Carrying of sub-stream in jpeg ai code stream

Abstract

A mechanism of processing video data is disclosed. The mechanism includes determining that a joint photographic experts group (JPEG) artificial intelligence (AI) bitstream contains one or more substreams, where each substream starts with a marker identifier (ID), followed by a size and payload data of the substream. A conversion between visual media data and the JPEG AI bitstream is performed based on the marker ID, the size and the payload data of the substream.

Description

[0001] Cross-references to related applications

[0002] This application claims priority and benefit to U.S. Provisional Patent Application No. 63 / 564,130, filed March 12, 2024. All of the aforementioned patent applications are incorporated herein by reference in their entirety. Technical Field

[0003] This patent document relates to the generation, storage, and use of digital audio and video media information in file formats. Background Technology

[0004] Digital video accounts for the largest share of bandwidth used on the internet and other digital communication networks. As the number of connected user devices capable of receiving and displaying video increases, the bandwidth demand for digital video is likely to continue to grow. Summary of the Invention

[0005] The first aspect relates to a method for processing video data, comprising: determining that a Joint Photo Experts Group (JPEG) Artificial Intelligence (AI) bitstream contains one or more substreams, wherein each substream begins with a marker identifier (ID), followed by substream size and payload data; and performing a conversion between visual media data and the JPEG AI bitstream based on the marker ID, the substream size and payload data.

[0006] The second aspect relates to an apparatus for processing video data, including a processor and a non-transitory memory having instructions thereon, wherein the instructions, when executed by the processor, cause the processor to perform any of the aforementioned aspects.

[0007] The third aspect relates to a non-transitory computer-readable medium including a computer program product for use by a video codec device, wherein the computer program product includes computer-executable instructions stored on the non-transitory computer-readable medium, such that when the computer-executable instructions are executed by a processor, the video codec device performs the methods of any of the preceding aspects.

[0008] The fourth aspect relates to a non-transitory computer-readable recording medium storing a JPEG AI stream of video generated by a method performed by a video processing apparatus, wherein the method includes: determining that the Joint Picture Experts Group (JPEG) Artificial Intelligence (AI) stream comprises one or more substreams, wherein each substream begins with a marker identifier (ID), followed by the size and payload data of the substream; and generating the JPEG AI stream based on the determination.

[0009] The fifth aspect relates to a method for storing a bitstream of video, comprising: determining that a Joint Picture Experts Group (JPEG) Artificial Intelligence (AI) bitstream contains one or more substreams, wherein each substream begins with a marker identifier (ID), followed by the size and payload data of the substream; generating a JPEG AI bitstream based on the determination; and storing the JPEG AI bitstream in a non-transitory computer-readable recording medium.

[0010] For clarity, any of the embodiments described above may be combined with one or more other embodiments described above to create new embodiments within the scope of this disclosure.

[0011] These and other features will become clearer through the following detailed description of the embodiments with reference to the accompanying drawings and claims. Attached Figure Description

[0012] To gain a more complete understanding of this disclosure, reference is now made to the following brief description, taken in conjunction with the accompanying drawings and detailed description, wherein like reference numerals denote like parts.

[0013] Figure 1 This is a block diagram illustrating an example video processing system.

[0014] Figure 2 This is a block diagram of an example video processing device.

[0015] Figure 3 This is a flowchart of an example method for video processing.

[0016] Figure 4 This is a block diagram illustrating an example video codec system.

[0017] Figure 5 This is a block diagram showing an example encoder.

[0018] Figure 6 This is a block diagram showing an example decoder.

[0019] Figure 7 This is a schematic diagram of an example encoder.

[0020] Figure 8 This is a flowchart of an example method for video processing. Detailed Implementation

[0021] First, it should be understood that although illustrative implementations of one or more embodiments are provided below, the disclosed systems and / or methods can be implemented using any number of techniques, whether currently known or yet to be developed. This disclosure should not be limited in any way to the illustrative embodiments, drawings, and techniques described below, including the exemplary designs and implementations shown and described herein, but can be modified within the scope of the appended claims and all their equivalents.

[0022] 1. Preliminary Discussion

[0023] This document relates to image and video encoding and decoding based on neural networks (NNs). Specifically, this disclosure relates to the carrying of substreams in image or video bitstreams based on neural networks (NNs), such as Joint Image Experts Group (JPEG) Artificial Intelligence (AI) bitstreams, which may also be referred to as bitstreams. These ideas can be applied individually or in various combinations to image and / or video encoding and decoding methods and specifications.

[0024] 2. Further discussion

[0025] Deep learning is rapidly developing across various fields, particularly in computer vision and image processing. Inspired by the tremendous success of deep learning in computer vision, many researchers have shifted their focus from image / video compression techniques to neural image / video compression. Neural networks are designed through interdisciplinary research in neuroscience and mathematics. They have demonstrated powerful capabilities in nonlinear transformations and classification. Significant progress has been made in neural network-based image / video compression techniques. Examples of neural network-based image compression algorithms have reportedly achieved rate-distortion (RD) performance comparable to Multifunctional Video Coding (VVC), a video coding standard developed by the Joint Video Experts Group (JVET), comprised of experts from the Moving Picture Experts Group (MPEG) and the Video Codec Experts Group (VCEG). With the continuous improvement in the performance of neural image compression, neural network-based video compression has become an actively developing research area. However, due to the inherent difficulties of the problem, neural network-based video coding is still in its early stages.

[0026] 2.1. Image / Video Compression

[0027] Image / video compression (also known as image / video encoding / decoding) generally refers to the computational techniques used to compress images or videos into binary code for easier storage and transmission. The binary code may or may not support lossless reconstruction of the original image or video; this is called lossless compression and lossy compression. Since lossless reconstruction is not necessary in most cases, most efforts focus on lossy compression. The performance of image or video compression algorithms is typically evaluated from two aspects: compression ratio and reconstruction quality. The compression ratio is directly related to the amount of binary code; less is better. Reconstruction quality is measured by comparing the reconstructed image or video with the original image or video; higher is better.

[0028] Image / video compression techniques can be divided into two branches: classical video encoding / decoding methods and neural network-based video compression methods. Classical video encoding / decoding schemes employ transform-based solutions, where researchers model dependencies in the quantization domain through carefully hand-designed entropy encoding / decoding, leveraging statistical dependencies in latent variables (e.g., Discrete Cosine Transform (DCT) or wavelet coefficients). Neural network-based video compression takes two forms: neural network-based encoding / decoding tools and end-to-end neural network-based video compression. The former is embedded as an encoding / decoding tool within a classical video codec, serving only as part of the framework; while the latter is an independent framework developed based on neural networks, independent of classical video codecs.

[0029] A series of classic video codec standards have been developed to accommodate the ever-increasing amount of visual content. The International Organization for Standardization (ISO) / International Electrotechnical Commission (IEC) has two expert groups, the Joint Group of Picture Experts (JPEG) and the Moving Picture Experts Group (MPEG), and the International Telecommunication Union (ITU) Telecommunication Standardization Sector (ITU-T) also has its own Video Codec Experts Group (VCEG) for standardizing image or video codec technologies. Influential video codec standards released by these organizations include JPEG, JPEG 2000, H.262, H.264 / Advanced Video Codec (AVC), and H.265 / High-Efficiency Video Codec (HEVC). Following H.265 / HEVC, the Joint Video Experts Group (JVET), comprised of MPEG and VCEG, began working on the Multi-Functional Video Codec (VVC) standard. The first version of VVC was released in July 2020. Compared to HEVC, VVC reduces the bit rate by an average of 50% while maintaining the same visual quality.

[0030] Many researchers are working on neural network-based image encoding and decoding for use in neural network-based image / video compression. However, the network architectures used in example designs are relatively shallow, resulting in unsatisfactory performance. Thanks to abundant data and powerful computing resources, neural network-based methods have been better utilized in a variety of applications. Currently, neural network-based image / video compression has shown promising improvements and proven its feasibility. However, this technology is still far from mature, and many challenges remain to be solved.

[0031] 2.2. Neural Networks

[0032] Neural networks, also known as artificial neural networks (ANNs), are computational models used in machine learning techniques. They typically consist of multiple processing layers, each composed of several simple but non-linear basic computational units. One advantage of these deep networks is their ability to process data with multiple levels of abstraction and transform it into different kinds of representations. Note that these representations are not manually designed; instead, they are learned from massive amounts of data using general machine learning procedures. Deep learning eliminates the need for manually designed representations and is therefore considered particularly suitable for processing raw, unstructured data, such as acoustic and visual signals, which has been a long-standing challenge in the field of artificial intelligence.

[0033] 2.3. Neural Networks for Image and Video Compression

[0034] Example neural networks used in image compression methods can be divided into two categories: pixel probability modeling and autoencoders. The former belongs to predictive encoding / decoding strategies, while the latter is a transform-based solution. Sometimes, these two methods are combined.

[0035] Similar to classic video encoding and decoding techniques, neural image compression is based on intra-frame compression in neural network-based video compression. Therefore, the development of neural network-based video compression technology lagged behind that of neural network-based image compression, but due to its complexity, it requires more effort to overcome the challenges. Some researchers are working on neural network-based video compression schemes. Compared to image compression, video compression requires effective methods to eliminate inter-frame redundancy. Inter-frame prediction is a key step in these works. Motion estimation and compensation have been employed, but only recently have they been implemented using trained neural networks.

[0036] Neural network-based video compression research can be broadly categorized into two types based on the target scenario: random access and low latency. In the case of random access, decoding can begin at any point in the sequence, which is divided into multiple individual segments, each of which can be decoded independently. Low latency aims to reduce decoding time by using earlier frames as reference frames to decode subsequent frames.

[0037] 2.4. JPEG AI Image Encoding and Decoding Standard and Signaling of JPEG AI Substreams

[0038] The JPEG AI image codec standard is an image codec standard managed by the JPEG Working Group (WG), which is WG 1 of ISO / IEC Joint Technical Committee (JTC) 1 SC 29. The ISO / IEC number for the JPEG AI standard is ISO / IEC 6048.

[0039] The JPEG AI draft specification example utilizes some of the neural network-based image encoding and decoding methods described above. The following describes or summarizes some features of the example JPEG AI specification, including signaling for various substreams.

[0040] 2.4.1 (2.2) Specification of Syntax Functions and Descriptors

[0041] The functions presented here are used for syntax description. These functions are expressed based on the value of a bitstream pointer, which indicates the position of the next bit read from the bitstream by the decoding process.

[0042] The `read_bits(n)` function reads the next `n` bits from the bitstream and moves the bitstream pointer forward by `n` bits. When `n` equals 0, `read_bits(n)` is specified to return a value equal to 0 and not advance the bitstream pointer.

[0043] The `read_bits_from_back(n)` function reads `n` bits from the end of a substream in the bitstream and moves the substream bitstream pointer forward by `n` bits. When `n` equals 0, `read_bits_from_back(n)` is specified to return a value equal to 0 and not move the substream bitstream pointer.

[0044] The following descriptors specify the parsing process for each syntax element:

[0045] - b(8): Bytes (8 bits) of any pattern with a bit string. The parsing process for this descriptor is specified by the return value of the function read_bits(8).

[0046] - f(n): A fixed-pattern bit string of n bits, written (from left to right) with the leftmost bit first. The parsing process for this descriptor is specified by the return value of the function read_bits(n).

[0047] - i(n): A signed integer using n bits. When n is "v" in the syntax table, the number of bits varies in a manner that depends on the values ​​of other syntax elements. The parsing process for this descriptor is specified by the return value of the function read_bits(n), which is interpreted as a two's complement integer representation, with the most significant bit written first.

[0048] - se(v): A syntax element for signed integer zero-order Exp-Golomb encoding / decoding, where the left bit comes first. The parsing procedure for this descriptor is specified in Appendix C, where the order k equals 0.

[0049] - u(n): An unsigned integer using n bits. When n is "v" in the syntax table, the number of bits varies in a manner that depends on the values ​​of other syntax elements. The parsing process for this descriptor is specified by the return value of the function read_bits(n), which is interpreted as a binary representation of the unsigned integer, with the most significant bit written first.

[0050] - ub(n): An unsigned integer using n bits. When n is "v" in the syntax table, the number of bits varies in a manner that depends on the values ​​of other syntax elements. The parsing process for this descriptor is specified by the return value of the function read_bits_from_back(n), which is interpreted as a binary representation of the unsigned integer, with the most significant bit written first.

[0051] 2.4.2 (6.2) Stream Layout ...

[0053] The bitstream consists of six parts with byte boundaries:

[0054] 1. Start of stream marker (SOC);

[0055] 2. Image header marker (PIH), followed by the image header;

[0056] 3. Tool Header Mark (TOH), followed by tool information;

[0057] 4. Start of Quality Quote (SOQ), followed by the bitstream;

[0058] 5. The Z-stream starts with a Start of Z (SOZ) marker, followed by the bitstream of the super-prior information tensor z, including... and ;

[0059] 6. The residual stream start marker (SORp) of the primary component, followed by the bitstream of the primary component residual, including... ;

[0060] 7. Secondary component residual stream start markers (SORs), followed by the bitstream of the secondary component residual, including... ;

[0061] 8. End of stream marker (EOC).

[0062] The overall grammatical structure of the image is as follows:

[0063]

[0064] Each bitstream begins with a 16-bit marker. All markers used in this specification are as follows:

[0065]

[0066] 2.4.3 (6.3) Image Header

[0067] This substream contains information about the image height. ,width Potential space piece location and size, control flags for each tool, scaling factors for primary and secondary components, - Learnable model index and displacement (primary components) for rate control parameters Secondary components are (information).

[0068] 2.4.2.1 (6.3.1) Syntax Table

[0069] ...

[0071] 2.4.2.3 (6.3.2) Image header semantics

[0072] The following service information is transmitted via signal:

[0073] picture_header_size is the number of bytes in the image header excluding the first two bytes of the marker;

[0074] Adding 64 to img_width specifies the width of the input image (from 64 to 65599).

[0075] img_height plus 64 specifies the height of the input image (from 64 to 65599); ...

[0077] 2.4.4. (6.4) Tool Head

[0078] This optional substream contains information about the tools. When the tools_header() flag segment is not present in the bitstream, the following tool enable flags are set to 0: Residual and Variance Scaling (RVS) enable flag (rvs_enable_flag), Latent Scaling Before Synthesis (LSBS) enable flag (lsbs_enable_flag), Gain Unit Refinement Scaling (GRFS) enable flag (grfs_enable_flag), Gain 3D (3D) enable flag (gain_3D_enable_flag), Inter-channel Correlation Information Filter (ICCI) enable flag (icci_enable_flag), Luminance Edge Filter (LEF) enable flag (LEF_enabled_flag), Enhancement Filter Extension (EFE) Linear Filter Enable flag (EFE_linear_filter_enabled_flag), and EFE Nonlinear Filter Enable flag (EFE_nonlinear_filter_enabled_flag).

[0079] 2.4.4.1 (6.4.1) Toolhead Syntax Table

[0080] ...

[0082] 2.4.4.2 (6.4.2) Tool Information Semantics

[0083] tools_header_size is the number of bytes in the tool header excluding the first two bytes of the marker;

[0084] rvs_enable_flag[comp] - This is a flag used in the RVS of the comp component. 0 indicates that RVS is disabled, and 1 indicates that RVS is enabled.

[0085] rvs_num_idx[comp] is an identifier for the operation point; 0 means 1 threshold for RVS, and 1 means 3 thresholds for RVS.

[0086] lsbs_enable_flag[comp] - This is a flag used in the LSBS mode of the comp component. 0 indicates LSBS is disabled, and 1 indicates LSBS is enabled. ...

[0088] 2.4.5. (6.5) Residual and Prior Information Tensor Encoding and Decoding ...

[0090] 2.4.5.1 (6.5.1.1) Syntax table of the super-prior information tensor

[0091]

[0092] 2.4.5.2 (6.5.1.2) Semantics of the Super-Prior Information Tensor

[0093] z_stream_size is the number of bytes in the super-prior information tensor bitstream excluding the first two bytes of the marker;

[0094] num_threads_z is the number of substreams in the prior information tensor bitstream that can be decoded in parallel; the maximum value of num_threads_z is 128.

[0095] thread_offsets_z[i] is the number of bytes between the start of the super-prior information tensor stream i and the start of the super-prior information tensor code stream (excluding the first two bytes of the marker).

[0096] 2.4.5.3 (6.5.2.1) Syntax table for quality graph information tensors

[0097]

[0098] 2.4.5.4 (6.5.2.2) Semantics of the Quality Graph Information Tensor

[0099] q_stream_size is the number of bytes in the quality graph bitstream excluding the first two bytes of the marker;

[0100] num_threads_q is the number of substreams in the quality graph bitstream that can be decoded in parallel; the maximum value of num_threads_q is 128.

[0101] thread_offsets_q[i] is the number of bytes between the start of quality graph substream i and the start of the quality graph bitstream (excluding the first two bytes of the marker).

[0102] 2.4.5.5 (6.5.3.1.1) Syntax table of primary residual tensors

[0103]

[0104] 2.4.5.6 (6.5.3.1.2) Semantics of the main residual tensor

[0105] r_primary_stream_size is the number of bytes in the primary component residual tensor bitstream excluding the first two bytes of the marker;

[0106] num_threads_r_primary is the number of substreams that can be decoded in parallel in the primary component residual tensor bitstream; the maximum value of num_threads_r_primary is 128, and it depends on the grade and level.

[0107] thread_offsets_r_primary[i] is the number of bytes between the start of the primary component residual tensor quantum stream i and the start of the primary component residual tensor bitstream (excluding the first two bytes of the marker);

[0108] 2.4.5.7 (6.5.3.2.1) Syntax table for minor residual tensors

[0109]

[0110] 2.4.5.8 (6.5.3.2.2) Semantics of minor residual tensors

[0111] r_secondary_stream_size is the number of bytes in the secondary component residual tensor bitstream excluding the first two bytes of the marker;

[0112] num_threads_r_secondary is the number of substreams that can be decoded in parallel in the minor component residual tensor bitstream; the maximum value of num_threads_r_secondary is 128, and it depends on the grade and level.

[0113] thread_offsets_r_secondary[i] is the number of bytes between the start of the secondary component residual tensor quantum stream i and the start of the secondary component residual tensor bitstream (excluding the first two bytes of the marker);

[0114] 3. The technical problem solved by the disclosed technical solution

[0115] The JPEG AI substream includes the image header substream (i.e., the `picture_header()` syntax structure), the tools header substream (i.e., the `tools_header()` syntax structure), the rendering information substream, the quality map substream (i.e., the `q_stream()` syntax structure), the super-prior information tensor stream (i.e., the `z_stream()` syntax structure), the primary residual substream (i.e., the `r_primary_stream()` syntax structure), the secondary residual substream (i.e., the `r_secondary_stream()` syntax structure), and some reserved type substreams associated with reserved substream tags. The image header substream, the primary residual substream, and the secondary residual substream are mandatory, and all other substreams are indicated as optional via a tag table.

[0116] The example design for signaling of the JPEG AI substream has the following problems:

[0117] First, the tags for the tool header, rendering information, and quality map are indicated as optional via a tag table. However, the syntax of the `picture()` syntax structure stipulates that the `tools_header()` and `q_stream()` syntax structures are unconditionally present within the `picture()` syntax structure. In other words, the tag table and the syntax of the `picture()` syntax structure contradict each other regarding the optionality of the tags (and corresponding substreams) for the tool header, rendering information, and quality map.

[0118] Second, the tag table specifies some reserved tag values ​​as optional. However, it is unclear how the version 1 JPEG AI decoder will handle JPEG AI bitstreams from future versions of the JPEG AI standard that contain reserved tags and associated substreams.

[0119] Third, the tag table specifies some reserved tag values ​​as optional. However, according to the current specification, the version 1 JPEG AI decoder will not be able to ignore substreams associated with tags reserved in JPEG AI version 1 that exist in JPEG AI bitstreams of future versions of the JPEG AI standard.

[0120] Fourth, in the JPEG AI stream, it is unclear which optional markers (and their corresponding substreams) will be placed before the media data substreams (i.e., the quality map substream, the super-prior information tensor quantum stream, the primary residual substream, and the secondary residual substream), and which optional markers (and their corresponding substreams) will be placed after the media data substreams, when they exist.

[0121] 4. List of solutions and implementation examples

[0122] To address the aforementioned issues, the following summarized methods are disclosed. These aspects should be considered as examples for explaining general concepts, and not interpreted in a narrow sense. Furthermore, these examples can be applied individually or in any combination.

[0123] 1) In one example, a substream in a JPEG AI stream is specified to consist of at least the following: an X-bit syntax element specifying the substream marker, a Y-bit syntax element specifying the length of the substream size, a Z-bit syntax element specifying the size (in bytes) of the substream payload, and the substream payload, where X, Y, and Z are integers greater than 0.

[0124] a. In one example, X equals 16.

[0125] b. In one example, Y equals 2.

[0126] c. In one example, Z is determined by the value Y.

[0127] d. In one example, Y equals 2, and the values ​​of Y 0, 1, 2, and 3 determine the values ​​of Z as 8, 16, 24, and 32, respectively.

[0128] e. In one example, the X-bit syntax element for the substream marker is specified as rsv_ss_marker.

[0129] f. In one example, the substream tag indicates the type of substream, such as image head substream, tool head substream, etc.

[0130] g. In one example, the Y-bit syntax element that specifies the length of the substream size is rsv_ss_size_len_in_bytes_minus1.

[0131] h. In one example, the Z-bit syntax element that specifies the size of the substream payload in bytes is rsv_ss_size_val.

[0132] i. In one example, the substream also consists of one or more byte alignment bits preceding the substream payload to ensure that the number of bits preceding the substream payload in the substream is an integer multiple of 8.

[0133] i. In one example, the alignment bit for each byte is specified to be 0.

[0134] ii. In one example, additionally, each byte alignment bit is reserved for future versions of the JPEG AI standard.

[0135] j. By applying this design to substreams associated with markers preserved in JPEG AI version 1, when the version 1 JPEG AI decoder encounters such a substream in a JPEG AI bitstream that exists in a future version of the JPEG AI standard, the decoder can parse the syntax elements rsv_ss_marker, rsv_ss_size_len_in_bytes_minus1, and rsv_ss_size_val, deduce the length of the substream payload, discard all data bytes of the substream payload, and then continue processing subsequent data bits in the JPEG AI bitstream.

[0136] 2) In one example, the existence of an optional substream in a JPEG AI stream is specified to depend on whether the value of the next M bits in the JPEG AI stream immediately following the syntax structure of the second substream is equal to the flag of the optional substream.

[0137] a. In one example, M equals 16.

[0138] 3) In one example, the specified value indicates that the reserved substream marker is reserved for future use by ISO / IEC and should not appear in the first version of the JPEG bitstream conforming to the JPEG AI standard.

[0139] a. In one example, additionally, it is specified that when a reserved substream marker is encountered, the first version of the decoder conforming to the JPEG AI standard must ignore the substream associated with that marker.

[0140] 4) In one example, optional substreams that can exist in the JPEG AI bitstream are defined as two categories, where the first category of substreams (when present) precedes the q_stream() syntax structure (when present) in decoding order, and the second category of substreams (when present) follows the always-present r_secondary_stream() syntax structure in decoding order, and precedes the always-present EOC marker in decoding order.

[0141] a. In one example, the optional sub-stream of the first category is called the prefix sub-stream.

[0142] b. In one example, the optional sub-stream of the second category is called the suffix sub-stream.

[0143] c. In one example, the optional substreams of the first category include substreams whose substream flag values ​​are in the range of 0xff83 to 0xff87 (inclusive).

[0144] d. In one example, the optional substreams of the second category include substreams whose substream flag values ​​are in the range of 0xff8c to 0xff8e (inclusive).

[0145] e. In one example, it is specified that one type of prefix substream is allowed to precede or follow any other type of prefix substream in the decoding order.

[0146] f. In one example, it is specified that one type of suffix substream is allowed to precede or follow any other type of suffix substream in the order of decoding.

[0147] 5) In one example, it is specified that all reserved type substreams are optional substreams.

[0148] 6) In one example, it is specified that when the SOQ flag exists and a second flag with a value in the range of 0xff83 to 0xff87 (inclusive) also exists, the substream associated with the second flag must precede the quality graph substream (i.e., the q_stream() syntax structure) associated with the SOQ flag in the decoding order.

[0149] 7) In one example, it is specified that when a tag with a value in the range of 0xff8c to 0xff8f (inclusive) exists, the substream associated with that tag must be after the minor component residual substream associated with the SORs tag in the decoding order.

[0150] 5. Examples

[0151] The following are some example implementations of the aspects summarized in Section 4.

[0152] Most of the relevant sections that have been added are shown in bold, or the deleted sections are shown in bold italics. Some other changes may exist as editable changes and are therefore not shown or highlighted. These changes are relative to the draft text of the JPEG AI standard.

[0153] 5.1 First Embodiment

[0154] This example is used for most of the projects summarized in Section 4.

[0155] 5.1.1. (2.2) Specification of Syntax Functions and Descriptors

[0156] The functions presented here are used for syntax description. These functions are expressed based on the value of a bitstream pointer, which indicates the position of the next bit read from the bitstream by the decoding process.

[0157] The `read_bits(n)` function reads the next `n` bits from the bitstream and moves the bitstream pointer forward by `n` bits. When `n` equals 0, `read_bits(n)` is specified to return a value equal to 0 and not advance the bitstream pointer.

[0158] The `read_bits_from_back(n)` function reads `n` bits from the end of a substream in the bitstream and moves the substream bitstream pointer forward by `n` bits. When `n` equals 0, `read_bits_from_back(n)` is specified to return a value equal to 0 and not move the substream bitstream pointer.

[0159] The next_bits(n) function provides the next bit in the bitstream for comparison purposes without advancing the bitstream pointer.

[0160] The following descriptors specify the parsing process for each syntax element:

[0161] - b(8): Bytes (8 bits) of any pattern with a bit string. The parsing process for this descriptor is specified by the return value of the function read_bits(8).

[0162] - f(n): A fixed-pattern bit string of n bits, written (from left to right) with the leftmost bit first. The parsing process for this descriptor is specified by the return value of the function read_bits(n).

[0163] - i(n): A signed integer using n bits. When n is "v" in the syntax table, the number of bits varies in a manner that depends on the values ​​of other syntax elements. The parsing process for this descriptor is specified by the return value of the function read_bits(n), which is interpreted as a two's complement integer representation, with the most significant bit written first.

[0164] - se(v): A syntax element for signed integer zero-order Exp-Golomb encoding / decoding, where the left bit comes first. The parsing procedure for this descriptor is specified in Appendix C, where the order k equals 0.

[0165] - u(n): An unsigned integer using n bits. When n is "v" in the syntax table, the number of bits varies in a manner that depends on the values ​​of other syntax elements. The parsing process for this descriptor is specified by the return value of the function read_bits(n), which is interpreted as a binary representation of the unsigned integer, with the most significant bit written first.

[0166] - ub(n): An unsigned integer using n bits. When n is "v" in the syntax table, the number of bits varies in a manner that depends on the values ​​of other syntax elements. The parsing process for this descriptor is specified by the return value of the function read_bits_from_back(n), which is interpreted as a binary representation of the unsigned integer, with the most significant bit written first.

[0167] 5.1.2. (6.2) Stream Layout ...

[0169] The bitstream consists of seven parts with byte boundaries, which are:

[0170] 1. SOC - Start of Stream Marker;

[0171] 2. PIH (Picture Header Marker), followed by the image header;

[0172] 3. TOH (Tool Head Marker), followed by tool information;

[0173] 4. RDI (Rendering Information Markup), followed by rendering information;

[0174] 5. SOQ (Start of Quality Graph), followed by the bitstream;

[0175] 6. SOZ (Start of Z-stream marker), followed by the bitstream of the super-prior information tensor z, including... and ;

[0176] 7. SORp (Start of Residual Stream Marker for Major Component), followed by the bitstream of the major component residual, including... ;

[0177] 8. SORs (Start of Residual Stream Markers for Secondary Components), followed by the bitstream of the secondary component residuals, including... ;

[0178] 9. EOC - End of stream marker.

[0179] The overall grammatical structure of the image is as follows:

[0180]

[0181]

[0182]

[0183]

[0184]

[0185] Each bitstream begins with a 16-bit marker. All markers used in this specification are as follows:

[0186]

[0187] When a SOQ flag exists and a second flag with a value in the range of 0xff83 to 0xff87 (inclusive) also exists, the substream associated with the second flag must precede the quality map substream associated with the SOQ flag in the decoding order.

[0188] When a tag with a value in the range of 0xff8c to 0xff8f (inclusive) exists, the substream associated with that tag must be after the minor component residual substream associated with the SORs tag in the decoding order.

[0189] Tags with reserved values ​​as indicated in the table above are reserved for future use by ISO / IEC and should not appear in this version of the JPEG bitstream conforming to the standard. When a reserved tag is encountered, the decoder of this version conforming to the standard must ignore the substream associated with that tag.

[0190] `rsv_ss_marker` specifies the marker for the substream. When a `reserved_substream()` syntax structure containing `rsv_ss_marker` is included within a `reserved_prefix_stream()` syntax structure, the value of `rsv_ss_marker` must be in the range of 0xff85 to 0xff87 (inclusive). When a `reserved_substream()` syntax structure containing `rsv_ss_marker` is included within a `reserved_suffix_stream()` syntax structure, the value of `rsv_ss_marker` must be in the range of 0xff8c to 0xff8e (inclusive).

[0191] The increment of 1 in rsv_ss_size_in_bytes_minus1 specifies the number of bytes used to represent the rsv_ss_size_val syntax element.

[0192] For this version of the JPEG bitstream conforming to this standard, `rsv_ss_reserved_zero_6bits` must be equal to 0. Values ​​greater than 0 for `rsv_ss_reserved_zero_6bits` are reserved for future use by ISO / IEC. When `rsv_ss_reserved_zero_6bits` is greater than 0, the decoder conforming to this version of the standard must ignore the `rsv_ss_reserved_zero_6bits` syntax element.

[0193] `rsv_ss_size_val` specifies the size (in bytes) of the substream payload (i.e., the substream excluding the syntax elements `rsv_ss_marker`, `rsv_ss_size_in_bytes_minus1`, `rsv_ss_reserved_zero_6bits`, and `rsv_ss_size_val`). The `rsv_ss_size_val` syntax element is represented by `rsv_ss_size_in_bytes_minus1 + 1` bytes.

[0194] reserved_ss_payload_byte (if it exists) has a value that will be specified by ISO / IEC in the future.

[0195] Figure 1This is a block diagram illustrating an example video processing system 4000 in which various techniques disclosed herein may be implemented. Various implementations may include some or all of the components of system 4000. System 4000 may include an input 4002 for receiving video content. The video content may be received in a raw or uncompressed format, such as 8 or 10-bit multi-component pixel values, or it may be received in a compressed or encoded format. Input 4002 may represent a network interface, a peripheral bus interface, or a storage interface. Examples of network interfaces include wired interfaces (such as Ethernet, Passive Optical Networking (PON), etc.) and wireless interfaces (such as Wi-Fi or cellular interfaces).

[0196] System 4000 may include an encoding component 4004 capable of implementing the various encoding / decoding or encoding methods described in this document. Encoding component 4004 can reduce the average bit rate from the video input 4002 to the output of encoding component 4004 to produce an encoded representation of the video. Encoding techniques are therefore sometimes referred to as video compression or video transcoding techniques. The output of encoding component 4004 may be stored or transmitted via a communication connection such as that represented by component 4006. The bitstream (or encoded) representation of the video received at input 4002, whether stored or transmitted via communication, may be used by component 4008 to generate pixel values ​​or displayable video to be transmitted to display interface 4010. The process of generating user-visible video from the bitstream representation is sometimes referred to as video decompression. Furthermore, although some video processing operations are referred to as “encoding” operations or tools, it should be understood that encoding tools or operations are used by the encoder, and the corresponding decoding tools or operations that reverse the encoded result will be performed by the decoder.

[0197] Examples of peripheral bus interfaces or display interfaces may include Universal Serial Bus (USB), High Definition Multimedia Interface (HDMI), or DisplayPort. Examples of storage interfaces include Serial Advanced Technology Attachment (SATA), Peripheral Component Interconnect (PCI), and Integrated Drive Electronics (IDE) interfaces. The technologies described in this document can be found in a variety of electronic devices, such as mobile phones, laptops, smartphones, or other devices capable of performing digital data processing and / or video display.

[0198] Figure 2This is a block diagram of an example video processing apparatus 4100. Apparatus 4100 can be used to implement one or more methods described herein. Apparatus 4100 can be embodied in a smartphone, tablet, computer, Internet of Things (IoT) receiver, etc. Apparatus 4100 may include one or more processors 4102, one or more memories 4104, and video processing circuitry 4106. The processors 4102(s) may be configured to implement one or more methods described herein. The memories 4104(s) may be used to store data and code for implementing the methods and techniques described herein. The video processing circuitry 4106 may be used to implement some of the techniques described herein in hardware circuitry. In some embodiments, the video processing circuitry 4106 may be at least partially included in the processor 4102, for example, a graphics coprocessor.

[0199] Figure 3 This is a flowchart of an example method 4200 for video processing. Method 4200 includes: in step 4202, determining data in a substream of a Joint Group of Image Experts (JPEG) Artificial Intelligence (AI) bitstream, the substream including an X-bit syntax element specifying the substream marker, a Y-bit syntax element specifying the length of the substream size, a Z-bit syntax element specifying the size (in bytes) of the substream payload, and the substream payload, where X, Y, and Z are integers greater than 0; and in step 4204, performing a conversion between visual media data and the JPEG AI bitstream based on the substream. The conversion may include encoding at the encoder, decoding at the decoder, or a combination thereof.

[0200] It should be noted that method 4200 can be implemented in an apparatus for processing video data, including a processor and a non-transitory memory having instructions thereon, such as a video encoder 4400, a video decoder 4500, and / or an encoder 4600. In this case, the instructions cause the processor to perform method 4200 when executed by the processor. Furthermore, method 4200 can be executed by a non-transitory computer-readable medium including a computer program product for use by a video encoding / decoding device. The computer program product includes computer-executable instructions stored on the non-transitory computer-readable medium, such that when the computer-executable instructions are executed by a processor, the video encoding / decoding device performs method 4200.

[0201] Figure 4 This is a block diagram illustrating an example video encoding / decoding system 4300 that can utilize the techniques of this disclosure. The video encoding / decoding system 4300 may include a source device 4310 and a target device 4320. The source device 4310 generates encoded video data, and this source device 4310 may be referred to as a video encoding device. The target device 4320 can decode the encoded video data generated by the source device 4310, and this target device 4320 may be referred to as a video decoding device.

[0202] Source device 4310 may include video source 4312, video encoder 4314, and input / output (I / O) interface 4316. Video source 4312 may include sources such as video capture devices, interfaces for receiving video data from video content providers, and / or computer graphics systems for generating video data, or combinations thereof. Video data may include one or more pictures. Video encoder 4314 encodes the video data from video source 4312 to generate a bitstream. The bitstream may include a sequence of bits forming a codec representation of the video data. The bitstream may include codec pictures and associated data. Codec pictures are codec representations of pictures. Associated data may include sequence parameter sets, picture parameter sets, and other syntax structures. I / O interface 4316 may include a modulator / demodulator (modem) and / or a transmitter. Encoded video data may be transmitted directly to target device 4320 via network 4330 through I / O interface 4316. Encoded video data may also be stored on storage medium / server 4340 for access by target device 4320.

[0203] Target device 4320 may include I / O interface 4326, video decoder 4324, and display device 4322. I / O interface 4326 may include a receiver and / or a modem. I / O interface 4326 may acquire encoded video data from source device 4310 or storage medium / server 4340. Video decoder 4324 may decode the encoded video data. Display device 4322 may display the decoded video data to a user. Display device 4322 may be integrated with target device 4320 or may be external to target device 4320, wherein target device 4320 may be configured to interface with an external display device.

[0204] The video encoder 4314 and the video decoder 4324 can operate according to video compression standards, such as the HEVC standard, the VVC standard, and other existing and / or further standards.

[0205] Figure 5 This is a block diagram illustrating an example of a video encoder 4400, wherein the video encoder 4400 may be... Figure 4 The system 4300 shown includes a video encoder 4314. The video encoder 4400 can be configured to perform any or all of the techniques disclosed herein. The video encoder 4400 includes multiple functional components. The techniques described in this disclosure can be shared among the various components of the video encoder 4400. In some examples, a processor can be configured to perform any or all of the techniques described in this disclosure.

[0206] The functional components of the video encoder 4400 may include a segmentation unit 4401, a prediction unit 4402 (which may include a mode selection unit 4403, a motion estimation unit 4404, a motion compensation unit 4405, and an intra-frame prediction unit 4406), a residual generation unit 4407, a transform processing unit 4408, a quantization unit 4409, an inverse quantization unit 4410, an inverse transform unit 4411, a reconstruction unit 4412, a buffer 4413, and an entropy coding unit 4414.

[0207] In other examples, the video encoder 4400 may include more, fewer, or different functional components. In one example, the prediction unit 4402 may include an intra-block copy (IBC) unit. The IBC unit can perform prediction in an IBC mode, where at least one reference picture is the picture in which the current video block is located.

[0208] Furthermore, some components such as the motion estimation unit 4404 and the motion compensation unit 4405 can be highly integrated, but for illustrative purposes, they are shown separately in the example of the video encoder 4400.

[0209] The segmentation unit 4401 can segment an image into one or more video blocks. The video encoder 4400 and the video decoder 4500 can support various video block sizes.

[0210] The mode selection unit 4403 can select one of several encoding / decoding modes (intra-frame encoding / decoding or inter-frame encoding / decoding), for example, based on error results, and provide the resulting intra-frame or inter-frame encoded / decoded block to the residual generation unit 4407 to generate residual block data, and to the reconstruction unit 4412 to reconstruct the coded block for use as a reference image. In some examples, the mode selection unit 4403 can select an intra-frame / inter-frame joint prediction (CIIP) mode, where prediction is based on inter-frame prediction signals and intra-frame prediction signals. In the case of inter-frame prediction, the mode selection unit 4403 can also select a resolution for the block based on motion vectors (e.g., sub-pixel precision or integer pixel precision).

[0211] To perform inter-frame prediction on the current video block, motion estimation unit 4404 can generate motion information for the current video block by comparing one or more reference frames from buffer 4413 with the current video block. Motion compensation unit 4405 can determine the predicted video block for the current video block based on the motion information and decoded samples of images from buffer 4413 other than the image associated with the current video block.

[0212] The motion estimation unit 4404 and the motion compensation unit 4405 can perform different operations on the current video block, for example, depending on whether the current video block is in an I-band, P-band, or B-band.

[0213] In some examples, motion estimation unit 4404 can perform unidirectional prediction on the current video block, and can search for a reference video block for the current video block in the reference images of list 0 or list 1. Motion estimation unit 4404 can then generate a reference index indicating the reference image containing the reference video block in list 0 or list 1, and a motion vector indicating the spatial displacement between the current video block and the reference video block. Motion estimation unit 4404 can output the reference index, prediction direction indicator, and motion vector as motion information for the current video block. Motion compensation unit 4405 can generate a predicted video block for the current block based on the reference video block indicated by the motion information of the current video block.

[0214] In other examples, motion estimation unit 4404 can perform bidirectional prediction on the current video block. Motion estimation unit 4404 can search for a reference video block for the current video block in the reference images in list 0, and can also search for another reference video block for the current video block in the reference images in list 1. Motion estimation unit 4404 can then generate reference indices indicating the reference images containing the reference video blocks in lists 0 and 1, and motion vectors indicating the spatial displacement between the reference video blocks and the current video block. Motion estimation unit 4404 can output the reference index and motion vector of the current video block as motion information for the current video block. Motion compensation unit 4405 can generate a predicted video block for the current video block based on the reference video blocks indicated by the motion information of the current video block.

[0215] In some examples, the motion estimation unit 4404 can output a complete set of motion information for use in the decoder's decoding process. In some examples, the motion estimation unit 4404 may not output a complete set of motion information for the current video. Instead, the motion estimation unit 4404 can reference the motion information of another video block to transmit the motion information of the current video block via a signal. For example, the motion estimation unit 4404 may determine that the motion information of the current video block is sufficiently similar to the motion information of neighboring video blocks.

[0216] In one example, the motion estimation unit 4404 may indicate a value to the video decoder 4500 in the syntax structure associated with the current video block, which indicates that the current video block has the same motion information as another video block.

[0217] In another example, motion estimation unit 4404 may identify another video block and motion vector difference (MVD) in the syntax structure associated with the current video block. The motion vector difference indicates the difference between the motion vector of the current video block and the motion vector of the indicated video block. Video decoder 4500 may use the motion vector of the indicated video block and the motion vector difference to determine the motion vector of the current video block.

[0218] As discussed above, the video encoder 4400 can transmit motion vectors via signaling in a predictive manner. Two examples of predictive signaling techniques that can be implemented by the video encoder 4400 include Advanced Motion Vector Prediction (AMVP) and Merge Pattern Signaling.

[0219] Intra-prediction unit 4406 can perform intra-prediction on the current video block. When intra-prediction unit 4406 performs intra-prediction on the current video block, it can generate prediction data for the current video block based on decoded samples from other video blocks in the same frame. The prediction data for the current video block can include the predicted video block and various syntax elements.

[0220] The residual generation unit 4407 can generate residual data for the current video block by subtracting (or more) predicted video blocks from the current video block. The residual data for the current video block may include residual video blocks corresponding to different sample components of the samples in the current video block.

[0221] In other examples, such as in skip mode, there may be no residual data for the current video block, and the residual generation unit 4407 may not perform subtraction operations.

[0222] The transform processing unit 4408 can generate one or more transform coefficient video blocks for the current video block by applying one or more transforms to the residual video blocks associated with the current video block.

[0223] After the transform processing unit 4408 generates a transform coefficient video block associated with the current video block, the quantization unit 4409 can quantize the transform coefficient video block associated with the current video block based on one or more quantization parameter (QP) values ​​associated with the current video block.

[0224] The inverse quantization unit 4410 and the inverse transform unit 4411 can apply inverse quantization and inverse transform to the transform coefficient video block respectively to reconstruct the residual video block from the transform coefficient video block. The reconstruction unit 4412 can add the reconstructed residual video block to the corresponding samples of one or more predicted video blocks generated by the prediction unit 4402 to generate a reconstructed video block associated with the current block and store it in the buffer 4413.

[0225] After the video block is reconstructed by reconstruction unit 4412, a loop filtering operation can be performed to reduce video block artifacts in the video block.

[0226] Entropy encoding unit 4414 can receive data from other functional components of video encoder 4400. When entropy encoding unit 4414 receives data, it can perform one or more entropy encoding operations to generate entropy-encoded data and output a bitstream including the entropy-encoded data.

[0227] Figure 6 This is a block diagram illustrating an example of a video decoder 4500, wherein the video decoder 4500 may be... Figure 4 The system 4300 shown includes a video decoder 4324. The video decoder 4500 can be configured to perform any or all of the techniques disclosed herein. In the example shown, the video decoder 4500 includes multiple functional components. The techniques described in this disclosure can be shared among the various components of the video decoder 4500. In some examples, a processor can be configured to perform any or all of the techniques described in this disclosure.

[0228] In the example shown, the video decoder 4500 includes an entropy decoding unit 4501, a motion compensation unit 4502, an intra-frame prediction unit 4503, an inverse quantization unit 4504, an inverse transform unit 4505, a reconstruction unit 4506, and a buffer 4507. In some examples, the video decoder 4500 can perform a decoding process that is generally contrasted with the encoding process described with respect to the video encoder 4400.

[0229] The entropy decoding unit 4501 can retrieve the encoded bitstream. The encoded bitstream may include entropy-encoded video data (e.g., encoded blocks of video data). The entropy decoding unit 4501 can decode the entropy-encoded video data, and based on the entropy-decoded video data, the motion compensation unit 4502 can determine motion information including motion vectors, motion vector precision, reference image list index, and other motion information. The motion compensation unit 4502 can determine this information, for example, by executing AMVP and Merge modes.

[0230] The motion compensation unit 4502 can generate motion compensation blocks and can perform interpolation based on an interpolation filter. The identifier of the interpolation filter to be used, with sub-pixel accuracy, can be included in the syntax element.

[0231] The motion compensation unit 4502 can use the interpolation filter used by the video encoder 4400 during the encoding of a video block to calculate the interpolation for sub-integer pixels of the reference block. The motion compensation unit 4502 can determine the interpolation filter used by the video encoder 4400 based on the received syntax information, and the motion compensation unit 4502 can use the interpolation filter to generate a prediction block.

[0232] The motion compensation unit 4502 may use some syntax information to determine the size of the blocks used to encode (multiple) frames and / or (multiple) stripes of the encoded video sequence, segmentation information describing how each macroblock of the image of the encoded video sequence is segmented, a mode indicating how each segment is encoded, one or more reference frames (and a list of reference frames) for each inter-frame codec block, and other information for decoding the encoded video sequence.

[0233] Intra-prediction unit 4503 can use, for example, an intra-prediction mode received in the bitstream to form prediction blocks from spatially adjacent blocks. Inverse quantization unit 4504 performs inverse quantization (i.e., dequantization) on the quantized video block coefficients provided in the bitstream and decoded by entropy decoding unit 4501. Inverse transform unit 4505 applies the inverse transform.

[0234] The reconstruction unit 4506 can add the residual block to the corresponding predicted block generated by the motion compensation unit 4502 or the intra-frame prediction unit 4503 to form a decoded block. If necessary, a deblocking filter can also be used to filter the decoded block to remove block artifacts. The decoded video block is then stored in a buffer 4507, which provides a reference block for subsequent motion compensation / intra-frame prediction, and the buffer 4507 also generates decoded video for presentation on a display device.

[0235] Figure 7 This is a schematic diagram of an example encoder 4600. Encoder 4600 is suitable for implementing VVC techniques. Encoder 4600 includes three loop filters: a deblocking filter (DF) 4602, a sample adaptive compensation (SAO) 4604, and an adaptive loop filter (ALF) 4606. Unlike DF 4602, which uses predefined filters, SAO 4604 and ALF 4606 utilize the original samples of the current image, reducing the mean square error between the original and reconstructed samples by adding an offset and by applying a finite impulse response (FIR) filter, respectively. The side information of the encoding and decoding is transmitted via the offset and filter coefficients in the signal transmission. ALF 4606 is located in the final processing stage of each image and can be considered as a tool for attempting to capture and repair artifacts caused by previous stages.

[0236] The encoder 4600 also includes an intra-frame prediction component 4608 and a motion estimation / compensation (ME / MC) component 4610 configured to receive input video. The intra-frame prediction component 4608 is configured to perform intra-frame prediction, while the ME / MC component 4610 is configured to perform inter-frame prediction using reference images obtained from a reference image buffer 4612. Residual blocks from inter-frame or intra-frame prediction are fed into a transform (T) component 4614 and a quantization (Q) component 4616 to generate quantized residual transform coefficients, which are then fed into an entropy coding component 4618. The entropy coding component 4618 entropy-codes the prediction results and the quantized transform coefficients and transmits them to a video decoder (not shown). The quantized components output from the quantization component 4616 can be fed into an inverse quantization (IQ) component 4620, an inverse transform component 4622, and a reconstruction (REC) component 4624. REC component 4624 is able to output images to DF 4602, SAO 4604 and ALF 4606 for filtering before these images are stored in reference image buffer 4612.

[0237] Figure 8 This is a flowchart of an example method 4700 for video processing. Method 4700 includes: in step 4702, determining that the Joint Picture Experts Group (JPEG) Artificial Intelligence (AI) bitstream contains one or more substreams, each substream beginning with a marker identifier (ID), followed by the substream's size and payload data; and in step 4704, performing a conversion between visual media data and the JPEG AI bitstream based on the marker ID, the substream's size, and the payload data. The conversion may include encoding at the encoder, decoding at the decoder, or a combination thereof.

[0238] It should be noted that method 4700 can be implemented in an apparatus for processing video data, including a processor and a non-transitory memory having instructions thereon, such as a video encoder 4400, a video decoder 4500, and / or an encoder 4600. In this case, the instructions, when executed by the processor, cause the processor to perform method 4700. Furthermore, method 4700 can be executed by a non-transitory computer-readable medium including a computer program product for use by a video codec device. The computer program product includes computer-executable instructions stored on the non-transitory computer-readable medium, causing the video codec device to perform method 4700 when executed by a processor.

[0239] The following is a list of some preferred solutions.

[0240] The following solutions illustrate examples of the techniques discussed in this article.

[0241] 1. A method for processing media data, comprising: determining data in a substream of a Joint Photo Experts Group (JPEG) Artificial Intelligence (AI) bitstream, wherein the substream includes an X-bit syntax element defining a substream marker, a Y-bit syntax element defining the length of the substream, a Z-bit syntax element defining the size of the substream payload in bytes, and a substream payload, wherein X, Y, and Z are integers greater than 0; and performing a conversion between visual media data and the JPEG AI bitstream based on the substream.

[0242] 2. The method described in Solution 1, wherein X equals 16, Y equals 2, and Z is determined by the value of Y.

[0243] 3. The method described according to solution 1 or 2, wherein Y equals 2, and the values ​​of Y 0, 1, 2, and 3 determine the values ​​of Z as 8, 16, 24, and 32, respectively.

[0244] 4. The method according to any one of solutions 1-3, wherein the X-bit syntax element specifying the substream marker is rsv_ss_marker, or wherein the substream marker indicates the type of the substream (including image header substream or tool header substream), or wherein the Y-bit syntax element specifying the length of the substream size is rsv_ss_size_len_in_bytes_minus1, or wherein the Z-bit syntax element specifying the size (in bytes) of the substream payload is rsv_ss_size_val.

[0245] 5. The method according to any one of solutions 1-4, wherein the substream further includes one or more byte alignment bits preceding the substream payload to ensure that the number of bits preceding the substream payload in the substream is an integer multiple of 8, or wherein each byte alignment bit is equal to 0, or wherein each byte alignment bit is reserved.

[0246] 6. The method according to any one of solutions 1-5, wherein the substream is associated with a marker preserved in JPEG AI version 1, and when the version 1 JPEG AI decoder encounters such a substream existing in the JPEG AI bitstream of another version of the JPEG AI standard, the decoder can parse the syntax elements rsv_ss_marker, rsv_ss_size_len_in_bytes_minus1, and rsv_ss_size_val, deduce the length of the substream payload, discard all data bytes of the substream payload, and then continue processing subsequent data bits in the JPEG AI bitstream.

[0247] 7. The method according to any one of solutions 1-6, wherein the existence of a substream in the JPEG AI stream depends on whether the value of the next M bits in the JPEG AI stream immediately following the syntax structure of the second substream is equal to the tag of the optional substream, or where M is equal to 16.

[0248] 8. The method according to any one of solutions 1-7, wherein the value indicating that a reserved substream marker should not appear in a first version of the JPEG bitstream conforming to the JPEG AI standard, or wherein when a reserved substream marker is encountered, the decoder of the first version of the JPEG AI standard must ignore the substream associated with the marker.

[0249] 9. The method according to any one of solutions 1-8, wherein substreams that may exist in the JPEG AI stream are defined as two categories, wherein substreams of the first category (when present) precede the q_stream() syntax structure (when present) in decoding order, and substreams of the second category (when present) follow the always-present r_secondary_stream() syntax structure in decoding order, and precede the always-present End of Stream (EOC) marker in decoding order.

[0250] 10. The method according to any one of solutions 1-9, wherein the optional sub-stream of the first category is referred to as a prefix sub-stream, or wherein the optional sub-stream of the second category is referred to as a suffix sub-stream, or wherein the optional sub-stream of the first category includes sub-streams whose sub-stream flag values ​​are in the range of 0xff83 to 0xff87 (inclusive), or wherein the optional sub-stream of the second category includes sub-streams whose sub-stream flag values ​​are in the range of 0xff8c to 0xff8e (inclusive).

[0251] 11. The method according to any one of solutions 1-10, wherein a type of prefix substream is allowed to precede or follow any other type of prefix substream in decoding order, or wherein a type of suffix substream is allowed to precede or follow any other type of suffix substream in decoding order.

[0252] 12. The method according to any one of solutions 1-11, wherein all reserved type sub-streams are optional sub-streams.

[0253] 13. The method according to any one of solutions 1-12, wherein when a quality graph start (SOQ) marker is present and a second marker having a value in the range of 0xff83 to 0xff87 (inclusive), the substream associated with the second marker must precede the quality graph substream (including the q_stream() syntax structure) associated with the SOQ marker in the decoding order.

[0254] 14. The method according to any one of solutions 1-13, wherein when a tag having a value in the range of 0xff8c to 0xff8f (inclusive) exists, the substream associated with the tag must be after the minor component residual substream associated with the SORs tag in the decoding order.

[0255] 15. The method according to any one of solutions 1-14, wherein the conversion includes encoding visual media data into a JPEG AI stream.

[0256] 16. The method according to any one of solutions 1-14, wherein the conversion includes decoding visual media data from a JPEG AI bitstream.

[0257] 17. An apparatus for processing video data, comprising a processor and a non-transitory memory having instructions thereon, wherein the instructions, when executed by the processor, cause the processor to perform the method of any one of solutions 1-16.

[0258] 18. A non-transitory computer-readable medium comprising a computer program product for use by a video codec device, the computer program product comprising computer-executable instructions stored on the non-transitory computer-readable medium, such that when the computer-executable instructions are executed by a processor, the video codec device performs the method of any one of solutions 1-16.

[0259] 19. A non-transitory computer-readable recording medium storing a Joint Photo Experts Group (JPEG) Artificial Intelligence (AI) bitstream of video generated by a method performed by a video processing apparatus, wherein the method comprises: determining data in a substream of the Joint Photo Experts Group (JPEG) Artificial Intelligence (AI) bitstream, wherein the substream includes an X-bit syntax element defining a substream marker, a Y-bit syntax element defining the length of the substream size, a Z-bit syntax element defining the size (in bytes) of the substream payload, and a substream payload, wherein X, Y, and Z are integers greater than 0; and generating the JPEG AI bitstream based on the determination.

[0260] 20. A method for storing a bitstream of video, comprising: determining data in a substream of a Joint Group of Image Experts (JPEG) Artificial Intelligence (AI) bitstream, wherein the substream includes an X-bit syntax element defining a substream marker, a Y-bit syntax element defining the length of the substream, a Z-bit syntax element defining the size of the substream payload in bytes, and a substream payload, wherein X, Y, and Z are integers greater than 0; generating the JPEG AI bitstream based on the determination; and storing the JPEG AI bitstream in a non-transitory computer-readable recording medium.

[0261] 21. A method, apparatus or system described in this document.

[0262] The following solutions illustrate further examples of the techniques discussed in this article.

[0263] 1. A method for processing video data, comprising: determining that a Joint Photo Experts Group (JPEG) Artificial Intelligence (AI) bitstream contains one or more substreams, wherein each substream begins with a marker identifier (ID), followed by the substream's size and payload data; and performing a conversion between visual media data and the JPEG AI bitstream based on the marker ID, the substream's size, and the payload data.

[0264] 2. The method according to Solution 1, wherein the tag ID is encoded and decoded into an unsigned sixteen-bit data element.

[0265] 3. The method according to solution 1 or 2, wherein each substream contains a byte-aligned syntax structure between the substream size and the payload data.

[0266] 4. The method according to any one of solutions 1-3, wherein each bit of the byte-aligned syntax structure is equal to 0.

[0267] 5. The method according to any one of solutions 1-4, wherein the tag ID and the size of the substream are configured such that when the decoder reads the tag ID and does not support the tag ID, the decoder must discard all remaining bits of the substream indicated by the size of the substream.

[0268] 6. The method according to any one of solutions 1-5, wherein the substream is configured such that when a reserved bitstream marker is encountered, the decoder must ignore subsequent bitstream segments.

[0269] 7. The method according to any one of solutions 1-6, wherein the reserved bitstream marker contains the least significant bit value in the range of 85 to 87 (inclusive).

[0270] 8. The method according to any one of solutions 1-7, wherein the tag ID is an X-bit syntax element specifying the substream tag, wherein each substream includes a Y-bit syntax element specifying the length of the substream size, wherein the substream size is a Z-bit syntax element specifying the size of the substream payload (in bytes), and wherein X, Y, and Z are integers greater than 0.

[0271] 9. The method according to any one of solutions 1-8, wherein X equals 16, Y equals 2, and Z is determined by the value of Y.

[0272] 10. The method according to any one of solutions 1-9, wherein Y equals 2, and the values ​​of Y 0, 1, 2, and 3 determine when the value of Z is 8, 16, 24, and 32, respectively.

[0273] 11. The method according to any one of solutions 1-10, wherein the X-bit syntax element is a marker ID and is represented as a reserved substream start marker (rsv_ss_marker), or wherein the marker ID indicates the type of substream including an image header substream or a tool header substream, or wherein the Y-bit syntax element is represented as the length (in bytes) of the reserved substream size minus 1 (rsv_ss_size_len_in_bytes_minus1), or wherein the Z-bit syntax element is represented as the reserved substream size value (rsv_ss_size_val).

[0274] 12. The method according to any one of solutions 1-11, wherein the byte-aligned syntax structure ensures that the number of bits preceding the payload data in the substream is an integer multiple of 8, or wherein each byte-aligned syntax structure bit is reserved.

[0275] 13. The method according to any one of solutions 1-12, wherein the existence of an optional substream in the JPEG AI bitstream depends on whether the value of the next M bits immediately following the syntax structure of the second substream in the JPEG AI bitstream is equal to the tag of the optional substream, or where M is equal to 16.

[0276] 14. The method according to any one of solutions 1-13, wherein the value indicates that a reserved substream marker should not appear in a first version of the JPEG AI bitstream conforming to the JPEG AI standard, or wherein when a reserved substream marker is encountered, the first version of the decoder conforming to the JPEG AI standard must ignore the substream associated with the reserved substream marker.

[0277] 15. The method according to any one of solutions 1-14, wherein substreams allowed to exist in the JPEG AI bitstream are defined as two categories, wherein substreams of the first category (when present) precede the quality stream (q_stream()) syntax structure (when present) in decoding order, and substreams of the second category (when present) follow the residual stream (r_secondary_stream()) syntax structure of the always-present second component in decoding order, and precede the end-of-stream (EOC) marker, which is also always present, in decoding order.

[0278] 16. The method according to any one of solutions 1-15, wherein the optional sub-stream of the first category is referred to as a prefix sub-stream, or wherein the optional sub-stream of the second category is referred to as a suffix sub-stream, or wherein the optional sub-stream of the first category includes sub-streams whose tag ID values ​​are in the range of 0xff83 to 0xff87 (inclusive), or wherein the optional sub-stream of the second category includes sub-streams whose tag ID values ​​are in the range of 0xff8c to 0xff8e (inclusive).

[0279] 17. The method according to any one of solutions 1-16, wherein a type of prefix substream is allowed to precede or follow any other type of prefix substream in decoding order, or wherein a type of suffix substream is allowed to precede or follow any other type of suffix substream in decoding order.

[0280] 18. The method according to any one of solutions 1-17, wherein all reserved type substreams are optional substreams.

[0281] 19. The method according to any one of solutions 1-18, wherein when a quality graph start (SOQ) marker is present and a second marker having a value in the range of 0xff83 to 0xff87 (inclusive), the substream associated with the second marker must precede the quality graph substream (including the q_stream() syntax structure) associated with the SOQ marker in the decoding order.

[0282] 20. The method according to any one of solutions 1-19, wherein when a tag ID with a value in the range of 0xff8c to 0xff8f (inclusive), the substream associated with the tag ID must, in the decoding order, follow the minor component residual substream associated with the minor component residual stream start (SORs) tag.

[0283] 21. The method according to any one of solutions 1-20, wherein the conversion includes encoding visual media data into a JPEG AI stream.

[0284] 22. The method according to any one of solutions 1-20, wherein the conversion includes decoding visual media data from a JPEG AI bitstream.

[0285] 23. An apparatus for processing video data, comprising a processor and a non-transitory memory having instructions thereon, wherein the instructions, when executed by the processor, cause the processor to perform the method of any one of claims 1-22.

[0286] 24. A non-transitory computer-readable medium comprising a computer program product for use by a video codec apparatus, wherein the computer program product comprises computer-executable instructions stored on the non-transitory computer-readable medium, such that when the computer-executable instructions are executed by a processor, the video codec apparatus performs the method of any one of claims 1-22.

[0287] 25. A non-transitory computer-readable recording medium storing a Joint Photo Experts Group (JPEG) Artificial Intelligence (AI) bitstream of video generated by a method performed by a video processing apparatus, wherein the method comprises: determining that the Joint Photo Experts Group (JPEG) Artificial Intelligence (AI) bitstream contains one or more substreams, wherein each substream begins with a tag identifier (ID), followed by substream size and payload data; and generating the JPEG AI bitstream based on the determination.

[0288] 26. A method for storing a bitstream of video, comprising: determining that a Joint Photo Experts Group (JPEG) Artificial Intelligence (AI) bitstream comprises one or more substreams, wherein each substream begins with a marker identifier (ID), followed by substream size and payload data; generating the JPEG AI bitstream based on the determination; and storing the JPEG AI bitstream in a non-transitory computer-readable recording medium.

[0289] In the described solution, the encoder conforms to the format rules by generating an encoded representation based on those rules. In the described solution, the decoder parses the syntax elements in the encoded representation using known information about their presence or absence, according to the format rules, to produce the decoded video.

[0290] In this document, the term "video processing" can refer to video encoding, video decoding, video compression, or video decompression. For example, a video compression algorithm can be applied during the conversion from the pixel representation of a video to the corresponding bitstream representation, and vice versa. For example, the bitstream representation of the current video block can correspond to bits at the same position in the bitstream defined by the syntax or bits propagated at different positions. For example, a macroblock can be encoded based on the error residual value after transformation and encoding / decoding, and can also use bits from the header and other fields in the bitstream. Furthermore, during the conversion, the decoder can parse the bitstream based on this determination, knowing whether some fields may be present or absent, as described in the solutions above. Similarly, the encoder can determine whether to include or exclude specific syntax fields and generate encoded representations accordingly by including or excluding syntax fields from the encoded representation.

[0291] The disclosed and other solutions, examples, embodiments, modules, and functional operations described in this document can be implemented in digital electronic circuits, or in computer software, firmware, or hardware, including the structures disclosed in this document and their structural equivalents, or in one or more combinations thereof. The disclosed embodiments and other embodiments can be implemented as one or more computer program products, i.e., one or more computer program instruction modules encoded on a computer-readable medium for execution by or control of the operation of a data processing apparatus. The computer-readable medium can be a machine-readable storage device, a machine-readable storage substrate, a storage device, a material composition affecting machine-readable propagation signals, or a combination thereof. The term "data processing apparatus" includes all means, devices, and machines for processing data, including, for example, a programmable processor, a computer, or multiple processors or computers. In addition to hardware, the apparatus may also include code that creates an execution environment for an associated computer program, such as code constituting processor firmware, a protocol stack, a database management system, an operating system, or a combination thereof. Propagation signals are artificially generated signals, such as machine-generated electrical signals, optical signals, or electromagnetic signals, which are generated to encode information for transmission to a suitable receiver device.

[0292] Computer programs (also known as programs, software, software applications, scripts, or code) can be written in any programming language, including compiled or interpreted languages, and can be deployed in any form, including standalone programs or modules, components, subroutines, or other units suitable for use in a computing environment. A computer program does not necessarily correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the related program, or in multiple co-located files (e.g., a file storing one or more modules, subroutines, or code portions). A computer program can be deployed to execute on one computer or on multiple computers located at a single site or distributed across multiple sites and interconnected by a communications network.

[0293] The processing and logic flows described in this document can be executed by one or more programmable processors that execute one or more computer programs to perform functions by manipulating input data and generating outputs. The processing and logic flows can also be executed by special-purpose logic circuitry, and the devices can be implemented as special-purpose logic circuitry, such as FPGAs (Field-Programmable Gate Arrays) or ASICs (Application-Specific Integrated Circuits).

[0294] Processors suitable for executing computer programs include, for example, general-purpose and special-purpose microprocessors, and any one or more processors of any type of digital computer. Typically, a processor receives instructions and data from read-only memory or random access memory, or both. The basic components of a computer are a processor that executes instructions and one or more storage devices that store the instructions and data. Typically, a computer will also include one or more mass storage devices for storing data, such as magnetic disks, magneto-optical disks, or optical disks, or receive data from or transfer data to one or more mass storage devices via operative coupling, or both. However, a computer does not necessarily have such devices. Computer-readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media, and storage devices, including, for example, semiconductor storage devices such as EPROM, EEPROM, and flash memory devices; magnetic disks, such as internal hard disks or removable hard disks; magneto-optical disks; and CD-ROMs and DVD-ROMs. Processors and memory may be supplemented by or incorporated into special-purpose logic circuitry.

[0295] While this patent document contains numerous details, these details should not be construed as limiting any subject matter or the scope of the claims, but rather as descriptions of features specific to particular embodiments of a particular technology. In this patent document, certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented separately in multiple embodiments, or in any suitable sub-combination. Furthermore, although features may function in certain combinations as described above, and even were originally claimed in this manner, in some cases one or more features in the claimed combination may be removed from that combination, and the claimed combination may be for sub-combinations or variations thereof.

[0296] Similarly, although operations are depicted in a specific order in the accompanying drawings, this should not be construed as requiring such operations to be performed sequentially in the particular order or sequence shown, or requiring all shown operations to be performed in order to achieve the desired result. Furthermore, the division of various system components in the embodiments described in this patent document should not be construed as requiring such division in all embodiments.

[0297] Only a few implementations and examples are described, and other implementations, improvements and variations can be made based on what is described and shown in this patent document.

[0298] When there is no intermediate component (other than a line, trace, or other medium between the first and second components), the first component is directly coupled to the second component. When there is an intermediate component between the first and second components other than a line, trace, or other medium, the first component is indirectly coupled to the second component. The term "coupled" and its variations include direct coupling and indirect coupling. The use of the term "about" means including a range of ±10% of the following figures, unless otherwise specified.

[0299] While several embodiments have been provided in this disclosure, it should be understood that the disclosed systems and methods may be embodied in many other specific forms without departing from the spirit or scope of this disclosure. The present examples are to be considered illustrative rather than restrictive and are not intended to be limited to the details set forth herein. For example, various elements or components may be combined or integrated into another system, or certain features may be omitted or not implemented.

[0300] Furthermore, the technologies, systems, subsystems, and methods described and illustrated as discrete or separate in the various embodiments can be combined or integrated with other systems, modules, technologies, or methods without departing from the scope of this disclosure. Other items shown or discussed as couplings can be directly connected or indirectly coupled or communicated through some interface, device, or intermediate component, whether electrically, mechanically, or otherwise. Those skilled in the art can identify examples of other changes, substitutions, and modifications, which can be made without departing from the spirit and scope of this disclosure.

Claims

1. A method for processing video data, comprising: The Joint Image Experts Group (JPEG) Artificial Intelligence (AI) bitstream is determined to contain one or more substreams, each of which begins with a tag identifier (ID), followed by the size and payload data of the substream; as well as The conversion between visual media data and the JPEG AI stream is performed based on the tag ID, the size of the substream, and the payload data.

2. The method as described in claim 1, wherein, The tag ID is encoded and decoded into an unsigned sixteen-bit data element.

3. The method as described in claim 1 or 2, wherein, Each substream contains a byte-aligned syntax structure between the substream's size and the payload data.

4. The method according to any one of claims 1-3, wherein, Each bit of the byte-aligned syntax structure is equal to 0.

5. The method according to any one of claims 1-4, wherein, The tag ID and the size of the substream are configured such that when the decoder reads the tag ID and does not support the tag ID, the decoder must discard all remaining bits of the substream as indicated by the size of the substream.

6. The method according to any one of claims 1-5, wherein, The substream is configured such that when a reserved bitstream marker is encountered, the decoder must ignore subsequent bitstream segments.

7. The method according to any one of claims 1-6, wherein, The reserved bitstream markers contain the lowest valid bit value in the range of 85 to 87 (inclusive).

8. The method according to any one of claims 1-7, wherein, The tag ID is an X-bit syntax element that specifies the substream tag, each substream includes a Y-bit syntax element that specifies the length of the substream's size, the substream's size is a Z-bit syntax element that specifies the size (in bytes) of the substream payload, and X, Y, and Z are integers greater than 0.

9. The method according to any one of claims 1-8, wherein, X equals 16, Y equals 2, and Z is determined by the value of Y.

10. The method according to any one of claims 1-9, wherein, Y equals 2, and the values ​​of Y 0, 1, 2, and 3 determine when the values ​​of Z are 8, 16, 24, and 32, respectively.

11. The method according to any one of claims 1-10, wherein, The X-bit syntax element is the marker ID and is represented as a reserved substream start marker (rsv_ss_marker), or the marker ID indicates the type of the substream including an image header substream or a toolhead substream, or the Y-bit syntax element is represented as the length (in bytes) of the reserved substream size minus 1 (rsv_ss_size_len_in_bytes_minus1), or the Z-bit syntax element is represented as the reserved substream size value (rsv_ss_size_val).

12. The method according to any one of claims 1-11, wherein, The byte alignment syntax structure ensures that the number of bits preceding the payload data in the substream is an integer multiple of 8, or that each byte alignment syntax structure bit in the byte alignment syntax structure is reserved.

13. The method according to any one of claims 1-12, wherein, The existence of an optional substream in the JPEG AI stream depends on whether the value of the next M bits immediately following the syntax structure of the second substream in the JPEG AI stream is equal to the flag of the optional substream, or M equals 16.

14. The method according to any one of claims 1-13, wherein, The value indicates that a reserved substream marker should not appear in a first version of the JPEG AI bitstream conforming to the JPEG AI standard, or that when a reserved substream marker is encountered, the first version of the decoder conforming to the JPEG AI standard must ignore the substream associated with the reserved substream marker.

15. The method according to any one of claims 1-14, wherein, Substreams allowed in the JPEG AI bitstream are defined as two categories, where substreams of the first category (when present) precede the quality stream (q_stream()) syntax structure (when present) in decoding order, and substreams of the second category (when present) follow the always-present residual stream (r_secondary_stream()) syntax structure in decoding order, and precede the always-present end-of-stream (EOC) marker in decoding order.

16. The method according to any one of claims 1-15, wherein, The first category of optional substreams is called prefix substreams, or the second category of optional substreams is called suffix substreams. Alternatively, the first category of optional substreams includes substreams whose tag ID values ​​are in the range of 0xff83 to 0xff87 (inclusive), or the second category of optional substreams includes substreams whose tag ID values ​​are in the range of 0xff8c to 0xff8e (inclusive).

17. The method according to any one of claims 1-16, wherein, Allow a prefix substream of one type to precede or follow any other prefix substream of another type in decoding order, or allow a suffix substream of one type to precede or follow any other suffix substream of another type in decoding order.

18. The method according to any one of claims 1-17, wherein, All reserved type substreams are optional substreams.

19. The method according to any one of claims 1-18, wherein, When a Start of Quality Graph (SOQ) marker is present and a second marker with a value in the range of 0xff83 to 0xff87 (inclusive) is also present, the substream associated with the second marker must precede the quality graph substream in the decoding order, the quality graph substream comprising the q_stream() syntax structure and associated with the SOQ marker.

20. The method according to any one of claims 1-19, wherein, When a tag ID with a value in the range of 0xff8c to 0xff8f (inclusive) exists, the substream associated with the tag ID must be after the minor component residual substream in the decoding order, and the minor component residual substream is associated with the minor component's Residual Stream Start (SOR) tag.

21. The method according to any one of claims 1-20, wherein, The conversion includes encoding the visual media data into the JPEG AI stream.

22. The method according to any one of claims 1-20, wherein, The conversion includes decoding the visual media data from the JPEG AI bitstream.

23. An apparatus for processing video data, comprising a processor and a non-transitory memory having instructions thereon, wherein, When executed by the processor, the instructions cause the processor to perform the method of any one of claims 1-22.

24. A non-transitory computer-readable medium comprising a computer program product for use with a video encoding / decoding device, wherein, The computer program product includes computer-executable instructions stored on the non-transitory computer-readable medium, such that when the computer-executable instructions are executed by a processor, the video codec device performs the method of any one of claims 1-22.

25. A non-transitory computer-readable recording medium storing a Joint Image Experts Group (JPEG) artificial intelligence (AI) bitstream of video generated by a method performed by a video processing apparatus, wherein, The method includes: The Joint Photographic Experts Group (JPEG) Artificial Intelligence (AI) bitstream is determined to contain one or more substreams, each substream beginning with a tag identifier (ID), followed by the substream's size and payload data; and The JPEG AI stream is generated based on the determination.

26. A method for storing a video bitstream, comprising: The Joint Image Experts Group (JPEG) Artificial Intelligence (AI) bitstream is determined to contain one or more substreams, each of which begins with a tag identifier (ID), followed by the size and payload data of the substream; Based on the determination, the JPEG AI bitstream is generated; and The JPEG AI stream is stored in a non-transitory computer-readable recording medium.

Images