. Technical problems addressed by the disclosed technical solutions Existing designs in recent VVC texts (JVET-Q2001-vE / v15) are It has the following problems: 1) The definition of an associated IRAP picture is the associated IR of a particular picture. The AP picture should be updated so that it belongs to the same layer as a specific picture. 2) The current definition of a trailing picture is as follows: Trailing picture: In the output order, following the associated IRAP picture, , not an STSA picture, but a non-IRAP picture. Therefore, the bitstream contains a trailing picture, which is why the IRAP picture A chat is required, and if there is no IRAP picture in the bitstream, the NAL unit The IP value TRAIL_NUT cannot be used. However, the non- STSA pictures must use the NAL unit type value TRAIL_NUT. . 3) Existing constraints on the output order of the picture preceding the IRAP picture in the decoding order are: It must be specified that it applies only to pictures within the layer. 4) GDR picture and relative to the picture before and after it in the decoding order. There are no constraints on the decoding order or output order. 5) Pictures associated with IRAP pictures and some non-reading pictures Existing constraints on the decoding order are specified to apply only to pictures within a layer. It is necessary.

[0079] 6) Currently, reading pictures and RADL pictures are associated with GDR pictures. RASL pictures are not supported. 7) Existing constraints on RPL for CRA pictures are only for pictures within layers. It needs to be stipulated that it should apply. 8) Trailing pictures associated with STSA pictures, GDR pictures, and GDR picture where NoOutputBeforeRecoveryFlag is equal to 0 In this case, the RP is not generated by the decoding process that generates an unavailable reference picture. The constraint on L's active entry is missing. 9) STSA picture, IDR picture, NoOutputBeforeRecover If the ryFlag is equal to 0, such as in a CRA picture, an unavailable reference picture will be generated. The constraints on RPL entries that are not generated by the decryption process are missing. 10) In the case of STSA pictures, the associated IR in the output order or decoding order. The constraint on active entries in RPL is missing so as not to precede AP pictures. They are doing it. 11) In the case of STSA pictures, the associated IR in the output order or decoding order. The constraint on entries in the RPL is missing to prevent them from preceding the AP picture.

[0080] 5. Examples of Implementations and Technical Solutions To solve the above-mentioned problems, the following methods are disclosed. The present invention is generally It should be considered an example to explain a general concept, and should not be interpreted in a narrow sense. Furthermore, the present invention may be applied individually or in any combination. stomach. 1) To solve Problem 1, the definition of an associated IRAP picture is a specific picture. The IRAP picture associated with the chat belongs to the same layer as the specific picture. It will be updated. 2) To solve problem 2, update the definition of trailing picture and trailing picture Kucha may be associated with GDR pictures. a. In addition, add the definition of the associated GDR picture and the associated IRA The definition of P-Picture has been updated, and each of the layers other than the first picture in the bitstream In the decoding order, among the IRAP or GDR pictures on the same layer prior to the picture Identify the closest possible match to the one that matches. b. In addition, the trailing picture is associated with an IRAP or GDR picture. A constraint is added requiring that the output order be followed.

[0081] 3) In order to solve problem 3, the picture that precedes the IRAP picture in the decoding order The existing constraints regarding the output order have been updated to only impose restrictions on pictures within layers. do. a. In one example, this constraint is defined as follows: nuh_layerId In the decoding order, the IRAP picture having a nuh_layer_id equal to the preceding one Any picture that has a nuh_layer_Id equal to a specific value layerId is In the output order, the IRAP picture and its associated RADL picture precede the IRAP picture. do. 4) To solve Problem 4, add one or more of the following constraints. a. Trailing pictures are output in the same order as the associated IRAP or GD. It follows R-Picture. For a GDR picture with a nuh_layer_id equal to b.layerId, In the decoding order, the preceding nuh_layer_id is equal to a specific value, layerId. Any picture you have will be output in the GDR picture and its associated picture in the output order. It precedes Kucha.

[0082] 5) To solve problem 5, IRAP pictures and some non-reading pictures Existing constraints on the decoding order of associated pictures can be applied to pictures within a single layer. Update it to only impose restrictions. a. In one example, this constraint is defined as follows: field_seq_ flag is equal to 0 and nuh_layer_id is equal to a specific value layerId If the current picture is the reading picture associated with the IRAP picture In total, this applies to all non-reading pictures associated with the same IRAP picture. In the decoding order, it shall take precedence. Otherwise, picA and picB shall be treated as follows: In each case, the first and last read data associated with the IRAP picture in the decoding order. Assuming it is a picture, in the decoding order it precedes picA and is equal to layerId. There is at most one non-reading picture with the nuh_layer_id, and, In the decoding order, the nuh_laye between picA and picB is equal to the layerId. Assume that no non-reading pictures with an r_id exist. b. In another example, this constraint is defined as follows: field_seq_ The flag is equal to 0, and the current picture is associated with the IRAP picture. If it is equal to the gpicture, it is associated with the same IRAP picture in the decoding order. It shall precede all non-reading pictures. Otherwise, pic A and picB, respectively, are associated with the IRAP picture in the decoding order. Assuming the first and last reading picture, the decoding order precedes picA. There is at most one non-reading picture associated with the IRAP picture, and In the decoding order, the IRAP picture associated with picA and picB Assume that there are no non-reading pictures.

[0083] 6) To solve problem 6, the reading picture associated with the GDR picture Defines and specifies RADL pictures and RASL pictures. a. The reading picture associated with the GDR picture is G in the decryption order. Following the DR picture, this picture precedes it in the output order. b. RADL pictures associated with GDR pictures are associated with GDR pictures. The reading picture is a nal_unit_t that is equal to RADL_NUT. It has a type. c. RASL pictures associated with GDR pictures are associated with GDR pictures. The reading picture is a nal_unit_t that is equal to RASL_NUT. It has a type.

[0084] 7) To solve problem 7, existing constraints regarding the RPL of CRA pictures are layered Update to impose restrictions only on internal pictures. a. In one example, this constraint is defined as follows: The current picture has a nuh_layer_id that is equal to a specific value, layerId. If present and it is a CRA picture, in the output order or decoding order, lay in the decoding order. A preceding IRAP picture with a nuh_layer_id equal to erId (if one exists) The preceding entry in RefPicList[0] or RefPicList[1] Assume that no picture referenced by the bird exists.

[0085] 8) In order to solve problem 8, the following constraints are defined. The current picture has a nuh_layer_id that is equal to a specific value layerId. CRAP with NoOutputBeforeRecoveryFlag equal to 1 RASL picture associated with Kucha, NoOutputBeforeR equal to 1 GDR picture with recoveryFlag, or NoOutput equal to 1 num_layer equal to BeforeRecoveryFlag and layerId If it is not a restored picture of a GDR picture with an _id, an unavailable reference picture will be generated. The RefPicList[0] or RefPi generated during the decoding process to achieve this The active entry in cList[1] does not refer to any picture. It is determined.

[0086] 9) In order to solve problem 9, the following constraints are defined. The current picture has a nuh_layer_id that is equal to a specific value layerId. CRAP with NoOutputBeforeRecoveryFlag equal to 1 In the decoding order, NoOutputBeforeRecovery is equal to 1. Preceding a reading picture associated with the same CRA picture that has the Flag Picture, with NoOutputBeforeRecoveryFlag equal to 1 NoOutputB is the reading picture associated with the CRA picture, equal to 1. A GDR picture with eforeRecoveryFlag, or No equal to 1. OutputBeforeRecoveryFlag is equal to layerId (nuh) Unavailable reference if it is not a restored picture of a GDR picture that has _layer_id RefPicList[0] generated during the decoding process for generating the picture, This means that there are no pictures referenced by the entries in RefPicList[1]. It is defined.

[0087] 10) In order to solve problem 10, the following constraints are defined. The current picture is associated with the IRAP picture, and the output order is IRAP picture If it follows a chat, in the output order or decoding order, it precedes the associated IRAP picture. The active element of RefPicList[0] or RefPicList[1] is executed. Assume there are no pictures referenced by the content. 11) In order to solve problem 11, the following constraints are defined. The current picture is associated with the IRAP picture, and in the output order, the IRAP picture is... Following this, and in both the decoding order and the output order, the same IRAP picture is associated. If it follows a reading picture, in the output order or decoding order, associated The IRAP picture precedes RefPicList[0] or RefPicLis Assume there are no pictures referenced by the entry in t[1].

[0088] 6. Embodiments The following are some illustrative examples of some embodiments of the inventions summarized in Chapter 5 above. This is a configuration that can be applied to VVC specifications. The modified text is from JVET-Q2001. -Based on the latest VVC text in vE / v15. Includes any additions or corrections already made. Relevant parts are highlighted in bold italics, and one of the deleted parts is also highlighted. Parts are marked with double brackets (for example, [[a]] means the character "a" is removed). (As shown). There may be other unemphasized changes, as it is inherently editable.

[0089] 6.1. First Embodiment This embodiment corresponds to sections 1, 2, 3, 4, 5, and 5a.

[0090] 3 Definition ...

[0091]

number

[0092] 7.4.2.2. Semantics of NAL Unit Headers ...

[0093]

number

[0094]

number

[0095] 7.4.3.7 Picture Header Structural Semantics ...

[0096]

number

[0097] [[If the current picture is a GDR picture, the variable RpPicOrderCntVa l is derived as follows: RpPicOrderCntVal=PicOrderCntVal+recove ry_poc_cnt (81)]] Note 2 - gdr_enabled_flag is equal to 1, and the current picture's PicOr recoveryPointPo of GDR picture associated with derCntVal If cVal[[RpPicOrderCntVal]] or greater, the current output order The current and subsequent decoded pictures are associated with the GDR pictures in the decoding order. By starting the decoding process from the IRAP picture that precedes the previous one (if one exists), It perfectly matches the corresponding picture that was generated. ...

[0098] 8.3.2 Decryption process for constructing the reference picture list ...

[0099]

number

[0100]

number

[0101]

number

[0102] -If the current picture is a RADL picture, RefPicList[0] or RefPicList[1] does not contain any active entries that fall under any of the following categories. Let's assume that. ○RASL Pictures [[Picture generated by the decryption process to generate an unavailable reference picture ]] ○In the decoding order, the picture preceding the associated IRAP picture -...

[0103] Figure 1 shows an exemplary image processing system 1 in which various technologies disclosed herein may be implemented. This is a block diagram of 900. Various implementation forms are components of System 1900. It may include part or all of the following. System 1900 is used to receive video content. It may also include input 1902. The video content is in raw or uncompressed format. For example, it may be received as an 8 or 10-bit multicomponent pixel value, or It may be received in a compressed or encoded format. Input 1902 is network Represents a bus interface, peripheral bus interface, or storage interface. This is also good. Examples of network interfaces include Ethernet (registered trademark), PON (registered trademark). Registered trademark; wired interface such as Passive Optical Network , and wireless interfaces such as Wi-Fi (registered trademark) or cellular interfaces Includes -.

[0104] System 1900 implements the various coding or encoding methods described herein. It may include a coding component 1904 that can be installed. Component 1904 codes the average bitrate of the video from input 1902. It can be reduced to the output of component 1904 and generate a coded representation of the video. Therefore, coding techniques are called video compression or video transcoding techniques. This can happen. The output of coding component 1904 is component 190 As represented by 6, it may be stored, or transmitted via connected communication. Good. Bitstance of the video received, stored or transmitted in input 1902. The REEM (or coded) representation is used by Component 1908. Then, generate pixel values ​​or displayable images to be sent to the display interface 1910. That's fine too. The process of generating video that the user can see from a bitstream representation is... It is sometimes called image unfolding. Furthermore, a specific image processing operation is called a "coding" operation. It is called a tool, but the coding tool or operation is the encoder and its corresponding The decoding tool or operation that reverses the coding result is performed by the decoder. This will be understood.

[0105] Examples of peripheral bus interfaces or display interfaces include USB® (registered trademark). Universal Serial Bus) or HDMI (registered trademark; High-Definition Digital) (Finition Multimedia Interface) or DisplayPort It may also include things like . An example of a storage interface is SATA (Serial Advanced Technology Attachment), PCI, IDE Including interfaces, etc. The technologies described herein are for mobile phones, laptops, etc. Smartphones, or other devices capable of performing digital data processing and / or image display, etc. This may be implemented in various electronic devices.

[0106] Figure 2 is a block diagram of the image processing device 3600. The device 3600 is described herein. It may be used to implement one or more of the methods. The device 3600 is a smart Phones, tablets, computers, IoT (Internet of Things) It may be implemented in a signaling device, etc. The device 3600 has one or more processors 3602 and 1 Alternatively, it may include multiple memory 3604 and video processing hardware 3606. Alternatively, multiple processors 3602 implement one or more of the methods described herein. It may be configured as follows: One or more memories 3604 are as described herein. Used to store data and code used to implement laws and technologies This is also acceptable. The video processing hardware 3606 uses the technology described herein in hardware. It may be used for implementation in the road.

[0107] Figure 4 shows an exemplary video coding system 100 that may utilize the techniques of this disclosure. This is a lock diagram.

[0108] As shown in Figure 4, the video coding system 100 consists of a source device 110 and a transmission device. The transmitting device 110 may include a receiving device 120 and a source device 110 which encodes video data It generates data and can also be called a video encoding device. Destination device 120 The encoded video data generated by the source device 110 may be decoded, and the video It can also be called a decoding device.

[0109] The source device 110 includes a video source 112, a video encoder 114, and an input / output (I / O) Interface 116 may be included.

[0110] Video source 112 is a source such as a video capture device, video content provider An interface for receiving video data from a device, and / or generating video data. A computer graphics system to achieve this, or a combination of these sources. It may include. The video data may contain one or more pictures. Video encoding D114 encodes the video data from the video source 112 and generates a bitstream. A bitstream is a sequence of bits that form the coded representation of video data. It may include Kens. The bitstream contains coded pictures and associations It may include the data that has been cut. Coated pictures are the coding of the picture. This is a censored expression. The associated data is the sequence parameter set, picture. It may include parameter sets and other syntactic structures. I / O interface 116 This may include a modem and / or transmitter. Encoded video data The destination device is transmitted via network 130a and I / O interface 116. The encoded video data may be sent directly to the chair 120. For access purposes, the data may be stored on the recording medium / server 130b.

[0111] The destination device 120 has an I / O interface 126, a video decoder 124, and The display device 122 may also be included.

[0112] The I / O interface 126 may include a receiver and / or a modem. The O interface 126 is either the source device 110 or the storage medium / server 130b The encoded video data may be obtained. The video decoder 124 receives the encoded video The data may be decoded. The display device 122 displays the decoded video data to the user. The display device 122 may be integrated with the destination device 120, or it may be external. Outside the destination device 120 which is configured to interface with the display device That's fine.

[0113] The video encoder 114 and video decoder 124 use HEVC (High Efficiency). iency Video Coding) standard, VVC (Versatile Video Coding) standard, o Coding) standards and other current and / or further conventions, etc., video compression standards Therefore, it may work.

[0114] Figure 5 is a block diagram showing an example of a video encoder 200, and this video encoder 2 00 may be the video encoder 114 in the system 100 shown in Figure 4.

[0115] The video encoder 200 is configured to perform any or all of the technologies of this disclosure. This may also be the case. In the example in Figure 5, the video encoder 200 includes multiple functional components. The techniques described in this disclosure are shared among the various components of the video encoder 200. In some examples, the processor may be any of the technologies described herein. It may be configured to do everything.

[0116] The functional components of the video encoder 200 are a splitting unit 201, a mode selection unit 203, A prediction unit which may include a motion estimation unit 204, a motion compensation unit 205, and an intra-prediction unit 206. 202, residual generation unit 207, conversion unit 208, quantization unit 209, inverse quantization unit 210, inverse conversion It includes a section 211, a reconstruction section 212, a buffer 213, and an entropy coding section 214. But that's fine.

[0117] In other examples, the video encoder 200 has more, fewer, or different functions. Components may be included. In one example, the prediction unit 202 is IBC (Intra It may include a Block Copy section. The IBC section contains at least one reference picture. In IBC mode, where the current video block is located, predictions can be made. can.

[0118] Furthermore, several components, such as the motion estimation unit 204 and the motion compensation unit 205, They may be highly integrated, but for illustrative purposes, they are represented separately in the example in Figure 5. .

[0119] The division section 201 may divide the picture into one or more video blocks. The coder 200 and video decoder 300 support various video block sizes. good.

[0120] The mode selection unit 203, for example, based on the result of an error, selects an intra or intercom Select one of the coding modes and obtain the intra or interconnect The generated blocks are supplied to the residual generation unit 207, which generates residual block data and then reconstructs the data. Even if you supply it to 12 and reconfigure the encoded block to use as a reference picture Good. In some examples, the mode selection unit 203 uses an interprediction signal and an intra CIIP (Combination of Intra You may also select the Inter Prediction mode. The selection unit 203, in the case of interpretation, determines the resolution of the block's motion vector (e.g., sub-pitch). You may choose to use either xels or integer pixel precision.

[0121] To perform interpretation on the current video block, the motion estimation unit 204 buff By comparing one or more reference frames from a213 with the current video block, Motion information may be generated for the current video block. Motion compensation unit 205 generates motion information for the current video Picture movement information from buffer 213 other than the picture associated with the image block. And based on the decoded sample, predict the video block for the current video block. You may make a judgment.

[0122] The motion estimation unit 204 and the motion compensation unit 205 assume that the current video block is an I-slice. For example, depending on whether it is a P-slice or a B-slice, the current video block You may perform different actions on the buck.

[0123] In some examples, the motion estimation unit 204 makes a unidirectional prediction for the current video block. The motion estimation unit 204 then performs the following for the current video block: The reference picture may be searched for in 0 or List 1. Then the motion estimation unit 204 searches for the reference The image block and the movement showing the spatial displacement between the current image block and the reference image block. A reference index that shows the reference picture in List 0 or List 1, including the vector. A s may be generated. The motion estimation unit 204 uses a reference index, a prediction direction indicator, The motion vector may also be output as motion information for the current video block. Motion compensation Section 205, based on the reference video block indicated by the motion information of the current video block, You may also generate predicted image blocks for the blocks.

[0124] In other examples, the motion estimation unit 204 may also predict the current video block in both directions. The motion estimation unit 204 finds the current video block from the reference pictures in list 0. You can also search for a reference video block to do so, or among the reference pictures in List 1 You can then search for another reference video block to determine the current video block. The motion estimation unit 204 uses the reference pitch in List 0 and List 1, which include the reference video block. A reference index indicating the block, and the space between the reference video block and the current video block. A motion vector indicating the displacement may be generated. The motion estimation unit 204 uses the current video block The reference index and motion vector of the block are output as motion information for the current video block. It is also possible. The motion compensation unit 205 uses the motion information of the current video block to determine the reference video block. Based on this, a predicted video block may be generated for the current video block.

[0125] In some cases, the motion estimation unit 204 processes motion information for the decoder's decoding process. You may output the full set.

[0126] In some cases, the motion estimation unit 204 uses the full set of motion information for the current video. Output is not necessary. Rather, the motion estimation unit 204 refers to the motion information of another video block. The motion information of the current video block may be notified as a signal. For example, motion estimation unit 204 This is because the motion information of the current video block is sufficiently similar to the motion information of neighboring video blocks. It may be determined that this is the case.

[0127] In one example, the motion estimation unit 204 uses the syntactic structure associated with the current video block. In this context, the current video block possesses the same motion information as another video block. The value shown in -300 may also be shown.

[0128] In another example, the motion estimation unit 204 uses the syntactic structure associated with the current video block. In this case, another video block and motion vector difference (MVD; Motion Vector The difference may be identified as the current video block. This shows the difference between the motion vector of the block and the motion vector of the displayed video block. -Da300 uses the motion vector and motion vector difference of the indicated video block The motion vector of the current video block may be determined.

[0129] As described above, the video encoder 200 may predictively signal motion vectors. Two examples of predictive signal notification techniques that can be implemented by the video encoder 200 are shown in AMV. P (Advanced Motion Vector Prediction) and M Includes notification of the call mode signal.

[0130] The intra prediction unit 206 may perform intra prediction for the current video block. When the intra prediction unit 206 predicts the current video block, the intra prediction unit 2 06 is based on decoded samples of other video blocks in the same picture, Predictive data may be generated for the current video block. The measurement data may include predicted image blocks and various syntactic elements.

[0131] The residual generation unit 207 generates a predicted image block from the current image block. By subtracting the lock (e.g., indicated by a minus sign), the current Residual data may be generated for the video block. The current residual data for the video block is , the residual video block corresponding to different sample components of the sample in the current video block It may include "ku".

[0132] In other examples, for instance, in skip mode, residuals for the current video block Data is not required, and the residual generation unit 207 does not need to perform subtraction calculations.

[0133] The conversion processing unit 208 assigns one residual video block to the current video block. This involves applying multiple transformations to one or more transformations for the current video block. You may generate a coefficient image block.

[0134] The conversion processing unit 208 generates a conversion coefficient video block associated with the current video block. After that, the quantization unit 209 performs one or more quantizations associated with the current image block. Based on the parameter (QP: Quantization Parameter) value, The transformation coefficients associated with the existing video blocks may be quantized.

[0135] The inverse quantization unit 210 and the inverse conversion unit 211 perform inverse quantization and inverse conversion on the conversion coefficient image block. Alternatively, the transformations may be applied to each component, and the residual image block may be reconstructed from the transformed image block. The reconstruction unit 212 selects from one or more predicted video blocks generated by the prediction unit 202. The reconstructed residual image block is added to the corresponding sample and associated with the current block. The reconstructed video blocks may be generated and stored in buffer 213.

[0136] After the reconstruction unit 212 reconstructs the video block, the video blocking in the video block Loop filtering may be performed to reduce artifacts.

[0137] The entropy coding unit 214 is configured to separate the other functional components of the video encoder 200 from the other functional components of the video encoder 200. Data may be received. When the entropy coding unit 214 receives data, the entropy coding unit The P-coding unit 214 performs one or more entropy coding operations, and the entropy code Generates encoded data and a bitstream containing entropy-encoded data. You may output it.

[0138] Figure 6 is a block diagram showing an example of the video decoder 300, and the video decoder 300 is This could also be the video decoder 114 in the system 100 shown in Figure 4.

[0139] The video decoder 300 is configured to perform any or all of the technologies of this disclosure. This is also acceptable. In the example in Figure 6, the video decoder 300 includes multiple functional components. The techniques described herein are shared among the various components of the video decoder 300. In some examples, the processor may be any of the technologies described in this disclosure or It may be configured to perform all of these actions.

[0140] In the example shown in Figure 6, the video decoder 300 includes an entropy decoding unit 301 and a motion compensation unit. 302, intra prediction unit 303, inverse quantization unit 304, inverse transformation unit 305, and reconstruction unit 3 06, and buffer 307. The video decoder 300, in some examples, includes video E Even if you perform a decoding path that is almost the reverse of the encoding path described for coder 200 (Figure 5), stomach.

[0141] The entropy decoding unit 301 extracts the encoded bitstream. The bitstream is entropy-encoded video data (for example, video data It may include an encoded block. The entropy decoding unit 301 is an entropy Encoded video data is decoded, and motion interpolation is performed from the entropy-decoded video data. The compensation unit 302 includes motion vector, motion vector precision, reference picture list index, and Motion information including other motion information may be determined. The motion compensation unit 302 may, for example, AM This information can also be determined by performing VP and merge modes.

[0142] The motion compensation unit 302 may generate motion-compensated blocks, and in some cases, compensate Interpolation is performed based on the inter-filter. The interpolation filter used has sub-pixel precision. The identifier for this may be included in the syntactic element.

[0143] The motion compensation unit 302 is used by the video encoder 200 during the encoding of the video block. An interpolation filter is used to calculate interpolation values ​​for sub-integer pixels of the reference block. It is also possible that the motion compensation unit 302 adjusts the video encoder 200 based on the received syntax information. You can determine the interpolation filter to use and generate prediction blocks using the interpolation filter. stomach.

[0144] The motion compensation unit 302 processes the frames and / or slices of the encoded video sequence. Syntax information for determining the size of the block used to encode, encoded This describes how each macroblock of the picture in the video sequence is divided. Division information, a mode indicating how each division is encoded, and each inter-encoded block One or more reference frames (and reference frame lists) for a given block, and encoding Some of the other information may be used to decode the video sequence.

[0145] The intra prediction unit 303, for example, receives an intra prediction model in the bitstream. A prediction block may be formed from spatially adjacent blocks using a code. Inverse quantum The processing unit 303 is provided to the bitstream and decoded by the entropy decoding unit 301. The quantized image block coefficients are inversely quantized. The inverse transformation unit 303 performs the inverse transformation appropriately. To use.

[0146] The reconstruction unit 306 combines the residual block with the motion compensation unit 202 or the intra prediction unit 303. Therefore, even if the corresponding predicted block that was generated is added together to form a decoded block Good. If desired, decrypt the blocks to remove block artifacts. A deblocking filter may be applied to filter out the decoded image. The image block is stored in buffer 307, and buffer 307 is used for subsequent motion compensation / input A reference block is provided for prediction, and the image is decoded for display on the display device. Generates.

[0147] Next, we will list some examples of embodiments that are suitable for this purpose.

[0148] The following items illustrate exemplary embodiments of the technology described in the previous chapter. (For example, item 1) shows an exemplary embodiment of the technology described.

[0149] 1.1 A video having one or more video layers including multiple video pictures, and This includes performing a conversion between the coded representation of the image (3002), and the coding The displayed representation is an intra-random access point picture of the second picture. The first and second video pictures are constrained to belong to the same video layer. A video processing method (for example, method 3000 shown in Figure 3) configured in accordance with the rules that define the .

[0150] The following items illustrate exemplary embodiments of the technology described in the previous chapter (for example, item 2).

[0151] 2.1 A video having one or more video layers including multiple video pictures, and This includes performing a conversion between the coded representation of the image and the coded table Currently, the first type of picture is a Cody, which is an intra-random access point. Trailing pictures in the rendered representation also use progressively decoded refresh pictures. A format that allows being associated with a second type of picture, including A video processing method that conforms to the rules.

[0152] 3. The formatting rules are as follows: For each layer, the first layer in the bitstream Each picture in layers other than the picture layer, in the decoding order, is the previous intra layer of the same layer. The closer of the random access point or the progressive decoder refresh picture A method of item 2, which further specifies that it is to be specified to be associated with a particular person.

[0153] The following items illustrate exemplary embodiments of the technology described in the previous chapter (for example, item 3).

[0154] 4.1 A video having one or more video layers including multiple video pictures, and This includes performing a conversion between the coded representation of the image and the coded table Currently, the output order of the picture precedes the intra-random access point in the decoding order. The constraints on this are defined to apply only to pictures within the same video layer in the same output order. A video processing method that conforms to the formatting rules.

[0155] 5. The constraint is that in the decoding order, the nuh_layer_id must be equal to the layerId. A specific value, layerId, preceding the intra random access point picture. Any picture with a nuh_layer_id equal to the one shown is in the output order. Random access point picture and all associated random access points The person described in item 1, who stipulates that it must precede the numberable reading picture. Law.

[0156] The following items illustrate exemplary embodiments of the technology described in the previous chapter (for example, item 4).

[0157] 6.1 Video having one or more video layers including multiple video pictures, and This includes performing a conversion between the coded representation of the image and the coded The representation is as follows: (1) Trailing picture, in output order, associated IRAP(I (Ntra Random Access Point) Picture or GDR (Grad (ual Decoder Refresh) must be followed by a picture, or ( 2) Pictures with the same layer ID as a GDR picture will be output in the same order as the GDR picture. The chat must precede all associated pictures of the GDR picture. A video processing method that conforms to format rules that define such constraints.

[0158] The following items illustrate exemplary embodiments of the technology described in the previous chapter (for example, item 5).

[0159] 7.1 Video having one or more video layers including multiple video pictures, and This involves performing a transformation between the coded representation of the image, and the transformation is subject to an ordering constraint. Picture, IRAP (Intra Random Access Point) picture And only if the non-reading picture is on the same layer, the picture, IRAP picture In accordance with the rules that apply to Kucha and non-reading pictures, this rule teeth, (a) A first rule that defines the values ​​and decoding order of the field sequence, or (b) The order of the reading and / or non-reading pictures of the layers On the other hand, there is the image processing method.

[0160] The following items illustrate exemplary embodiments of the technology described in the previous chapter (for example, item 6).

[0161] 8.1 Video having one or more video layers including multiple video pictures, and This involves performing a conversion between the coded representation of the image and the reading Picture, RADL (Random Access Decodable Leading) g) Pictures and GDR (Gradual Decoding Refresh) RASL (Random Access Skipped Le) associated with Kucha (adding) A video processing method that conforms to rules that define the order of pictures.

[0162] The following items illustrate exemplary embodiments of the technology described in the previous chapter (for example, item 7).

[0163] 9.1 Video having one or more video layers including multiple video pictures, and This involves performing a conversion between the coded representation of the image and the clean run. This specifies that the constraints on the reference picture list for dam access pictures are limited to layers. A video processing method that complies with regulations.

[0164] 10. The constraint is that for layers with clean random access pictures, decoding is performed. Alternatively, the preceding intra-random access point picture in the output order is the reference picture. The method described in item 9, which specifies that an entry in the list is not referenced.

[0165] The following items illustrate exemplary embodiments of the technology described in the previous chapter (for example, item 8).

[0166] 11.1 A video having one or more video layers including multiple video pictures, This includes performing a conversion between the current picture and its coded representation, and the conversion is performed by the current picture. The reference picture generated by the decoding process for generating an unavailable reference picture It complies with the rules that specify the conditions under which it is permitted to refer to entries in the Kucharist. Image processing methods.

[0167] 12. The condition is that the current picture is equal to 1. NoOutputBeforeReco CRA (Clean Random Access) picture with veryFlag RASL (Random Access Skipped Leading) associated with ng) Picture, NoOutputBeforeRecoveryFlag equal to 1 A GDR (Gradual Decoder Refresh) picture, or 1 GDR picture with NoOutputBeforeRecoveryFlag equal to The method described in item 11, which is to restore the picture.

[0168] The following items are examples of the technologies described in the previous chapter (for example, items 9, 10, and 11). Embodiments are shown.

[0169] 13.1 A video having one or more video layers including multiple video pictures, This includes performing a conversion between the current picture and its coded representation, and the conversion is performed by the current picture. The images conform to the ordering rules between the chat and the reference picture list corresponding to the current picture. Processing method.

[0170] 14. The rule is that the current nuh_layer_id has a value equal to a specific value layerId. The current picture has a NoOutputBeforeRecoveryFlag equal to 1. The CRA (Clean Random Access) picture has, in the encoding order The same CRA with NoOutputBeforeRecoveryFlag equal to 1 The picture preceding the leading picture associated with the picture, equal to 1 (NoO) Associated with CRA pictures that have outputBeforeRecoveryFlag The reading picture, equal to 1 NoOutputBeforeRecovery A GDR picture with a Flag, or NoOutputBeforeR equal to 1. Pass nuh_layer_id, which is equal to ecoveryFlag and layerId. If it is not a restored GDR picture, it will generate an unavailable reference picture. RefPicList[0] or RefPicList generated by the decoding process Item 13 stipulates that there are no pictures referenced by the entry in [1]. Methods used.

[0171] 15. The rule is that the current picture is IRAP (Intra Random Access). (s Point) Associated with the picture and following the IRAP picture in the output order. In this case, in the output order or decoding order, the R that precedes the associated IRAP picture The active entry in efPicList[0] or RefPicList[1] The method described in item 13, which stipulates that there are no referenced pictures.

[0172] 16. The rule is that the current picture is IRAP (Intra Random Access). (s Point) Associated with the picture, and in the output order, following the IRAP picture, Furthermore, in both the decoding order and the output order, the same IRAP picture associated with the same image. If it follows a DingPicture, the associated IRAP in output order or decoding order. The RefPicList[0] or RefPicList[1] that precedes the picture The method described in item 13, which stipulates that there are no pictures referenced by the entry.

[0173] 17. Conversion includes encoding video into a coded representation, items 1 to 1 The method described in any of 6.

[0174] 18. The conversion involves decoding the coded representation to generate the pixel values ​​of the image. Including any method described in item 1 through 16.

[0175] 19. Configured to implement one or more of the methods described in items 1 through 18. A video decoding device equipped with a processor.

[0176] 20. Configured to implement one or more of the methods described in items 1 through 18. A video encoding device equipped with a processor.

[0177] 21. In a computer program product in which computer code is stored, When executed by the processor, the processor will perform one of the actions described in items 1 through 18. Implement the law.

[0178] 22. The method, apparatus, or system described herein.

[0179] Section 2 presents exemplary examples of the techniques discussed in the previous section (for example, items 1-7).

[0180] 1. One or more video pictures containing one or more video images in accordance with the format rules. Performing a conversion between video having an image layer and the video bitstream 712 Including formatting rules, the associated intra-random access of the second picture. The first and second picture, both point pictures, belong to the same video layer. A method of image processing that stipulates that it is constrained to be such (for example, method 710 shown in Figure 7A) ).

[0181] 2. Between the first and second video pictures in the decoding order, the same video layer The method described in item 1, which does not include a progressively decoded refresh picture.

[0182] 3. The first video picture and the second picture are part of the video coding layer network. The identifier of the layer to which the abstraction layer unit belongs, and the non-video coding layer network The method described in item 1, wherein the identifier of the layer to which the abstract layer unit is applied is the same. .

[0183] 4. In the decoding order, the first video picture and the second picture have the same identifier. If there is no progressive decoding refresh picture, use one of the methods described in items 1 through 3.

[0184] 5. The formatting rules stipulate that the trailing picture is associated with the output order. Intra-random access point picture or progressively decoded refresh picture The method described in item 1, which further specifies what follows.

[0185] 6. One or more video pictures containing one or more video images in accordance with the format rules. Performing a conversion between video having an image layer and the video bitstream 722 Including formatting rules, trailing pictures in the bitstream are progressive. A video processing method that specifies permission to associate the decoded refresh picture with the refresh picture. (For example, method 720 shown in Figure 7B).

[0186] 7. Trailing picture is a network abstraction unit for each video coding layer. The item is a picture that has the Trail Network Abstract Layer Unit type. The method described in 6.

[0187] 8. Trailing pictures are associated with intra-random access point pictures. The method described in item 6 or 7, which is permitted.

[0188] 9. Intra-random access point picture or progressively decoded refresh picture Trailing pictures associated with a chat are, in the decoding order, intra-random. Items 6 through 8 follow the access point picture or novel decryption refresh picture. One of the methods described above.

[0189] 10. Following the associated intra-random access point picture in the output order Furthermore, the associated intra-random access point pictures in the decoding order The picture preceding it is not permitted, as described in any of items 6 through 8.

[0190] 11. The formatting rules are as follows: For each layer, the first layer in the bitstream Each picture in layers other than the picture in the decoding order is the previous input of the same layer The closer of the random access point or the progressive decoder refresh picture. A method of item 6 that further specifies how to be associated with something.

[0191] 12. Trailing pictures are associated with the intra-random elements in the output order. The item is required to be followed by an access point or a progressive decoder refresh picture. The method described in item 6 or 11.

[0192] 13. One or more video pictures in accordance with the format rules. Performing a conversion between video with a video layer and a video bitstream 732 The format rules include prioritizing the intra-random access point in the decoding order. The constraints on the output order of the pictures being processed can be applied to pictures within the same video layer. A video processing method that defines the following (for example, method 730 shown in Figure 7C).

[0193] 14. The constraint is that the NAL (Network Abstraction) is equal to a specific value. The Layer unit header has a layer identifier, and in the decoding order, it is equal to a specific value. Intra-random access point picture with NAL unit header layer identifier Any picture preceding it will be output in the intra-random access point picture order. To the chat and all associated random access decryptable reading pictures A method of item 13 that specifies that precedence is required.

[0194] 15. NAL (Network Abstraction Layer) Unit Head The Dalaye identifier is nuh_layer_id, as described in item 14.

[0195] 16. NAL (Network Abstraction Layer) Unit Head The DAREIER identifier is the layer to which the video coding layer network abstraction layer unit belongs. The ear identifier, or non-video coding layer network abstraction layer unit, is applied. The method described in item 14, which defines the identifier of the layer to be performed.

[0196] 17. One or more video pictures in accordance with the format rules. Performing a conversion between video with a video layer and a video bitstream 742 The format rules include (1) Trailing pictures are associated in the output order. Refreshing an intra-random access point picture or a progressive decoder Following the picture, or (2) the same NAL(N) for the progressive decoder refresh picture. (etwork Abstraction Layer) Unit Header Layer Identifier The picture being output is a progressive decoder refresh picture and a progressive decoder The constraint is that it precedes all associated pictures of the CODA refresh picture. A video processing method that defines this (for example, method 740 shown in Figure 7D).

[0197] 18. In accordance with the rules, one or more video layers containing one or more video pictures This includes performing a conversion between the video and the video bitstream, and the rules This includes pictures, intra-random access point pictures, and non-reading pictures. Intra random access points are associated only if the chat is on the same layer. This specifies that constraints must be applied to the decoding order of readable and non-readable pictures. , an image processing method (for example, method 750 shown in Figure 7E).

[0198] 19. The constraint is that the value of the field sequence flag is equal to 0 and equal to a specific value. iNAL (Network Abstraction Layer) Unit Header Ray A picture with a Ya identifier picture is associated with an intra-random access point picture. If it is an attached reading picture, the picture is intra in the decoding order. All non-reading pictures associated with random access point pictures The method described in item 18, which stipulates that the preceding action will occur.

[0199] A field sequence flag value equal to 20.0 corresponds to the coded layer image. The method described in item 19, which indicates that a sequence transmits pictures representing frames.

[0200] 21. The constraint is that the value of the field sequence flag is equal to 0, and the picture is This is a reading picture associated with an InterRandomAccess Point picture. In this case, the picture is related to the intra random access point picture in the decryption order. Item 18 stipulates that it precedes all linked non-reading pictures. Method of loading.

[0201] 22. In accordance with the rules, one or more video layers containing one or more video pictures The rules include performing a conversion between the video and the video bitstream. This includes reading pictures, random-access decryptable reading pictures, and Random access skip ready associated with progressively decoded refresh picture A video processing method (for example, method 760 shown in Figure 7F) that defines the order of the images.

[0202] 23. The reading picture associated with the progressively decoded refresh picture is In the decoding order, progressive decoding of the refresh picture is followed by progressive decoding in the output order. The method described in item 22, which precedes the refresh picture.

[0203] 24. Random access decryption associated with progressively decrypted refresh pictures. The reading picture is associated with the progressively decoded refresh picture. It is a reading picture, and a coding of a random-access decryptable reading picture. NAL (Network Abstraction Layer) corresponding to the sliced The method according to item 22, having unit type er).

[0204] 25. Random access decryption associated with progressively decrypted refresh pictures. The reading picture is associated with the progressively decoded refresh picture. It is a reading picture, and a coding of a random-access decryptable reading picture. NAL (Network Abstraction Layer) corresponding to the sliced The method according to item 22, having unit type er).

[0205] 26. In accordance with the rules, one or more video layers containing one or more video pictures This includes performing a conversion between the video and the video bitstream, and the rules This is for the reference picture list for slicing clean random access pictures. A video processing method that specifies limiting constraints to layers (for example, method 77 shown in Figure 7G) 0).

[0206] 27. The constraint is that for layers with clean random access pictures, decoding is performed. Alternatively, in the output order, the preceding intra random access point picture is the reference picture. The person described in item 26, who is not referred to by entries in the charity. Law.

[0207] 28. Conversion includes encoding video into a bitstream, as described in items 1 through 27. A method using either of the above.

[0208] 29. Conversion includes decoding video from a bitstream, items 1 through 27. The method described in any of the following.

[0209] 30. Conversion involves generating a bitstream from the video, and the method is to generate a bitstream This further includes storing the ream on a non-temporary computer-readable storage medium, item 1 to The method described in any of the 27 methods.

[0210] 31. A program configured to implement one or more of the methods described in items 1 through 30. Image processing device equipped with a sensor.

[0211] 32. The method described in any of items 1 through 30, and the bitstream to non-temporary container This includes storing a video bitstream, which further includes storing it on a computer-readable recording medium. method.

[0212] 33. When executed, the processor is subjected to one or more of the methods described in items 1 through 30. A computer-readable medium for storing program code to be implemented.

[0213] 34. A computer that stores the bitstream generated according to one of the methods described above. A computer-readable medium.

[0214] 35. A device configured to implement one or more of the methods described in items 1 through 30. A video processing device for storing stream representations.

[0215] Section 3 provides exemplary embodiments of the technologies described in the previous chapter (for example, items 8 and 9). vinegar.

[0216] 1. In accordance with the rules, have one or more video layers containing one or more video pictures. The rules include performing a conversion between the video and the video bitstream, and the rules are , The picture generated by the decoding process to produce an unavailable reference picture, The active entry in the current slice's reference picture list is the current picture. A video processing method (for example, the method shown in Figure 8A) defines the condition that it will not be referenced. 0).

[0217] 2. Active entries are reference indexes in the current picture's interpretation. The method described in item 1, corresponding to an entry available for use as such.

[0218] 3. The condition is that the NAL (Network Abstraction L) is equal to a specific value. The current picture with the ayer) unit header layer identifier is equal to 1 before restoration. Associated with clean random access pictures that have a variable indicating no output. Random access skip reading picture, gradual decode with variable equal to 1 A flash picture, or a variable equal to 1 and a NAL unit equal to a specific value. Not a restored picture of a progressive decoder refresh picture that has a header layer identifier. The method described in item 1.

[0219] 4. In accordance with the rules, one or more video layers containing one or more video pictures The rules include performing a conversion between the video and the video bitstream, and the rules are , The picture generated by the decoding process to produce an unavailable reference picture, The current picture is referenced by an entry in the current slice's reference picture list. A video processing method (for example, method 820 shown in Figure 8B) that defines the condition that it cannot be done.

[0220] 5. The condition is that the NAL (Network Abstraction L) is equal to a specific value. The current picture with the ayer) unit header layer identifier is equal to 1 before restoration. Clean random access pictures with a variable indicating no output, in the order of decoding. A ready associated with a clean random access picture that has a variable equal to 1. A picture preceding a ping picture, or a variable equal to 1 and NA equal to a specific value Restoration picture of a progressive decoder refresh picture having an L-unit header layer identifier The method described in item 4, which is not a cha.

[0221] 6. Conversion includes any of items 1 through 5, including encoding video into a bitstream. The method used.

[0222] 7. The conversion includes decoding the video from the bitstream, any of items 1 through 5. A method using one of the following methods.

[0223] 8. The conversion includes generating a bitstream from the video, and the method is to generate a bitstream Items 1 to 5 further include storing the data on a non-temporary computer-readable recording medium. One of the methods described above.

[0224] 9. A program configured to implement one or more of the methods described in items 1 through 8. Image processing device equipped with a transistor.

[0225] 10. The method described in any of items 1 through 8, and the bitstream is not a temporary compilation. A method for storing a video bitstream, which further includes storing it on a computer-readable recording medium. Law.

[0226] 11. When executed, process one or more of the methods described in items 1 through 8. A computer-readable medium for storing program code to be implemented by a computer.

[0227] 12. A computer that stores the bitstream generated according to one of the methods described above. A computer-readable medium.

[0228] 13. Configured to implement one or more of the methods described in items 1 through 12. Also, a video processing device for storing bitstream representations.

[0229] The set in Section 4 is an exemplary embodiment of the techniques discussed in the previous section (for example, items 10 and 1). This shows 1).

[0230] 1. In accordance with the rules, one or more video frames including the current picture including the current slice. Includes 912 performing a conversion between video having ears and the video bitstream. The rule is that the current slice's reference picture list is in the decoding order or output order. P This specifies a condition that does not allow having an active entry that references Kucha. A video processing method (for example, method 910 shown in Figure 9A).

[0231] 2. Active entries are reference indexes in the current picture's interpretation. The method described in item 1, corresponding to an entry available for use as such.

[0232] 3. The rule is that the current picture's reference picture list is, in the decoding order, the current picture The picture preceding the intra random access point picture associated with the chat Item 1 or Item 1 specifies the conditions under which you are not permitted to have an active entry that references The method described in 2.

[0233] 4. The rule is that the current picture's reference picture list is in output order, the current picture The picture preceding the intra random access point picture associated with the camera is used as a reference. Items 1 to 3 specify the conditions under which having an active entry is not permitted. One of the methods described above.

[0234] 5. The condition is that the current picture is associated with an intra-random access point picture. Furthermore, in the decoding order and / or output order, intra-random access point pins The method described in any of items 1 through 4, which follows the Kucha.

[0235] 6. The current picture belongs to the video coding layer network abstraction layer unit. Layer identifier, or non-video coding layer network abstract layer unit Intra-random access point picture with the same value as the identifier of the layer to which it applies The following items 1 to 4 are followed in the order of decoding and / or output: A method using either of the above.

[0236] 7. In accordance with the rules, one or more video frames including the current picture including the current slice. Includes 922, which performs conversion between video with ears and the video bitstream. The rule is that the current slice's reference picture list is in the decoding order or output order. P The video specifies a condition that does not allow having an active entry pointing to Kucha. Processing method (Method 920 shown in Figure 9B).

[0237] 8. The rule is that the current picture's reference picture list is in the decoding order of the current picture. The picture preceding the intra random access point picture associated with the chat The method described in item 7, which specifies the conditions under which an entry is not permitted to have a reference.

[0238] 9. The rule is that the current picture's reference picture list is in output order, the current picture The picture preceding the intra random access point picture associated with the camera is used as a reference. Those who are not permitted to have an entry as described in item 7 or 8 Law.

[0239] 10. Intra-random access point pictures have 0 or more reading pictures. Associated with, the condition is that the current picture is an intra-random access point picture. Associated with, in the decoding order and / or output order, intra-random action Following the setpoint picture, both the decoding order and the output order are intra-random. Following zero or more reading pictures associated with the access point picture A method described in any of items 7 through 9.

[0240] 11. Intra-random access point pictures have 0 or more reading pictures. Associated with this, the condition is that the current picture is video coding layer network abstraction Identifier of the layer to which the layer unit belongs, or non-video coding layer network An intra-random array having the same value as the identifier of the layer to which the abstraction layer unit is applied. Access point picture, and 0 or more reading pictures, in the decryption order and / The method described in any of items 7 through 9, which is to be the order in which the output is performed.

[0241] 12. Conversion includes encoding video into a bitstream, as described in items 1 through 11. A method using either of the above.

[0242] 13. Conversion includes decoding the video from the bitstream, as described in items 1 through 11. One of the methods described above.

[0243] 14. The conversion includes generating a bitstream from the video, and the method is to generate a bitstream This further includes storing the ream on a non-temporary computer-readable storage medium, item 1 to The method described in any of the 11.

[0244] 15. Configured to implement one or more of the methods described in items 1 through 14. A video processing device equipped with a processor.

[0245] 16. The method described in any of items 1 through 14, and the bitstream to non-temporary container This includes storing a video bitstream, which further includes storing it on a computer-readable recording medium. method.

[0246] 17. When executed, process one or more of the methods described in items 1 through 14. A computer-readable medium for storing program code to be implemented by the computer.

[0247] 18. A computer that stores the bitstream generated according to one of the methods described above. A computer-readable medium.

[0248] 19. Configured to implement one or more of the methods described in items 1 through 14. Also, a video processing device for storing bitstream representations.

[0249] In this specification, the term "video processing" includes video encoding, video decoding, video compression, and This can refer to video processing. For example, a video compression algorithm uses the pixel representation of the video to determine the corresponding... It may be applied during the conversion to a bitstream representation, or vice versa. The bitstream representation of an image block is, for example, defined by the syntax, bit This may correspond to bits that are spread to the same or different locations within the stream. Example For example, one macroblock is, in terms of the transformed and coded error residual values, Furthermore, the bitstream uses the header and bits in other fields to sign It may be converted to a number. Furthermore, during conversion, the decoder, as described in the above solution, Based on the judgment, we have knowledge that some fields may or may not exist. It is also possible to parse the bitstream. Similarly, an encoder can parse a specific syntactic file. Determine whether a 'gold' should be included or not, and the syntactic feel You can either include or exclude the character "do" from the coded expression. This may generate a correspondingly coded representation.

[0250] Disclosed and other solutions, examples, embodiments, and modules described herein. The implementation forms of the structure and its functional operation are as disclosed herein and their structural equivalents. This includes digital electronic circuits, or computer software, firmware, or This may be done in hardware, or in one or more combinations thereof. The disclosed and other embodiments may be one or more computer programs. Because it is implemented by RAM products, i.e., data processing devices, or data processing devices A computer program encoded on a computer-readable medium to control its operation. This can be implemented as one or more modules of instructions. The body includes machine-readable memory devices, machine-readable memory substrates, memory devices, and machine-readable propagating signals. The resulting substance may be a composition of substances, or one or more combinations thereof. The term "processing device" is used, for example, to refer to a programmable processor, a computer, and This includes multiple processors, or computers, and all other devices for processing data. This includes the device, machine, and other components. In addition to the hardware, this device includes the computer Code that creates the execution environment for a program, such as processor firmware, protocols, etc. Tack, database management system, operating system, or one of these It can contain codes that make up multiple combinations. The propagated signal is artificially generated. A signal, such as an electrical, optical, or electromagnetic signal generated by a machine, is sent to a suitable receiving device. It is generated to encode information for transmission.

[0251] Computer programs (programs, software, software applications) A script (also called code) is a compiled language or an interpreted language. It can be written in any form of programming language, including languages, and it is also a st A module suitable for use as an arron program or in a computing environment. Developed in any form, including as a route, component, subroutine, or other unit. It can be opened. Computer programs do not necessarily have file systems. It may not support files. A program may hold other programs or data. A portion of a file (for example, one or more scripts stored in a markup language document) ) may be recorded in a separate file, or it may be stored in a single file dedicated to that program. Okay, multiple adjustment files (for example, one or more modules, subprograms, etc.) It may also be stored in a file that stores part of the code. The communication is performed by a single computer located at one site, or distributed across multiple sites. Deploying to run on multiple computers interconnected on a network It is also possible.

[0252] The processing and logic flows described herein operate on input data and produce outputs. Execute one or more computer programs to perform a function by doing so. This can be done by one or more programmable processors. Processing and Logic Flow also provides logic circuits for specific applications, such as FPGAs (Field Pro). Grammable Gate Array) or ASIC (Application This can be done by a Specific Integrated Circuit. The device can also be implemented as a special-purpose logic circuit.

[0253] Processors suitable for running computer programs include, for example, general-purpose and dedicated microphones. Both of the loprocessors, and any one or more of any type of digital computer Includes the processor. Generally, the processor has read-only memory or random access It receives instructions and data from either or all of the following: an essential function of a computer. The system consists of a processor for executing instructions and one or more for storing instructions and data. It is a combination of multiple memory devices. Generally, a computer has one memory device to store data. or multiple mass storage devices, for example, magnetic, magneto-optical disks, or optical disks May include, or receive data from, these mass storage devices, or this They may be coupled in a way that allows them to transfer data. However, the computer It is not necessary to have such a device. Computer program instructions and data Computer-readable media suitable for storing data include all forms of non-volatile memory, media, and memory devices, such as EPROM, EEPROM, and flash memory. Magnetic disks, such as internal hard disks or removable disks, magneto-optical disks. This includes semiconductor storage devices such as CD-ROMs and DVD-ROM discs. The separator and memory may be complemented by application-specific logic circuits, or specific It may be incorporated into the logic circuitry for its intended use.

[0254] This patent specification contains many details, but these do not constitute the scope of any subject matter or claims. It should not be interpreted as limiting the scope, but rather as being specific to a particular embodiment of a particular technology. This should be interpreted as a description of possible features. In this patent document, a different embodiment The specific features described in the context can be combined and implemented in one example. Good. Conversely, various features described in the context of one example may appear in multiple embodiments. These may be implemented separately or in any appropriate subcombination. Furthermore, the features are: As described above, they act in specific combinations, and were initially claimed to do so. It may be, however, one or more features from the claimed combination may, in some cases, be combined It can be extracted from the combination, and the asserted combination is a subcombination or This may be directed towards variations of subcombinations.

[0255] Similarly, the operation is shown in a specific order in the drawings, which is to achieve the desired result. In order to do so, these actions must be performed in a specific or sequential order as indicated. It should not be understood as requiring all indicated actions to be performed. Furthermore, the separation of the various system components in the examples described in this patent specification is It should not be understood that such separation is necessary in all embodiments.

[0256] Only a few implementation forms and examples are described and illustrated in this patent document. Other embodiments, extensions, and modifications are possible based on the content.

Claims

1. A method of image processing, Performing a conversion between a video having one or more video layers containing one or more video pictures and the bitstream of said video, in accordance with the first rule. It has, The first rule specifies that the first video picture, which is the associated intra-random access point picture of the second picture, and the second picture are constrained to belong to the same video layer. Between the first video picture and the second picture in the decoding order, there is no progressive decoding refresh picture in the same video layer. The aforementioned transformation is carried out according to the second rule, The second rule states that if the first current picture having a NAL (Network Abstraction Layer) unit header layer identifier equal to the first specific value laverId is a specific picture, then in the decoding order, there are no pictures referenced by entries in RefPicList[0] or RefPicList[1] that precede any preceding intra-random access point picture having a NAL unit header layer identifier equal to laverId in the output order or decoding order. method.

2. The aforementioned transformation is carried out according to the third rule, The third rule specifies that the first constraint applies to the decoding order of the second current picture only if the second current picture associated with the intra-random access point picture, the intra-random access point picture, and the non-reading picture are on the same layer. The method according to claim 1.

3. The first constraint stipulates that if the value of the field sequence flag is equal to 0 and the second current picture having a NAL (Network Abstraction Layer) unit header layer identifier equal to a second specific value is a reading picture associated with the intra-random access point picture, then the second current picture precedes all non-reading pictures associated with the intra-random access point picture in the decoding order. The method according to claim 2.

4. If the value of the field sequence flag is equal to 1, then, assuming picA and picB are the first and last pictures associated with the intra-random access point picture in the decoding order, there is at most one non-reading picture that precedes picA in the decoding order and has a corresponding NAL unit header layer identifier equal to the second specific value laverId, and there is no non-reading picture between picA and picB in the decoding order that has a corresponding NAL unit header layer identifier equal to the second specific value laverId. The method according to claim 3.

5. The value of the field sequence flag equal to 0 indicates that the coded layer video sequence transmits a picture representing a frame. The method according to claim 3.

6. The aforementioned specific picture is a clean random access picture. The method according to claim 1.

7. The first video picture and the second picture have the same identifier as the identifier of the layer to which the video coding layer network abstraction layer unit belongs, or the identifier of the layer to which the non-video coding layer network abstraction layer unit applies. The method according to claim 1.

8. The aforementioned transformation is carried out in accordance with the fourth rule, The fourth rule states that a trailing picture in the bitstream is permitted to be associated with a progressively decoded refresh picture or an associated intra-random access point picture. The method according to any one of claims 1 to 7.

9. The trailing picture is a picture in which each video coding layer network abstraction layer unit has a trail network abstraction layer unit type. The method according to claim 8.

10. The trailing picture associated with the associated intra-random access point picture or progressive decoding refresh picture follows the associated intra-random access point picture or progressive decoding refresh picture in the decoding order or output order. The method according to claim 8.

11. Pictures that follow the associated intra-random access point picture in the output order, and that precede the associated intra-random access point picture in the decoding order, are not permitted. The method according to claim 8.

12. The fourth rule further stipulates that, for each layer, with the exception of a first picture in the bitstream in the layer, each picture in the layer relates in decoding order to the closer of the previous intra-random access point or progressive decoding refresh picture in the same layer. The method according to claim 8.

13. The aforementioned transformation is carried out in accordance with the fifth rule, The fifth rule stipulates that the fifth constraint on the output order of pictures preceding an intra-random access point in the decoding order is applicable to pictures on the same video layer. The method according to any one of claims 1 to 12.

14. The fifth constraint stipulates that any picture having a Network Abstraction Layer (NAL) unit header layer identifier equal to a third specific value, and that in the decoding order precedes an intra-random access point having the NAL unit header layer identifier equal to the third specific value, is required to precede the intra-random access point picture and all associated random access decodeable reading pictures in the output order. The method according to claim 13.

15. The conversion includes encoding the video into the bitstream. The method according to any one of claims 1 to 14.

16. The conversion includes decoding the video from the bitstream. The method according to any one of claims 1 to 14.

17. A device for processing video data, comprising a processor and non-temporary memory containing instructions, When the aforementioned instruction is executed by the processor, the processor will be instructed to: Performing a conversion between a video having one or more video layers containing one or more video pictures and the bitstream of said video, in accordance with the first rule. Have them do it, The first rule specifies that the first video picture, which is the associated intra-random access point picture of the second picture, and the second picture are constrained to belong to the same video layer. Between the first video picture and the second picture in the decoding order, there is no progressive decoding refresh picture in the same video layer. The aforementioned transformation is carried out according to the second rule, The second rule states that if the first current picture having a NAL (Network Abstraction Layer) unit header layer identifier equal to the first specific value laverId is a specific picture, then in the decoding order, there are no pictures referenced by entries in RefPicList[0] or RefPicList[1] that precede any preceding intra-random access point picture having a NAL unit header layer identifier equal to laverId in the output order or decoding order. Device.

18. In the processor, Performing a conversion between a video having one or more video layers containing one or more video pictures and the bitstream of said video, in accordance with the first rule. Store the command to execute, The first rule specifies that the first video picture, which is the associated intra-random access point picture of the second picture, and the second picture are constrained to belong to the same video layer. Between the first video picture and the second picture in the decoding order, there is no progressive decoding refresh picture in the same video layer. The aforementioned transformation is carried out according to the second rule, The second rule states that if the first current picture having a NAL (Network Abstraction Layer) unit header layer identifier equal to the first specific value laverId is a specific picture, then in the decoding order, there are no pictures referenced by entries in RefPicList[0] or RefPicList[1] that precede any preceding intra-random access point picture having a NAL unit header layer identifier equal to laverId in the output order or decoding order. Non-temporary computer-readable storage medium.

19. A method for storing a video bitstream, The aforementioned method, Performing a conversion between a video having one or more video layers containing one or more video pictures and the bitstream of said video, in accordance with the first rule, The bitstream is stored in a non-temporary computer-readable recording medium, It has, The first rule specifies that the first video picture, which is the associated intra-random access point picture of the second picture, and the second picture are constrained to belong to the same video layer. Between the first video picture and the second picture in the decoding order, there is no progressive decoding refresh picture in the same video layer. The aforementioned transformation is carried out according to the second rule, The second rule states that if the first current picture having a NAL (Network Abstraction Layer) unit header layer identifier equal to the first specific value laverId is a specific picture, then in the decoding order, there are no pictures referenced by entries in RefPicList[0] or RefPicList[1] that precede any preceding intra-random access point picture having a NAL unit header layer identifier equal to laverId in the output order or decoding order. method.

Images