Method and apparatus for signaling the number of channels in a bitstream

Abstract

The present disclosure provides an improved decoding method and improved method of encoding a bitstream to ensure backwards compatibility of bitstreams with legacy decoders. The decoding method (1400) comprises: receiving (1401) a bitstream including encoded data of an input signal, wherein the bitstream is formed of multiple substreams, a first substream comprising a parameter indicating a number of channels in a second substream for a component of an image coding format of the bitstream; parsing (1402) the first substream, the parsing of the first substream comprising parsing syntax elements of the parameter to thereby determine the number of channels in the second substream for the component of the image coding format; and parsing (1403) the second substream in dependence on the determined number of channels in the second substream for the component of the image coding format. This may allow the number of channels to be decoded in the second bitstream to be signaled in the first bitstream and allow the number of channels used for reconstruction to be reduced.

Description

[0001] METHOD AND APPARATUS FOR SIGNALING THE NUMBER OF CHANNELS IN A BITSTREAM

[0002] TECHNICAL FIELD

[0003] Embodiments of the present disclosure generally relate to the field of encoding and decoding databased on a neural network architecture. In particular, some embodiments relate to methods and apparatuses for such encoding and decoding images and / or videos from a bitstream using a plurality of processing layers.

[0004] BACKGROUND

[0005] Hybrid image and video codecs have been used for decades to compress image and video data. In such codecs, signal is typically encoded block- wisely by predicting a block and by further coding only the difference between the original bock and its prediction. In particular, such coding may include transformation, quantization and generating the bitstream, usually including some entropy coding. Typically, the three components of hybrid coding methods - transformation, quantization, and entropy coding - are separately optimized. Modem video compression standards like High-Efficiency Video Coding (HEVC), Versatile Video Coding (VVC) and Essential Video Coding (EVC) also use transformed representation to code residual signal after prediction.

[0006] Recently, neural network architectures have been applied to image and / or video coding. In general, these neural network (NN) based approaches can be applied in various different ways to the image and video coding. For example, some end- to-end optimized image or video coding frameworks have been discussed. Moreover, deep learning has been used to determine or optimize some parts of the end-to-end coding framework such as selection or compression of prediction parameters or the like. Besides, some neural network based approached have also been discussed for usage in hybrid image and video coding frameworks, e.g. for implementation as a trained deep learning model for intra or inter prediction in image or video coding.

[0007] The end-to-end optimized image or video coding applications discussed above have in common that they produce some feature map data, which is to be conveyed between encoder and decoder.

[0008] Neural networks are machine learning models that employ one or more layers of nonlinear units based on which they can predict an output for a received input. Some neural networks include one or more hidden layers in addition to an output layer. A corresponding feature map may be provided as an output of each hidden layer. Such corresponding feature map of each hidden layer may be used as an input to a subsequent layer in the network, i.e., a subsequent hidden layer or the output layer. Each layer of the network generates an output from a received input in accordance with current values of a respective set of parameters. In a neural network that is split between devices, e.g. between encoder and decoder, a device and a cloud or between different devices, a feature map at the output of the place of splitting (e.g. a first device) is compressed and transmitted to the remaining layers of the neural network (e.g. to a second device).

[0009] Further improvement of encoding and decoding using trained network architectures may be desirable.

[0010] SUMMARY

[0011] The foregoing and other objectives are achieved by the subject matter of the independent claims. Further implementation forms are apparent from the dependent claims, the description and the figures.

[0012] Particular embodiments are outlined in the attached independent claims, with other embodiments in the dependent claims. According to a first aspect, the present disclosure relates to a decoding method, implemented by a decoder, the method comprising: receiving a bitstream including encoded data of an input signal, wherein the bitstream is formed of multiple substreams, a first substream of the multiple substreams comprising a parameter indicating a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream; parsing the first substream, the parsing of the first substream comprising parsing syntax elements of the parameter to thereby determine the number of channels in the second substream for the component of the image coding format of the bitstream; and parsing the second substream in dependence on the determined number of channels in the second substream for the component of the image coding format of the bitstream.

[0013] This may allow the number of channels to be decoded in the second bitstream to be signaled in the first bitstream. This may allow the number of channels used for reconstruction to be reduced, for example by a transcoder, without actual reconstruction and may allow to reduce the size of the stream by removing unused channels of residuals.

[0014] The foregoing and other embodiments can each optionally include one or more of the following features, alone or in combination.

[0015] In a possible implementation, the second substream is a residual substream for the component of the image coding format of the bitstream. This may allow the number of residual channels to be decoded to be signaled.

[0016] In a possible implementation, the component of the image coding format may be indicated by a component value. If the component value is 0, the parameter may relate to a number of residual channels in a primary residual substream. If the component value is 1 , the parameter may relate to a number of residual channels in a secondary residual substream. This may allow the number of residual channels to be decoded for each component to be signaled.

[0017] In a possible implementation, the parameter is signaled as 8 bits for a primary component of the image coding format and 7 bits for a secondary component of the image coding format of the bitstream. This may be a convenient implementation that can result in a bit saving. For example where the primary component is luma, this component has 160 channels and costs 8 bits, and where the secondary component is chroma, this component has 96 channels, and 7 bits can be used to encode the number.

[0018] In a possible implementation, the parameter is signaled as 8 bits for each of a primary component of the image coding format and a secondary component of the image coding forma of the bitstream. This may be a convenient implementation, as 8 bits can be signaled for both the primary and secondary components.

[0019] In a possible implementation, the component of the image coding format is luma or chroma. This may allow the number of channels to be decoded for different color components.

[0020] In a possible implementation, the parameter is specified in a field num_chs[comp] in the first substream, where comp is the component of the image coding format of the bitstream. This may allow the number of channels to be decoded to be signaled for each component.

[0021] In a possible implementation, the maximum value of the field num_chs[comp] is 160 where comp is equal to 0 (and / or for the primary component) and 96 where comp is equal to 1 (and / or for the secondary component). In a possible implementation, the first substream is a picture header substream of the bitstream. This may allow the parameter to be included in a header containing other information relevant to the image.

[0022] In a possible implementation, each of the multiple substreams comprises an Marker ID indicating a data type of the respective substream, wherein the marker ID for the picture header substream is PIH and wherein the parsing comprises parsing the marker ID of the picture header substream. This may allow the decoder to identify the picture image header during decoding.

[0023] In a possible implementation, each of the multiple substreams comprises an Marker ID indicating a data type of the respective substream, wherein the marker ID for the second substream is SORc, where c indicates the component of the image coding format of the bitstream. This may allow the residual substream for the component to be identified.

[0024] In a possible implementation, the method further comprises, subsequent to determining the number of channels in the second substream for the component of an image coding format of the bitstream, configuring the parsing process for the second substream. The method may comprise limiting a number of channels to be parsed from the second substream. Only num_chs[comp] number of channels may be parsed from the second substream. All other channels of the residual tensor in the second substream can be zeroed (i.e. set to zero).

[0025] In a possible implementation, the method further comprises, subsequent to determining the number of channels in the second substream for the component of an image coding format of the bitstream, configuring the parsing process for the array grfs_channel_flag[comp] and limiting the number of parsed elements to num_chs[comp] . All other elements of the array grfs_channel_flag[comp] can be zeroed.

[0026] According to second aspect, the present disclosure related to a device for decoding data for picture or video processing from a bitstream, the device comprising: a receiving unit configured to receive a bitstream including encoded data of an input signal, wherein the bitstream is formed of multiple substreams, a first substream of the multiple substreams comprising a parameter indicating a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream; and a parsing unit configured to: parse the first substream, the parsing of the first substream comprising parsing syntax elements of the parameter to thereby determine the number of channels in the second substream for the component of the image coding format of the bitstream; and parse the second substream in dependence on the determined number of channels in the second substream for the component of the image coding format of the bitstream.

[0027] The method according to the first aspect of the present disclosure may be performed by the apparatus according to the second aspect of the present disclosure. Further features and implementations of the method according to the first aspect of the present disclosure correspond to respective features and implementations of the apparatus according to the second aspect of the present disclosure. The advantages of the method according to the first aspect can be the same as those for the corresponding implementation of the apparatus according to the second aspect.

[0028] According to a third aspect, the present disclosure relates to an encoding method, implemented by an encoder, the method comprising: receiving an input signal; and encoding the input signal into a bitstream, the bitstream being formed of multiple substreams, a first substream of the multiple substreams comprising a parameter indicating a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream. The foregoing and other embodiments can each optionally include one or more of the following features, alone or in combination.

[0029] In a possible implementation, the encoding comprises generating the parameter. This may allow the encoder to signal the number of channels to be decoded to the decoder.

[0030] In a possible implementation, the second substream is a residual substream for the component of the image coding format of the bitstream. This may allow the number of residual channels to be decoded to be signaled to the deocder.

[0031] In a possible implementation, the component of the image coding format is indicated by a component value, wherein if the component value is 0, the parameter relates to a number of residual channels in a primary residual substream. In a possible implementation, the component of the image coding format is indicated by a component value, wherein if the component value is 1 , the parameter relates to a number of residual channels in a second residual substream. This may allow the number of residual channels to be decoded for each component to be signaled.

[0032] In a possible implementation, the parameter is signaled as 8 bits for a primary component of the image coding format and 7 bits for a secondary component of the image coding format of the bitstream. This may be a convenient implementation that can result in a bit saving. For example where the primary component is luma, this component has 160 channels and costs 8 bits, and where the secondary component is chroma, this component has 96 channels, and 7 bits can be used to encode the number.

[0033] In a possible implementation, the parameter is signaled as 8 bits for each of a primary component of the image coding format and a secondary component of the image coding forma of the bitstream. This may be a convenient implementation, as 8 bits can be signaled for both the primary and secondary components.

[0034] In a possible implementation, the component of the image coding format is luma or chroma. The primary component may be luma. The secondary component may be chroma. This may allow the parameter to be signaled for each component.

[0035] In a possible implementation, the parameter is specified in a field num_chs[comp] in the first substream, where comp is the component of the image coding format of the bitstream. This may allow the number of channels to be decoded to be signaled for each component.

[0036] In a possible implementation, the maximum value of the field num_chs[comp] is 160 where comp is equal to 0 (and / or for the primary component) and 96 where comp is equal to 1 (and / or for the secondary component).

[0037] In a possible implementation, the method comprising for each substream of the multiple substreams: generating an Marker ID indicating a data type encoded within the substream; and inserting the Marker ID within the substream. This may allow the encoder to encode data that allows the decoder to identify the different substreams during decoding.

[0038] In a possible implementation, the first substream is a picture header substream of the bitstream. In a possible implementation, the marker ID for the picture header substream is PIH. This may allow the parameter to be included in a header containing other information relevant to the image. This may allow the encoder to indicate the picture header so that the decoder can identify the picture image header during decoding. In a possible implementation, the marker ID for the second substream is SORc, where c indicates the component of the image coding format of the bitstream. This may allow the residual substream for the component to be identified.

[0039] In a possible implementation, the bitstream comprises one or more substreams, the method comprising for each substream of the one or more substreams: generating an Marker ID indicating a data type encoded within the substream; and inserting the Marker ID within the substream. This may allow the encoder to encode data that allows the decoder to identify the different substreams during decoding.

[0040] According to a fourth aspect, the present disclosure relates to a device for encoding data for picture or video processing to obtain a bitstream, the device comprising: a receiving unit configured to receive an input signal; and an encoding unit configured to encode the input signal into a bitstream, the bitstream being formed of multiple substreams, a first substream of the multiple substreams comprising a parameter indicating a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream.

[0041] The method according to the third aspect of the present disclosure may be performed by the apparatus according to the fourth aspect of the present disclosure. Further features and implementations of the method according to the third aspect of the present disclosure correspond to respective features and implementations of the apparatus according to the fourth aspect of the present disclosure . The advantages of the method according to the third aspect can be the same as those for the corresponding implementation of the apparatus according to the fourth aspect.

[0042] The decoding and encoding methods above may be used in progressive decoding.

[0043] According to a fifth aspect, the present disclosure relates to a decoder comprising one or more processors and a non- transitory computer-readable storage medium coupled to the one or more processors, wherein the storage medium stores programming for execution by the one or more processors, wherein the programming, when executed by the one or more processors, configures the decoder to carry out a method as set out above.

[0044] According to a sixth aspect, the present disclosure relates to a storage medium storing in non-transitory form a bitstream encoded by a method as set out above.

[0045] According to a seventh aspect, the present disclosure relates to a computer program stored in non-transitory form and including code instructions, which, when executed on one or more processor, causes the one or more processor to execute a method as set out above.

[0046] According to an eighth aspect, the present disclosure relates to a system for delivering a bitstream, the system including at least one storage medium configured to store at least one bitstream generated by an encoding method as set out above.

[0047] In a possible implementation, the system comprises a video streaming device configured to obtain the bitstream from the at least one storage medium and send the bitstream to a terminal device, wherein the video streaming device comprises a content server or a content delivery server.

[0048] In a possible implementation the system further comprises one or more processors configured to perform encryption processing on at least one bitstream to obtain at least one encrypted bitstream, the at least one storage medium configured to store the encrypted bitstream; or the one or more processor configured to convert the bitstream in a first format into a bitstream in a second format, the at least one storage medium configured to store the bitstream in the second format. The storage medium can then store a usefully encoded form of data.

[0049] In a possible implementation the system further comprises: a receiver configured to receive a first operation request; wherein the one or more processor is configured to determine a target bitstream in the at least one storage medium in response to the first operation request; a transmitter configured to send the target bitstream to a terminal-side apparatus. The target bitstream can then be processed by the terminal side apparatus.

[0050] In a possible implementation the one or more processor is further configured to encapsulate the bitstream to obtain a transport stream in a first format, wherein the transmitter is further configured to: send the transport stream in the first format to a terminal-side apparatus for display; or send the transport stream in the first format to storage space for storage. In this way the transport stream can be processed as appropriate.

[0051] According to a further aspect, there is provided a decoding apparatus comprising processing circuitry configured to execute steps of the decoding method having any of the features described above.

[0052] According to a further aspect, there is provided an encoding apparatus comprising processing circuitry configured to execute steps of the encoding method having any of the features described above.

[0053] According to a further aspect, there is provided a non-transitory storage medium comprising a bitstream encoded by the encoding method having any of the features described above.

[0054] According to a further aspect, there is provided a bitstream including encoded data of an input signal, wherein the bitstream is formed of multiple substreams, a first substream of the multiple substreams comprising a parameter indicating a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream.

[0055] BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In the following embodiments of the present disclosure are described in more detail with reference to the attached figures and drawings, in which:

[0057] Fig. 1 is a schematic drawing illustrating channels processed by layers of a neural network;

[0058] Fig. 2 is a schematic drawing illustrating an autoencoder type of a neural network;

[0059] Fig. 3 A is a schematic drawing illustrating an exemplary network architecture for encoder and decoder side including a hyperprior model;

[0060] Fig. 3B is a schematic drawing illustrating a general network architecture for encoder side including a hyperprior model;

[0061] Fig. 3C is a schematic drawing illustrating a general network architecture for decoder side including a hyperprior model;

[0062] Fig. 4 is a schematic drawing illustrating an exemplary network architecture for encoder and decoder side including a hyperprior model;

[0063] Fig. 5 is a block diagram illustrating a structure of a cloud-based solution for machine based tasks such as machine vision tasks;

[0064] Fig. 6A is a block diagram illustrating end-to-end video compression framework based on a neural networks;

[0065] Fig. 6B is a block diagram illustrating some exemplary details of application of a neural network for motion field compression; Fig. 6C is a block diagram illustrating some exemplary details of application of a neural network for motion compensation;

[0066] Fig. 7 is a block diagram illustrating an example of an encoding apparatus or a decoding apparatus;

[0067] Fig. 8 is a block diagram illustrating another example of an encoding apparatus or a decoding apparatus;

[0068] Fig. 9 shows an exemplary codestream structure for a bitstream;

[0069] Fig. 10 schematically illustrates an example of a picture header substream;

[0070] Fig. I la schematically illustrates an example of a primary residual substream;

[0071] Fig. 11b schematically illustrates an example of a secondary residual substream;

[0072] Fig. 12 shows proposed amendments to the picture header substream to include the parameter num_chs[comp];

[0073] Fig. 13 schematically illustrates the inclusion of the parameter num_chs[comp] in the picture header substream;

[0074] Fig. 14 is a flow diagram illustrating an exemplary method for decoding data such as image or video data or a component thereof;

[0075] Fig. 15 is a flow diagram illustrating an exemplary method for encoding data such as image or video data or a component thereof;

[0076] Fig. 16 shows a device for decoding for processing by a neural network-based unit;

[0077] Fig. 17 shows a device for encoding for processing by a neural network-based unit;

[0078] Fig. 18 is a block diagram showing an example of a video coding system configured to implement embodiments of the present disclosure;

[0079] Fig. 19 is a block diagram showing another example of a video coding system configured to implement embodiments of the present disclosure;

[0080] Fig. 20 is a block diagram illustrating an example of an encoding apparatus or a decoding apparatus;

[0081] Fig. 21 is a block diagram illustrating another example of an encoding apparatus or a decoding apparatus;

[0082] Fig. 22 is a block diagram illustrating another example of an encoding apparatus or a decoding apparatus.

[0083] Like reference numbers and designations in different drawings may indicate similar elements.

[0084] DETAILED DESCRIPTION OF THE EMBODIMENTS

[0085] In the following description, reference is made to the accompanying figures, which form part of the disclosure, and which show, by way of illustration, specific aspects of embodiments of the present disclosure or specific aspects in which embodiments of the present disclosure may be used. It is understood that embodiments of the present disclosure may be used in other aspects and comprise structural or logical changes not depicted in the figures. The following detailed description, therefore, is not to be taken in a limiting sense, and the scope of the present disclosure is defined by the appended claims.

[0086] For instance, it is understood that a disclosure in connection with a described method may also hold true for a corresponding device or system configured to perform the method and vice versa. For example, if one or a plurality of specific method steps are described, a corresponding device may include one or a plurality of units, e.g. functional units, to perform the described one or plurality of method steps (e.g. one unit performing the one or plurality of steps, or a plurality of units each performing one or more of the plurality of steps), even if such one or more units are not explicitly described or illustrated in the figures. On the other hand, for example, if a specific apparatus is described based on one or a plurality of units, e.g. functional units, a corresponding method may include one step to perform the functionality of the one or plurality of units (e.g. one step performing the functionality of the one or plurality of units, or a plurality of steps each performing the functionality of one or more of the plurality of units), even if such one or plurality of steps are not explicitly described or illustrated in the figures. Further, it is understood that the features of the various exemplary embodiments and / or aspects described herein may be combined with each other, unless specifically noted otherwise. In the following, an overview over some of the used technical terms and framework within which the embodiments of the present disclosure may be employed is provided.

[0087] Artificial neural networks

[0088] Artificial neural networks (ANN) or connectionist systems are computing systems vaguely inspired by the biological neural networks that constitute animal brains. Such systems "learn" to perform tasks by considering examples, generally without being programmed with task-specific rules. For example, in image recognition, they might learn to identify images that contain cats by analyzing example images that have been manually labeled as "cat" or "no cat" and using the results to identify cats in other images. They do this without any prior knowledge of cats, for example, that they have fur, tails, whiskers and cat-like faces. Instead, they automatically generate identifying characteristics from the examples that they process.

[0089] An ANN is based on a collection of connected units or nodes called artificial neurons, which loosely model the neurons in a biological brain. Each connection, like the synapses in a biological brain, can transmit a signal to other neurons. An artificial neuron that receives a signal then processes it and can signal neurons connected to it.

[0090] In ANN implementations, the "signal" at a connection is a real number, and the output of each neuron is computed by some non-linear function of the sum of its inputs. The connections are called edges. Neurons and edges typically have a weight that adjusts as learning proceeds. The weight increases or decreases the strength of the signal at a connection. Neurons may have a threshold such that a signal is sent only if the aggregate signal crosses that threshold. Typically, neurons are aggregated into layers. Different layers may perform different transformations on their inputs. Signals travel from the first layer (the input layer), to the last layer (the output layer), possibly after traversing the layers multiple times.

[0091] The original goal of the ANN approach was to solve problems in the same way that a human brain would. Over time, attention moved to performing specific tasks, leading to deviations from biology. ANNs have been used on a variety of tasks, including computer vision, speech recognition, machine translation, social network filtering, playing board and video games, medical diagnosis, and even in activities that have traditionally been considered as reserved to humans, like painting.

[0092] The name “convolutional neural network” (CNN) indicates that the network employs a mathematical operation called convolution. Convolution is a specialized kind of linear operation. Convolutional networks are neural networks that use convolution in place of a general matrix multiplication in at least one of their layers.

[0093] Fig. 1 schematically illustrates a general concept of processing by a neural network such as the CNN. A convolutional neural network consists of an input and an output layer, as well as multiple hidden layers. Input layer is the layer to which the input (such as a portion 11 of an input image as shown in Fig. 1 ) is provided for processing. The hidden layers of a CNN typically consist of a series of convolutional layers that convolve with a multiplication or other dot product. The result of a layer is one or more feature maps (illustrated by empty solid-line rectangles), sometimes also referred to as channels. There may be a resampling (such as subsampling) involved in some or all of the layers. As a consequence, the feature maps may become smaller, as illustrated in Fig. 1. It is noted that a convolution with a stride may also reduce the size (resample) an input feature map. The activation function in a CNN is usually a ReLU (Rectified Linear Unit) layer or Leaky ReLU, and is subsequently followed by additional convolutions such as pooling layers, fully connected layers and normalization layers, referred to as hidden layers because their inputs and outputs are masked by the activation function and final convolution. Though the layers are colloquially referred to as convolutions, this is only by convention. Mathematically, it is technically a sliding dot product or cross-correlation. This has significance for the indices in the matrix, in that it affects how the weight is determined at a specific index point.

[0094] When programming a CNN for processing images, as shown in Fig . 1 , the input is a tensor with shape (number of images) x (image width) x (image height) x (image depth). It should be known that the image depth can be constituted by channels of an image. After passing through a convolutional layer, the image becomes abstracted to a feature map, with shape (number of images) x (feature map width) x (feature map height) x (feature map channels). A convolutional layer within a neural network should have the following attributes.

[0095] Convolutional kernels defined by a width and height (hyper-parameters). The number of input channels and output channels (hyper-parameter). The depth of the convolution filter (the input channels) should be equal to the number channels (depth) of the input feature map.

[0096] In the past, traditional multilayer perceptron (MLP) models have been used for image recognition. However, due to the full connectivity between nodes, they suffered from high dimensionality, and did not scale well with higher resolution images. A 1000xl000-pixel image with RGB color channels has 3 million weights, which is too high to feasibly process efficiently at scale with full connectivity. Also, such network architecture does not take into account the spatial structure of data, treating input pixels which are far apart in the same way as pixels that are close together. This ignores locality of reference in image data, both computationally and semantically. Thus, full connectivity of neurons is wasteful for purposes such as image recognition that are dominated by spatially local input patterns.

[0097] Convolutional neural networks are biologically inspired variants of multilayer perceptrons that are specifically designed to emulate the behavior of a visual cortex. These models mitigate the challenges posed by the MLP architecture by exploiting the strong spatially local correlation present in natural images. The convolutional layer is the core building block of a CNN. The layer's parameters consist of a set of learnable filters (the above-mentioned kernels), which have a small receptive field, but extend through the full depth of the input volume. During the forward pass, each filter is convolved across the width and height of the input volume, computing the dot product between the entries of the filter and the input and producing a 2-dimensional activation map of that filter. As a result, the network learns filters that activate when it detects some specific type of feature at some spatial position in the input.

[0098] Stacking the activation maps for all filters along the depth dimension forms the full output volume of the convolution layer. Every entry in the output volume can thus also be interpreted as an output of a neuron that looks at a small region in the input and shares parameters with neurons in the same activation map. A feature map, or activation map, is the output activations for a given filter. Feature map and activation has same meaning. In some papers it is called an activation map because it is a mapping that corresponds to the activation of different parts of the image, and also a feature map because it is also a mapping of where a certain kind of feature is found in the image. A high activation means that a certain feature was found.

[0099] Another important concept of CNNs is pooling, which is a form of non-linear down- sampling. There are several nonlinear functions to implement pooling among which max pooling is the most common. It partitions the input image into a set of non-overlapping rectangles and, for each such sub-region, outputs the maximum.

[0100] Intuitively, the exact location of a feature is less important than its rough location relative to other features. This is the idea behind the use of pooling in convolutional neural networks. The pooling layer serves to progressively reduce the spatial size of the representation, to reduce the number of parameters, memory footprint and amount of computation in the network, and hence to also control overfitting. It is common to periodically insert a pooling layer between successive convolutional layers in a CNN architecture. The pooling operation provides another form of translation invariance.

[0101] The pooling layer operates independently on every depth slice of the input and resizes it spatially. The most common form is a pooling layer with filters of size 2x2 applied with a stride of 2 at every depth slice in the input by 2 along both width and height, discarding 75% of the activations. In this case, every max operation is over 4 numbers. The depth dimension remains unchanged. In addition to max pooling, pooling units can use other functions, such as average pooling or £2-norm pooling. Average pooling was often used historically but has recently fallen out of favour compared to max pooling, which often performs better in practice. Due to the aggressive reduction in the size of the representation, there is a recent trend towards using smaller filters or discarding pooling layers altogether. "Region of Interest" pooling (also known as ROI pooling) is a variant of max pooling, in which output size is fixed and input rectangle is a parameter. Pooling is an important component of convolutional neural networks for object detection based on Fast R-CNN architecture.

[0102] The above-mentioned ReLU is the abbreviation of rectified linear unit, which applies the non-saturating activation function. It effectively removes negative values from an activation map by setting them to zero. It increases the nonlinear properties of the decision function and of the overall network without affecting the receptive fields of the convolution layer. Other functions are also used to increase nonlinearity, for example the saturating hyperbolic tangent and the sigmoid function. ReLU is often preferred to other functions because it trains the neural network several times faster without a significant penalty to generalization accuracy.

[0103] Leaky Rectified Linear Unit, or Leaky ReLU, is a type of activation function based on a ReLU, but it has a small slope for negative values instead of a flat slope. The slope coefficient is determined before training, i.e. it is not learnt during training. This type of activation function is popular in tasks where it suffers from sparse gradients, for example training generative adversarial networks. Leaky ReLU applies the element-wise function:

[0104] LeakyReLU(x)=max(0,x)+negative_slope*min(0

[0105] Leaky

[0106] JReLU(x)

[0107] Among them, parameters: negative_slope - Controls the angle of the negative slope. Default: le-2 inplace - can optionally do the operation in-place. Default: False.

[0108] After several convolutional and max pooling layers, the high-level reasoning in the neural network is done via fully connected layers. Neurons in a fully connected layer have connections to all activations in the previous layer, as seen in regular (non-convolutional) artificial neural networks. Their activations can thus be computed as an affine transformation, with matrix multiplication followed by a bias offset (vector addition of a learned or fixed bias term).

[0109] The "loss layer" (including calculating of a loss function) specifies how training penalizes the deviation between the predicted (output) and true labels and is normally the final layer of a neural network. Various loss functions appropriate for different tasks may be used. Softmax loss is used for predicting a single class of K mutually exclusive classes. Sigmoid cross-entropy loss is used for predicting K independent probability values in [0, 1], Euclidean loss is used for regressing to real- valued labels. In summary, Fig. 1 shows the data flow in a typical convolutional neural network. First, the input image is passed through convolutional layers and becomes abstracted to a feature map comprising several channels, corresponding to a number of filters in a set of learnable filters of this layer. Then, the feature map is subsampled using e.g. a pooling layer, which reduces the dimension of each channel in the feature map. Next, the data comes to another convolutional layer, which may have different numbers of output channels. As was mentioned above, the number of input channels and output channels are hyper-parameters of the layer. To establish connectivity of the network, those parameters need to be synchronized between two connected layers, such that the number of input channels for the current layers should be equal to the number of output channels of the previous layer. For the first layer which processes input data, e.g. an image, the number of input channels is normally equal to the number of channels of data representation, for instance 3 channels for RGB or YUV representation of images or video, or 1 channel for grayscale image or video representation. The channels obtained by one or more convolutional layers (and possibly resampling layer(s)) may be passed to an output layer. Such output layer may be a convolutional or resampling in some implementations. In an exemplary and non-limiting implementation, the output layer is a fully connected layer.

[0110] Autoencoders and unsupervised learning

[0111] An autoencoder is a type of artificial neural network used to learn efficient data codings in an unsupervised manner. A schematic drawing thereof is shown in Fig. 2. The autoencoder includes an encoder side 210 with an input x inputted into an input layer of an encoder subnetwork 220 and a decoder side 250 with output x’ outputted from a decoder subnetwork 260. The aim of an autoencoder is to learn a representation (encoding) 230 for a set of data x, typically for dimensionality reduction, by training the network 220, 260 to ignore signal “noise”. Along with the reduction (encoder) side subnetwork 220, a reconstructing (decoder) side subnetwork 260 is learnt, where the autoencoder tries to generate from the reduced encoding 230 a representation x’ as close as possible to its original input x, hence its name. In the simplest case, given one hidden layer, the encoder stage of an autoencoder takes the input x and maps it to h: h = ct(Wx + b).

[0112] This image h is usually referred to as code 230, latent variables, or latent representation. Here, ct is an element-wise activation function such as a sigmoid function or a rectified linear unit. W is a weight matrix b is a bias vector. Weights and biases are usually initialized randomly, and then updated iteratively during training through Backpropagation. After that, the decoder stage of the autoencoder maps h to the reconstruction x'of the same shape as x: x' = ct'(W'h' + b') where a’ , W' and b' for the decoder may be unrelated to the corresponding a, W and b for the encoder.

[0113] Variational autoencoder models make strong assumptions concerning the distribution of latent variables. They use a variational approach for latent representation learning, which results in an additional loss component and a specific estimator for the training algorithm called the Stochastic Gradient Variational Bayes (SGVB) estimator. It assumes that the data is generated by a directed graphical model pe(x|h) and that the encoder is learning an approximation q,|,(h|x) to the posterior distribution pe(h|x) where <f> and 0 denote the parameters of the encoder (recognition model) and decoder (generative model) respectively.

[0114] The probability distribution of the latent vector of a VAE typically matches that of the training data much closer than a standard autoencoder. The objective of VAE has the following form: Here, DKLstands for the Kullback-Leibler divergence. The prior over the latent variables is usually set to be the centered isotropic multivariate Gaussian pe(h) = N (0, / ) . Commonly, the shape of the variational and the likelihood distributions are chosen such that they are factorized Gaussians: where p(x) and a>2(x) are the encoder output, while p( / i) and cr2(h) are the decoder outputs.

[0115] Recent progress in artificial neural networks area and especially in convolutional neural networks enables researchers’ interest of applying neural networks based technologies to the task of image and video compression. For example, End- to-end Optimized Image Compression has been proposed, which uses a network based on a variational autoencoder.

[0116] Accordingly, data compression is considered as a fundamental and well-studied problem in engineering, and is commonly formulated with the goal of designing codes for a given discrete data ensemble with minimal entropy. The solution relies heavily on knowledge of the probabilistic structure of the data, and thus the problem is closely related to probabilistic source modeling. However, since all practical codes must have finite entropy, continuous- valued data (such as vectors of image pixel intensities) must be quantized to a finite set of discrete values, which introduces an error.

[0117] In this context, known as the lossy compression problem, one must trade off two competing costs: the entropy of the discretized representation (rate) and the error arising from the quantization (distortion). Different compression applications, such as data storage or transmission over limited-capacity channels, demand different rate-distortion tradeoffs.

[0118] Joint optimization of rate and distortion is difficult. Without further constraints, the general problem of optimal quantization in high-dimensional spaces is intractable. For this reason, most existing image compression methods operate by linearly transforming the data vector into a suitable continuous-valued representation, quantizing its elements independently, and then encoding the resulting discrete representation using a lossless entropy code. This scheme is called transform coding due to the central role of the transformation.

[0119] For example, JPEG uses a discrete cosine transform on blocks of pixels, and JPEG 2000 uses a multi-scale orthogonal wavelet decomposition. Typically, the three components of transform coding methods - transform, quantizer, and entropy code - are separately optimized (often through manual parameter adjustment). Modem video compression standards like HEVC, VVC and EVC also use transformed representation to code residual signal after prediction. The several transforms are used for that purpose such as discrete cosine and sine transforms (DCT, DST), as well as low frequency non-separable manually optimized transforms (LFNST).

[0120] Variational image compression

[0121] Variable Auto-Encoder (VAE) framework can be considered as a nonlinear transforming coding model. The transforming process can be mainly divided into four parts. This is exemplified in Fig. 3A showing a VAE framework. Q representing the quantizer function. The entropy model, or the hyper encoder / decoder (also known as hyperprior) 103 estimates the distribution of the quantized latent representation y to get the minimum rate achievable with a lossless entropy source coding.

[0122] The latent space can be understood as a representation of compressed data in which similar data points are closer together in the latent space. Latent space is useful for learning data features and for finding simpler representations of data for analysis. The quantized latent representation T, y and the side information z of the hyperprior 3 are included into a bitstream 2 (are binarized) using arithmetic coding (AE). Furthermore, a decoder 104 is provided that transforms the quantized latent representation to the reconstructed image x, x = g(y)- The signal x is the estimation of the input image x. It is desirable that x is as close to x as possible, in other words the reconstruction quality is as high as possible. However, the higher the similarity between x and x, the higher the amount of side information necessary to be transmitted. The side information includes bitstreaml and bitstream2 shown in Fig. 3A, which are generated by the encoder and transmitted to the decoder. Normally, the higher the amount of side information, the higher the reconstruction quality. However, a high amount of side information means that the compression ratio is low. Therefore, one purpose of the system described in Fig. 3 A is to balance the reconstruction quality and the amount of side information conveyed in the bitstream.

[0123] In Fig. 3A the component AE 105 is the Arithmetic Encoding module, which converts samples of the quantized latent representation y and the side information z into a binary representation bitstream 1. The samples of y and z might for example comprise integer or floating point numbers. One purpose of the arithmetic encoding module is to convert (via the process of binarization) the sample values into a string of binary digits (which is then included in the bitstream that may comprise further portions corresponding to the encoded image or further side information).

[0124] The arithmetic decoding (AD) 106 is the process of reverting the binarization process, where binary digits are converted back to sample values. The arithmetic decoding is provided by the arithmetic decoding module 106.

[0125] It is noted that the present disclosure is not limited to this particular framework. Moreover, the present disclosure is not restricted to image or video compression, and can be applied to object detection, image generation, and recognition systems as well.

[0126] In Fig. 3 A there are two sub networks concatenated to each other. A subnetwork in this context is a logical division between the parts of the total network. For example, in Fig. 3A the modules 101, 102, 104, 105 and 106 are called the “Encoder / Decoder” subnetwork. The “Encoder / Decoder” subnetwork is responsible for encoding (generating) and decoding (parsing) of the first bitstream “bitstreaml”. The second network in Fig. 3A comprises modules 103, 108, 109, 110 and 107 and is called “hyper encoder / decoder” subnetwork. The second subnetwork is responsible for generating the second bitstream “bitstream2”. The purposes of the two subnetworks are different.

[0127] • parsing the bitstream 1 via AD using the arithmetic decoding module 106, and

[0128] • reconstructing 104 the reconstructed image (x) using the parsed data. The purpose of the second subnetwork is to obtain statistical properties (e.g. mean value, variance and correlations between samples of bitstream 1) of the samples of “bitstream 1”, such that the compressing of bitstream 1 by first subnetwork is more efficient. The second subnetwork generates a second bitstream “bitstream2”, which comprises the said information (e.g. mean value, variance and correlations between samples of bitstream 1).

[0129] The second network includes an encoding part which comprises transforming 103 of the quantized latent representation y into side information z, quantizing the side information z into quantized side information z, and encoding (e.g. binarizing) 109 the quantized side information z into bitstream2. In this example, the binarization is performed by an arithmetic encoding (AE). A decoding part of the second network includes arithmetic decoding (AD) 110, which transforms the input bitstream2 into decoded quantized side information z'. The z' might be identical to z, since the arithmetic encoding end decoding operations are lossless compression methods. The decoded quantized side information z' is then transformed 107 into decoded side information y'. y' represents the statistical properties of y (e.g. mean value of samples of y, or the variance of sample values or like). The decoded latent representation y' is then provided to the above-mentioned Arithmetic Encoder 105 and Arithmetic Decoder 106 to control the probability model ofy.

[0130] The Fig. 3A describes an example of VAE (variational auto encoder), details of which might be different in different implementations. For example in a specific implementation additional components might be present to more efficiently obtain the statistical properties of the samples of bitstream 1. In one such implementation a context modeler might be present, which targets extracting cross-correlation information of the bitstream 1. The statistical information provided by the second subnetwork might be used by AE (arithmetic encoder) 105 and AD (arithmetic decoder) 106 components.

[0131] Fig. 3A depicts the encoder and decoder in a single figure. As is clear to those skilled in the art, the encoder and the decoder may be, and very often are, embedded in mutually different devices.

[0132] Fig. 3B depicts the encoder and Fig. 3C depicts the decoder components of the VAE framework in isolation. As input, the encoder receives, according to some embodiments, a picture. The input picture may include one or more channels, such as color channels or other kind of channels, e.g. depth channel or motion information channel, or the like. The output of the encoder (as shown in Fig. 3B) is a bitstreaml and a bitstream2. The bitstreaml is the output of the first sub-network of the encoder and the bitstream2 is the output of the second subnetwork of the encoder.

[0133] Similarly, in Fig. 3C, the two bitstreams, bitstreaml and bitstream2, are received as input and z, which is the reconstructed (decoded) image, is generated at the output. As indicated above, the VAE can be split into different logical units that perform different actions. This is exemplified in Figs. 3B and 3C so that Fig. 3B depicts components that participate in the encoding of a signal, like a video and provided encoded information. This encoded information is then received by the decoder components depicted in Fig. 3C for encoding, for example. It is noted that the components of the encoder and decoder denoted with numerals 12x and 14x may correspond in their function to the components referred to above in Fig. 3A and denoted with numerals lOx.

[0134] Specifically, as is seen in Fig. 3B, the encoder comprises the encoder 121 that transforms an input x into a signal y which is then provided to the quantizer 322. The quantizer 122 provides information to the arithmetic encoding module 125 and the hyper encoder 123. The hyper encoder 123 provides the bitstream2 already discussed above to the hyper decoder 147 that in turn provides the information to the arithmetic encoding module 105 (125).

[0135] The output of the arithmetic encoding module is the bitstreaml . The bitstreaml and bitstream2 are the output of the encoding of the signal, which are then provided (transmitted) to the decoding process. Although the unit 101 (121) is called “encoder”, it is also possible to call the complete subnetwork described in Fig. 3B as “encoder”. The process of encoding in general means the unit (module) that converts an input to an encoded (e.g. compressed) output. It can be seen from Fig. 3B, that the unit 121 can be actually considered as a core of the whole subnetwork, since it performs the conversion of the input x into y, which is the compressed version of the x. The compression in the encoder 121 may be achieved, e.g. by applying a neural network, or in general any processing network with one or more layers. In such network, the compression may be performed by cascaded processing including downsampling which reduces size and / or number of channels of the input. Thus, the encoder may be referred to, e.g. as a neural network (NN) based encoder, or the like.

[0136] The remaining parts in the figure (quantization unit, hyper encoder, hyper decoder, arithmetic encoder / decoder) are all parts that either improve the efficiency of the encoding process or are responsible for converting the compressed output y into a series of bits (bitstream). Quantization may be provided to further compress the output of the NN encoder 121 by a lossy compression. The AE 125 in combination with the hyper encoder 123 and hyper decoder 127 used to configure the AE 125 may perform the binarization which may further compress the quantized signal by a lossless compression. Therefore, it is also possible to call the whole subnetwork in Fig. 3B an “encoder”.

[0137] A majority of Deep Learning (DL) based image / video compression systems reduce dimensionality of the signal before converting the signal into binary digits (bits). In the VAE framework for example, the encoder, which is a non-linear transform, maps the input image x into y, where y has a smaller width and height than x. Since the y has a smaller width and height, hence a smaller size, the (size of the) dimension of the signal is reduced, and, hence, it is easier to compress the signal y. It is noted that in general, the encoder does not necessarily need to reduce the size in both (or in general all) dimensions. Rather, some exemplary implementations may provide an encoder which reduces size only in one (or in general a subset of) dimension.

[0138] In J. Balle, L. Valero Laparra, and E. P. Simoncelli (2015). “Density Modeling of Images Using a Generalized Normalization Transformation”, In: arXiv e-prints, Presented at the 4th Int. Conf, for Learning Representations, 2016 (referred to in the following as “Balle”) the authors proposed a framework for end-to-end optimization of an image compression model based on nonlinear transforms. The authors optimize for Mean Squared Error (MSE), but use a more flexible transforms built from cascades of linear convolutions and nonlinearities. Specifically, authors use a generalized divisive normalization (GDN) joint nonlinearity that is inspired by models of neurons in biological visual systems, and has proven effective in Gaussianizing image densities. This cascaded transformation is followed by uniform scalar quantization (i.e., each element is rounded to the nearest integer), which effectively implements a parametric form of vector quantization on the original image space. The compressed image is reconstructed from these quantized values using an approximate parametric nonlinear inverse transform.

[0139] Such example of the VAE framework is shown in Fig. 4, and it utilizes 6 downsampling layers that are marked with 401 to 406. The network architecture includes a hyperprior model. The left side (ga, gs) shows an image autoencoder architecture, the right side (ha, hs) corresponds to the autoencoder implementing the hyperprior. The factorized-prior model uses the identical architecture for the analysis and synthesis transforms gaand gs. Q represents quantization, and AE, AD represent arithmetic encoder and arithmetic decoder, respectively. The encoder subjects the input image x to ga, yielding the responses y (latent representation) with spatially varying standard deviations. The encoding gaincludes a plurality of convolution layers with subsampling and, as an activation function, generalized divisive normalization (GDN). The responses are fed into ha, summarizing the distribution of standard deviations in z. z is then quantized, compressed, and transmitted as side information. The encoder then uses the quantized vector z to estimate a, the spatial distribution of standard deviations which is used for obtaining probability values (or frequency values) for arithmetic coding (AE), and uses it to compress and transmit the quantized image representation y (or latent representation). The decoder first recovers z from the compressed signal. It then uses hsto obtain y, which provides it with the correct probability estimates to successfully recover y as well. It then feeds y into gsto obtain the reconstructed image.

[0140] The layers that include downsampling is indicated with the downward arrow in the layer description. The layer description „Conv N,kl ,2J“ means that the layer is a convolution layer, with N channels and the convolution kernel is klxkl in size. For example, kl may be equal to 5 and k2 may be equal to 3. As stated, the 2 (means that a downsampling with a factor of 2 is performed in this layer. Downsampling by a factor of 2 results in one of the dimensions of the input signal being reduced by half at the output. In Fig. 4, the 2 (indicates that both width and height of the input image is reduced by a factor of 2. Since there are 6 downsampling layers, if the width and height of the input image 414 (also denoted with x) is given by w and h, the output signal z ' 413 is has width and height equal to w / 64 and h / 64 respectively. Modules denoted by AE and AD are arithmetic encoder and arithmetic decoder, which are explained with reference to Figs. 3A to 3C. The arithmetic encoder and decoder are specific implementations of entropy coding. AE and AD can be replaced by other means of entropy coding. In information theory, an entropy encoding is a lossless data compression scheme that is used to convert the values of a symbol into a binary representation which is a revertible process. Also, the “Q” in the figure corresponds to the quantization operation that was also referred to above in relation to Fig. 4 and is further explained above in the section “Quantization”. Also, the quantization operation and a corresponding quantization unit as part of the component 413 or 415 is not necessarily present and / or can be replaced with another unit.

[0141] In Fig. 4, there is also shown the decoder comprising upsampling layers 407 to 412. A further layer 420 is provided between the upsampling layers 411 and 410 in the processing order of an input that is implemented as convolutional layer but does not provide an upsampling to the input received. A corresponding convolutional layer 430 is also shown for the decoder. Such layers can be provided in NNs for performing operations on the input that do not alter the size of the input but change specific characteristics. However, it is not necessary that such a layer is provided.

[0142] When seen in the processing order of bitstream2 through the decoder, the upsampling layers are run through in reverse order, i.e. from upsampling layer 412 to upsampling layer 407. Each upsampling layer is shown here to provide an upsampling with an upsampling ratio of 2, which is indicated by the f. It is, of course, not necessarily the case that all upsampling layers have the same upsampling ratio and also other upsampling ratios like 3, 4, 8 or the like may be used. The layers 407 to 412 are implemented as convolutional layers (conv). Specifically, as they may be intended to provide an operation on the input that is reverse to that of the encoder, the upsampling layers may apply a deconvolution operation to the input received so that its size is increased by a factor corresponding to the upsampling ratio. However, the present disclosure is not generally limited to deconvolution and the upsampling may be performed in any other manner such as by bilinear interpolation between two neighboring samples, or by nearest neighbor sample copying, or the like.

[0143] In the first subnetwork, some convolutional layers (401 to 403) are followed by generalized divisive normalization (GDN) at the encoder side and by the inverse GDN (I GDN) at the decoder side. In the second subnetwork, the activation function applied is ReLu. It is noted that the present disclosure is not limited to such implementation and in general, other activation functions may be used instead of GDN or ReLu. Cloud solutions for machine tasks

[0144] The Video Coding for Machines (VCM) is another computer science direction being popular nowadays. The main idea behind this approach is to transmit a coded representation of image or video information targeted to further processing by computer vision (CV) algorithms, like object segmentation, detection and recognition. In contrast to traditional image and video coding targeted to human perception the quality characteristic is the performance of computer vision task, e.g. object detection accuracy, rather than reconstructed quality. This is illustrated in Fig. 5.

[0145] Video Coding for Machines is also referred to as collaborative intelligence and it is a relatively new paradigm for efficient deployment of deep neural networks across the mobile-cloud infrastructure. By dividing the network between the mobile side 510 and the cloud side 590 (e.g. a cloud server), it is possible to distribute the computational workload such that the overall energy and / or latency of the system is minimized. In general, the collaborative intelligence is a paradigm where processing of a neural network is distributed between two or more different computation nodes; for example devices, but in general, any functionally defined nodes. Here, the term “node” does not refer to the above-mentioned neural network nodes. Rather the (computation) nodes here refer to (physically or at least logically) separate devices / modules, which implement parts of the neural network. Such devices may be different servers, different end user devices, a mixture of servers and / or user devices and / or cloud and / or processor or the like. In other words, the computation nodes may be considered as nodes belonging to the same neural network and communicating with each other to convey coded data within / for the neural network. For example, in order to be able to perform complex computations, one or more layers may be executed on a first device (such as a device on mobile side 510) and one or more layers may be executed in another device (such as a cloud server on cloud side 590). However, the distribution may also be finer and a single layer may be executed on a plurality of devices. In this disclosure, the term “plurality” refers to two or more. In some existing solution, a part of a neural network functionality is executed in a device (user device or edge device or the like) or a plurality of such devices and then the output (feature map) is passed to a cloud. A cloud is a collection of processing or computing systems that are located outside the device, which is operating the part of the neural network. The notion of collaborative intelligence has been extended to model training as well. In this case, data flows both ways: from the cloud to the mobile during back-propagation in training, and from the mobile to the cloud (illustrated in Fig. 5) during forward passes in training, as well as inference.

[0146] Some works presented semantic image compression by encoding deep features and then reconstructing the input image from them. The compression based on uniform quantization was shown, followed by context-based adaptive arithmetic coding (CABAC) from H.264. In some scenarios, it may be more efficient, to transmit from the mobile part 510 to the cloud 590 an output of a hidden layer (a deep feature map) 550, rather than sending compressed natural image data to the cloud and perform the object detection using reconstructed images. It may thus be advantageous to compress the data (features) generated by the mobile side 510, which may include a quantization layer 520 for this purpose. Correspondingly, the cloud side 590 may include an inverse quantization layer 560. The efficient compression of feature maps benefits the image and video compression and reconstruction both for human perception and for machine vision. Entropy coding methods, e.g. arithmetic coding is a popular approach to compression of deep features (i.e. feature maps).

[0147] Nowadays, video content contributes to more than 80% internet traffic, and the percentage is expected to increase even further. Therefore, it is critical to build an efficient video compression system and generate higher quality frames at given bandwidth budget. In addition, most video related computer vision tasks such as video object detection or video object tracking are sensitive to the quality of compressed videos, and efficient video compression may bring benefits for other computer vision tasks. Meanwhile, the techniques in video compression are also helpful for action recognition and model compression. However, in the past decades, video compression algorithms rely on hand-crafted modules, e.g., block based motion estimation and Discrete Cosine Transform (DCT), to reduce the redundancies in the video sequences, as mentioned above. Although each module is well designed, the whole compression system is not end-to-end optimized. It is desirable to further improve video compression performance by jointly optimizing the whole compression system.

[0148] End-to-end image or video compression

[0149] DNN based image compression methods can exploit large scale end-to-end training and highly non-linear transform, which are not used in the traditional approaches. However, it is non-trivial to directly apply these techniques to build an end-to-end learning system for video compression. First, it remains an open problem to learn how to generate and compress the motion information tailored for video compression. Video compression methods heavily rely on motion information to reduce temporal redundancy in video sequences.

[0150] A straightforward solution is to use the learning based optical flow to represent motion information. However, current learning based optical flow approaches aim at generating flow fields as accurate as possible. The precise optical flow is often not optimal for a particular video task. In addition, the data volume of optical flow increases significantly when compared with motion information in the traditional compression systems and directly applying the existing compression approaches to compress optical flow values will significantly increase the number of bits required for storing motion information. Second, it is unclear how to build a DNN based video compression system by minimizing the rate-distortion based objective for both residual and motion information. Rate-distortion optimization (RDO) aims at achieving higher quality of reconstructed frame (i.e., less distortion) when the number of bits (or bit rate) for compression is given. RDO is important for video compression performance. In order to exploit the power of end-to-end training for learning based compression system, the RDO strategy is required to optimize the whole system.

[0151] In Guo Lu, Wanli Ouyang, Dong Xu, Xiaoyun Zhang, Chunlei Cai, Zhiyong Gao; „DVC: An End-to-end Deep Video Compression Framework". Proceedings of the IEEE / CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 11006-11015, authors proposed the end-to-end deep video compression (DVC) model that jointly learns motion estimation, motion compression, and residual coding.

[0152] Such encoder is illustrated in Figure 6A. In particular, Figure 6A shows an overall structure of end-to-end trainable video compression framework. In order to compress motion information, a CNN was designated to transform the optical flow vLto the corresponding representations mtsuitable for better compression. Specifically, an auto-encoder style network is used to compress the optical flow. The motion vectors (MV) compression network is shown in Figure 6B. The network architecture is somewhat similar to the ga / gs of Figure 4. In particular, the optical flow vLis fed into a series of convolution operation and nonlinear transform including GDN and IGDN. The number of output channels c for convolution (deconvolution) is here exemplarily 128 except for the last deconvolution layer, which is equal to 2 in this example. The kernel size is k, e.g. k=3. Given optical flow with the size of M x N x 2, the MV encoder will generate the motion representation mtwith the size of M / 16xN / 16x 128. Then motion representation is quantized (Q), entropy coded and sent to bitstream as mt. The MV decoder receives the quantized representation mtand reconstruct motion information vLusing MV encoder. In general, the values for k and c may differ from the above mentioned examples as is known from the art.

[0153] Figure 6C shows a structure of the motion compensation part. Here, using previous reconstructed frame xt-i and reconstructed motion information, the warping unit generates the warped frame (normally, with help of interpolation filter such as bi-linear interpolation filter). Then a separate CNN with three inputs generates the predicted picture. The architecture of the motion compensation CNN is also shown in Figure 6C. The residual information between the original frame and the predicted frame is encoded by the residual encoder network. A highly non-linear neural network is used to transform the residuals to the corresponding latent representation. Compared with discrete cosine transform in the traditional video compression system, this approach can better exploit the power of non-linear transform and achieve higher compression efficiency.

[0154] From above overview it can be seen that CNN based architecture can be applied both for image and video compression, considering different parts of video framework including motion estimation, motion compensation and residual coding. Entropy coding is popular method used for data compression, which is widely adopted by the industry and is also applicable for feature map compression either for human perception or for computer vision tasks.

[0155] Video Coding for Machines

[0156] The Video Coding for Machines (VCM) is another computer science direction being popular nowadays. The main idea behind this approach is to transmit the coded representation of image or video information targeted to further processing by computer vision (CV) algorithms, like object segmentation, detection and recognition. In contrast to traditional image and video coding targeted to human perception the quality characteristic is the performance of computer vision task, e.g. object detection accuracy, rather than reconstructed quality.

[0157] A recent study proposed a new deployment paradigm called collaborative intelligence, whereby a deep model is split between the mobile and the cloud. Extensive experiments under various hardware configurations and wireless connectivity modes revealed that the optimal operating point in terms of energy consumption and / or computational latency involves splitting the model, usually at a point deep in the network. Today’s common solutions, where the model sits fully in the cloud or fully at the mobile, were found to be rarely (if ever) optimal. The notion of collaborative intelligence has been extended to model training as well. In this case, data flows both ways: from the cloud to the mobile during back-propagation in training, and from the mobile to the cloud during forward passes in training, as well as inference.

[0158] Lossy compression of deep feature data has been studied based on HEVC intra coding, in the context of a recent deep model for object detection. It was noted the degradation of detection performance with increased compression levels and proposed compression-augmented training to minimize this loss by producing a model that is more robust to quantization noise in feature values. However, this is still a sub-optimal solution, because the codec employed is highly complex and optimized for natural scene compression rather than deep feature compression.

[0159] The problem of deep feature compression for the collaborative intelligence has been addressed by an approach for object detection task using popular YOLOv2 network for the study of compression efficiency and recognition accuracy tradeoff. Here the term deep feature has the same meaning as feature map. The word ‘deep’ comes from the collaborative intelligence idea when the output feature map of some hidden (deep) layer is captured and transferred to the cloud to perform inference. That appears to be more efficient rather than sending compressed natural image data to the cloud and perform the object detection using reconstructed images.

[0160] The efficient compression of feature maps benefits the image and video compression and reconstruction both for human perception and for machine vision. Said about disadvantages of state-of-the art autoencoder based approach to compression are also valid for machine vision tasks. Functional modules

[0161] Variable bitrate module

[0162] An encoder can output bitstreams at different bit rates. Therefore, in some methods, an output of an encoding network is scaled (for example, each channel is multiplied by a corresponding scaling factor that is also referred to as a target gain value), and an input of a decoding network is inversely scaled (for example, each channel is multiplied by a corresponding scaling factor reciprocal that is also referred to as a target inverse gain value), as shown in FIG. 7. The scaling factor may be preset. Different quality levels or quantization parameters correspond to different target gain values. If the output of the encoding network is scaled to a smaller value, a bitstream size may be decreased. Otherwise, the bitstream size may be increased.

[0163] Color format transform

[0164] RGB and YUV are common color spaces. Conversion between RGB and YUV may be performed according to an equation specified in standards such as CCIR 601 and BT.709.

[0165] Separate structure for luma and chroma

[0166] Some VAE-based codecs use the YUV color space as an input of an encoder and an output of a decoder, as shown in FIG. 8. A Y component indicates luma, and a UV component indicates chroma. Resolution of the UV component may be the same as or lower than that of the Y component. Typical formats include YUV4:4:4, YUV4:2:2, and YUV4:2:0. The Y component is converted into a feature map F_Y through a network, and an entropy encoding module generates a bitstream of the Y component based on the feature map F_Y. The UV component is converted into a feature map F_UV through another network, and the entropy encoding module generates a bitstream of the UV component based on the feature map F_UV Under this structure, the feature map of the Y component and the feature map of the UV component may be independently quantized, so that bits are flexibly allocated for luma and chroma. For example, for a color-sensitive image, a feature map of a UV component may be less quantized, and a quantity of bitstream bits for a UV component may be increased, to improve reconstruction quality of the UV component and achieve better visual effect.

[0167] In some other methods, an encoder concatenates (concatenate) a Y component and a UV component and then sends to a UV component processing module (for converting image information into a feature map). In addition, a decoder concatenates a reconstructed feature map of the Y component and a reconstructed feature map of the UV component and then sends to a UV component processing module 2 (for converting a feature map into image information). In this method, a correlation between the Y component and the UV component may be used to reduce a bitstream of the UV component.

[0168] In the present specification, in the context of a neural network, a ‘parameter’ may be a value used in an operation process of each layer forming a neural network, and for example, may include a weight used when an input value is applied to a certain operation expression. Here, the parameter may be expressed in a matrix form. The parameter is a value set as a result of training, and may be updated through separate training data when necessary.

[0169] In the present specification, in the context of a bitstream, a ‘parameter’ may be a sequence of bits used to encode some property, flag, or value that defines or identifies an aspect or property of the bitstream.

[0170] Bitstream structure

[0171] A bitstream generally comprises, or consists of, one or more substreams. For example, a bitstream may comprise a plurality of substreams. Each substream contains a payload having a particular data type. In order to signal to a decoder what the data type of a substream is, each substream comprises a marker ID. Preferably, the marker ID is formed of a sequence of bits that are leading bits in the substream. Fig. 9 shows an exemplary codestream layout of a bitstream for image or video coding.

[0172] The bitstream 900 is composed of start and end of stream markers, and multiple marker segments, which are parts of a codestream started with a marker. Each maker segment my have a byte aligned boundary. The marker and marker segments shown are:

[0173] - SOC - start of codestream marker, 901 ;

[0174] - PIH - start of picture header marker segment followed by picture header substream, 902:

[0175] - SOZ - start of z-stream marker segment followed by substream of hyper tensor z, including z[0] ("stream zY") and z[l] ("stream zuv"), 903;

[0176] - SORp - start of residual stream for primary component marker segment followed by substream of primary component residual, which includes r[0] data ("stream rY"), 904;

[0177] - SORs - start of residual stream for secondary component marker segment followed by substream of secondary component residual, which includes r[l] data ("stream ruv), 905;

[0178] - TOH - start of tools header marker segment followed by tools information substream, 906;

[0179] - RDI - start of rendering information marker segment followed by rendering information substream, 907;

[0180] - UDI - start of used defined information marker segment followed by substream, 908;

[0181] - SOQ - start of quality map marker segment followed by substream, 909;

[0182] - EOC - end of codestream marker, 910.

[0183] As mentioned above, a bitstream may comprise multiple substreams. Each marker segment can correspond to a substream of the bitstream.

[0184] In Fig. 9, the marker segments SORp and SORs are residual substreams for the primary and secondary components of the image coding format. The primary and second components may be luma and chroma respectively

[0185] As described above, the picture header substream has a marker ID ‘PIH’. An example of a picture header substream 1000 of a bitstream is schematically illustrated in Fig. 10. The picture image header 1001 starts with marker ID (marker_id) and the substream comprises entropy encoded data 1002 in the pay load.

[0186] The picture header substream may contain information about image height H. width W, latent domain tiles location and sizes, control flags for each tool, scaling factors for primary and secondary components (e.g. luma and chroma), learnable model index and displacement for rate control parameters ( pYfor the primary component (luma) and puvfor the secondary component (chroma)).

[0187] Figures I la and 11b respectively schematically illustrate substreams 1100 and 1150 for the primary and secondary residual substreams respectively. The primary residual substream comprises residual information for a primary component of the image coding format (for example, luma) and the secondary residual substream comprises residual information for a secondary component of the image coding format (for example, chroma).

[0188] The primary residual header 1011 starts with the marker ID (marker_id) SORp and the substream comprises entropy encoded data 1102 in the pay load. The secondary residual header 1151 starts with the marker ID (marker_id) SORs and the substream comprises entropy encoded data 1152 in the pay load. The primary and secondary residual substreams each comprise one or more channels, preferably multiple channels, to be decoded by the decoder.

[0189] Tensor sizes

[0190] Table 1 - Tensor sizes

[0191] Skip mode

[0192] Skip mode is a tool that can be used in the codec to tell the entropy coder which residual elements should be decoded from the bitstream and which should not. Some functions relevant to skip mode will now be described.

[0193] To generate the mask, four syntax elements are used, namely Threshold, GreaterFlag, Mode and BlockSize. These syntax elements can be included in the picture header. Mask generation is generally a three-step process.

[0194] A first step is pooling. If the BlockSize is greater than 1, a pooling operation is applied to the input sigma samples tensor first. Based on the value of the Mode syntax element, the pooling operation is either max pooling, average pooling or min pooling, with a kernel size equal to BlockSize in horizontal and vertical dimensions.

[0195] After the pooled mask samples tensor, maskp, is obtained, and if the BlockSize is greater than 1, an up- sampling operation is applied to maskp to obtain the final mask samples tensor. The up-sampling operation can be based on nearest neighbors. If the BlockSize is equal to 1 , the up-sampling operation is skipped. If the BlockSize is greater than 1 , a cropping operation is applied after up- sampling resulting in an output mask tensor with size [C, h4,w4]

[0196] Parameter for signaling the number of channels in a substream

[0197] Exemplary methods and devices according to particular embodiments of the present disclosure will now be described in further detail. Embodiments of the present disclosure may be used for an image coding scenario. Generally, embodiments allow backwards compatibility of bitstreams encoded using subsequent standards with existing decoders and legacy decoders. The codec has an internal image coding format. The input image converts to the internal coded format and is divided into the primary (luma) and the secondary (chroma) components of the image coding format. Each component of the coded image converts to the latent representation (latent tensor). The latent tensor is separated into prediction and residual parts. The residual part (r) is encoded into the bitstream. The residual part can be encoded into the bitstream for the primary and secondary components in a primary residual bitstream and a secondary residual bitstream respectively, as described above.

[0198] As mentioned above, the picture header substream (with marker ID ‘PIH’) contains information about image height H, width W, latent domain tiles location and sizes, control flags for tools, scaling factors for primary and secondary components, learnable model index and displacement for rate control parameters (pYfor primary and puvfor secondary component).

[0199] The present disclosure introduces a parameter in a first substream of the bitsteam, which can conveniently be the picture header substream. The parameter in the first substream of the multiple substreams indicates a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream.

[0200] The decoder parses the first substream. The decoder parses syntax elements of the parameter to determine the number of channels in the second substream for the component of the image coding format of the bitstream. In dependence on the determined number of channels, the decoder parses the second substream. The decoder may parse only the determined number of channels in the second bitstream. The channels that are not to be parsed may be removed. This may reduce the final size of the bitstream. This may allow to reduce the size of the bitstream without reconstruction for transcoding.

[0201] The first substream and the second substream may not be consecutively parsed substreams of the substream. Preferably, the first substream is the first substream that is parsed in the parsing of the bitstream, but it need not be. The second substream may be parsed after the first substream, but not necessarily consecutively . Generally, the terms first and second are used as labels to differentiate two different substreams. There may be one or more other substreams that are parsed between the first and the second substreams.

[0202] In the implementations described herein, the second substream is a residual substream for the component of the image coding format of the bitstream. The signaled number of channels are channels of the residual (r) for reconstruction of the image.

[0203] In a possible implementation, the parameter is specified in a field num_chs[comp] in the first substream, where comp is the component of the image coding format of the bitstream. The component of the image coding format may be indicated by a component value (comp). For example, if the component value is 0, the parameter relates to a number of residual channels in a primary residual substream. If the component value is 1 , the parameter relates to a number of residual channels in a secondary residual substream.

[0204] Fig. 12 shows exemplary changes to the picture header syntax (showing a reduced version of the picture image header) to include the number of decoded channels for the component of the image coding format of the bitstream, as shown at 1201, 1202 in the “model information for secondary compoemmnts” and “RVS information” seconds of the picture image header respectively. grfs_channel_flag[comp] is an array with 1 -bit flags. Its size depends on value of the comp. For comp equal to 0 (primary component) the size is Cpand for comp equal to 1 (secondary component) the size is Cs. For channels with grfs_channel_flag[comp] equal to 0, the absolute value of the comp component residual tensor is reduced, for channels with grfs_channel_flag[comp] equal to 1 the absolute value of comp component residual tensor is amplified.

[0205] Subsequent to determining the number of channels in the second substream for the component of an image coding format of the bitstream, the parsing process for the array grfs_channel_flag[comp] can be configured by limiting the number of parsed elements to num_chs[comp] . All other elements of the array grfs_channel_flag[comp] can be zeroed.

[0206] Fig. 13 schematically illustrates the parameter 1301 being encoded in the first substream 1302.

[0207] As illustrated in the picture header syntax table shown in Fig. 12, the parameter num_chs is parsed according to the u(n) descriptor which specifies the parsing process of each sytax element of the parameter. The descriptor u(n) indicates an unsigned integer using n bits. When n is "v" in the syntax table, the number of bits varies in a manner dependent on the value of other syntax elements. The parsing process for this descriptor is specified by the return value of the function read_bits(n) interpreted as a binary representation of an unsigned integer with most significant bit written first.

[0208] The parameter may be an integer value. As mentioned above, the parameter comprises syntax elements that are parsed by the decoder. In a possible implementation, the parameter is signaled as 8 bits for a primary component of the image coding format and 7 bits for a secondary component of the image coding format of the bitstream. In another implementation, the parameter is signaled as 8 bits for each of a primary component of the image coding format and a secondary component of the image coding forma of the bitstream.

[0209] As mentioned above, each of the multiple substreams comprises an Marker ID indicating a data type of the respective substream. The marker ID for the second substream may be SORc, where c indicates the component of the image coding format of the bitstream. For example, where the parameter relates to the primary component of the image coding format (for example, luma), the Marker ID may be SOPp and where the parameter relates to the second component of the image coding format (for example, chroma), the Marker ID may be SOPp.

[0210] Suitably, the indicated number of channels is less than Cpor Cs(as indicated in Table 1 ), the channels sizes of the primary and secondary latent tensor respectively, for the primary and secondary components respectively.

[0211] In addition to the first and second substreams, the bitstream may comprise a third substream, and optionally one or more additional substreams. In one example, the first substream is the picture header substream (with marker ID PIH). The second and third substreams are the residual substreams for the primary and secondary components of the image coding format respectively (with marker IDs SORp and SORs respectively). The picture header substream comprises parameters indicating the number of channels in the second and third substreams. These parameters can be specified in field num_chs[0] and num_chs[l] for the primary and secondary residual substreams respectively. The decoder can parse the picture header substream, which comprises parsing syntax elements of the parameters for the primary and secondary residual bitstreams to determine the number of channels in the primary and secondary residual bitstreams. The decoder can then parse the second and third substreams and reconstruct the image signaled by the bitstream in dependence on the determined numbers of channels for each component. The second and third substreams may be parsed in any order; that is the primary and secondary residual substreams may be parsed in any order (i.e. primary then secondary, or secondary then primary).

[0212] The number of channels to be parsed from the second substream (and optionally any additional substreams for components of the image coding format, such as the third substream mentioned above) may be limited. Only num_chs[comp] number of channels may be parsed from the second or additional substream. All other channels of the residual tensor in the second or additional substream can be zeroed (i.e. set to zero).

[0213] As discussed above, residual data decoding can be performed for two colour components comp=0..1. The components may be luma (0) and chroma (1). These components are signaled in primary and secondary residual bitstreams respectively.

[0214] In an example, the input of this process is:

[0215] - CodeStreamR[comp] for primary (comp =0) and secondary (comp =1) component residual;

[0216] - stream_size =CodeStreamSizeR[comp];

[0217] - mask_skip[comp] [C, / / 4,W4] generated by skip mask generation (as described above), comp=0..1 ,

[0218] C=(comp==0?Cp:Cs),

[0219] - sigma_Idx[comp] [C, / / 4,W4] - tensor of size which is the output sigma quantization,

[0220] - NumberOfThreads[comp] (the number of threads in the substream for the component)

[0221] - NumVerSubTensorSplits, NumVerSubTensorSplits.

[0222] Where:

[0223] - CodeStreamR is the residual codestream marker segment;

[0224] - CodeStreamSizeR is the size of residual codestream marker segment in bytes (excluding the first two bytes for marker and the bytes for marker segment length);

[0225] - NumberOfThreadsR is the number of threads in residual codestream marker segment,

[0226] - tensorSize [ NumRegions ] array of sizes (conumted in the number elements) in each region residual sub-tensor,

[0227] - sigmaldxlD [ NumRegions ] [0..tensorSize -1] is the one-dimensional array of quantized log-scale variance indexes for each region residual sub-tensor,

[0228] - maskSkiplD [ NumRegions ] [0.. tensorSize -1] is the one-dimensional array of skip mask for each region residual sub-tensor. num_ver_splits_minusl plus 1 specifies the number of vertical splits of residual tensor and tensors in latent prediction and reconstruction process. When num_ver_splits_minusl is not present, its value is inferred to be equal to 0. The variable NumVerSubTensorSplits is derived to be equal to num_ver_splits_minusl + 1. The maximum and the minimum value of NumVerSubTensorSplits can be constrained constrained depending on the profile and the level to which the codestream conforms.

[0229] The output of this process is:

[0230] - r [comp] [C, / i4,W4] - residual tensor, comp = 0 corresponds to primary (rr) and for the secondary component (fuv) residual tensor which is later used as an input for latent tensor reconstruction process.

[0231] Here C= num_chs[comp]. The values of C|>.C, / AI.M' I are defined in Table 1.

[0232] - First, arrays tensorSize , sigmaldxlD and maskSkiplD are determined though following steps.For comp =0..1

[0233] - C= num_chs[comp]

[0234] - For i_ver=0..NumVerSubTensorSplits- 1

[0235] - For j _hor=0..NumHorSubT ensorSplits- 1

[0236] - sudldx = i_ver * NumHorSubTensorSplits + j_hor

[0237] - tensorSize[ sudldx ]= RCVer[4][i_ver][2]* RCHor[4]|j_hor][2]*C

[0238] - Set £ = 0

[0239] - For c=0..C-l o For i= RCVer[4][i_ver][0].. RCVer[4][i_ver][l]-l

[0240] - For j= RCHor [4] [ j_hor] [0] ..RCHor [4] [ j_hor]

[0001] - 1

[0241] - sigmaldxlD [sudldx][k] = sigma_Idx[comp] [c, i, j\

[0242] - maskSkip ID [sudldx] [k] = mask_skip[comp] [c,i, j]

[0243] - k = k + 1

[0244] Second, the residual stream parsing process is performed with syntax as described above. The semantics of the residual tensor used in the process may be as below:

[0245] For comp=0..1

[0246] - If region_residual_in_its_own_substream_flag is equal to 1, BytesOfRegionInfo[comp] is set to be I :

[0247] - Otherwise, BytesOfRegionInfo[comp] is set to be the number of bytes that signal the res_stream_segment _offset[comp] [i] syntax elements and byte_alignment(), i=0..NumRegions-2. When NumRegions is equal to 1, BytesOfRegionlnfo [comp] is set to be 0.

[0248] - The variable ThreadMeanSizeR[comp] [sudldx] derived as follows:

[0249] - RegionStreamSizeRfcomp] [sudldx] = (sudldx < NumRegions) ? (res_stream_segment _offset[comp][sudldx+l]- res_stream_segment _offset[comp] [sudldx]) (CodeStreamSizeRfcomp] - BytesOfRegionlnfofcomp] - res_stream_segment _offset[comp] [NumRegions- 1 ] ).

[0250] - For i=0..NumRegions - 1

[0251] - ThreadMeanSizeR[comp] [sudldx] = floor( (RegionStreamSizeRfcomp] [sudldx] -

[0252] BytesOfThreadlnfo [comp] [sudldx] ) / NumberO fThreadsR [ comp ] ). res_stream_segment_offset[comp] [sudldx] is the number of bytes counted from the start of the O-th residual region codestream segment till the start of the sudldx-th residual region codestream segment. When region_residual_in_its_own_substream_flag is equal to 1 , the start of the O-th residual region codestream segment is the position right after parsing the region_idx[comp] syntax elements; otherwise, the start of the O-th residual region codestream segment is the position right after parsing the res_stream_segment _offset[comp] [i] syntax elements and byte_alignment(), i=0.. NumRegions -2. thread_size_delta_r[comp] [sudldx] [i] is the signed difference between ThreadMeanSizeR[ comp ] [sudldx] and the number of bytes of the i-th thread codestream segment in the sudldx- th residual region codestream segment.

[0253] BytesOfThreadInfo[comp][sudIdx] is the number of bytes that signal the thread_size_delta_r[comp] [sudldx] [i] syntax elements and byte_alignment() in the sudldx-th residual region codestream segment, i=O..NumberOfThreadsR-2. - state2[i] - array of values state2 for me-tANS decoding of i-th thread codestream segment (0<=i< NumberOffhreadsR).

[0254] - region_idx[comp] specifies the index, in the list of all regions in raster-scan order, of the region corresponding to the current residual tensor codestream for comp=0..1. Codestream shall not contain no more than one primary (comp=0) and no more than one secondary (comp=l) residual marker segment for each value of region_idx.

[0255] The residual stream parsing process receives as inputs above sigmaIdxlD[comp] and maskSkip lD[comp] and CodeStreamR [comp] inputs. The residual stream parsing process returns a one-dimensional rlD[ comp ] residual array.

[0256] - For comp =O..l

[0257] - Set

[0258] - CodeStreamR = CodeStreamR[comp],

[0259] - CodeStreamSizeR = CodeStreamSizeR[comp],

[0260] - NumberOfThreads = NumberOfThreads[comp],

[0261] - tensorSize= tensorS ize[ ],

[0262] - sigmaldxlD = sigma!dxlD[ ][comp],

[0263] - maskSkiplD= maskSkiplD[ ][comp],

[0264] - rlD= rlD[ ][comp]

[0265] - Invoke residual stream parsing process residual_stream (comp, CodeStreamR, CodeStreamSizeR).

[0266] Output r[comp] can be produced as follows:

[0267] - For comp =O..l

[0268] - C= num_chs[comp]

[0269] - Initialize r[comp] as zeros.

[0270] - For i_ver=0.. NumVerSubTensorSplits-1

[0271] - For j _hor=0..NumHorSubT ensorSplits- 1

[0272] - sudldx = i_ver * NumHorSubTensorSplits + j_hor

[0273] - Set £ = 0

[0274] - For c=0..C-l

[0275] - For i= RCVer[4][i_ver][0]..RCVer[4][i_ver][l]-l,

[0276] - For j= RCHor [4][ j_hor][0]..RCHor [4][ j_hor][l] -1

[0277] - r[comp] [c,z,j]= rlD[ sudldx ] [comp] [k]

[0278] - k = k + l

[0279] The following amendment to the picture header syntax table is proposed: num_chs[comp] is a number of residual channels in the primary residual substream for comp equal to 0 and in the secondary residual substream for comp equal to 1.

[0280] The parameter described herein may be used for progressive decoding of a bitstream comprising residual substreams for the primary and secondary components of the image coding format.

[0281] Fig. 14 is a flow diagram illustrating an exemplary method for decoding an image based on a neural network architecture. This method 1400 describes the decoding method in general terms, wherein, as a first step 1401, the method comprises receiving a bitstream including encoded data of an input signal, wherein the bitstream is formed of multiple substreams, a first substream of the multiple substreams comprising a parameter indicating a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream. At step 1402, the method comprises parsing the first substream, the parsing of the first substream comprising parsing syntax elements of the parameter to thereby determine the number of channels in the second substream for the component of the image coding format of the bitstream. At step 1403, the method comprises parsing the second substream in dependence on the determined number of channels in the second substream for the component of the image coding format of the bitstream.

[0282] Fig. 15 is a flow diagram illustrating an exemplary method for encoding. The embodiment according to FIG. 15 may be configured to provide output readily decoded by the decoding method described with reference to FIG. 14. The method 1500 of FIG. 15 will be described as being performed by a neural network system of one or more computers located in one or more locations. For example, a system configured to perform image compression, e.g., the neural network of FIG. 1 can perform the method 1500.

[0283] This method 1500 describes the encoding method in general terms, wherein, as a first step 1501, the method comprises receiving an input signal. At step 1502, the method comprises encoding the input signal into a bitstream, the bitstream being formed of multiple substreams, a first substream of the multiple substreams comprising a parameter indicating a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream.

[0284] While operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Moreover, the separation of various system modules and components in the embodiments described above should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

[0285] Particular embodiments of the subject matter have been described. Other embodiments are within the scope of the following claims. For example, the actions recited in the claims can be performed in a different order and still achieve desirable results. As one example, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In certain implementations, multitasking and parallel processing may be advantageous.

[0286] Fig. 16 shows a device 1600 for decoding for processing by a neural network based unit. The device comprises a receiving unit 16a configured to receive a bitstream including encoded data of an input signal, wherein the bitstream is formed of multiple substreams, a first substream of the multiple substreams comprising a parameter indicating a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream. The device comprises a parsing unit 16b configured to parse the first substream, the parsing of the first substream comprising parsing syntax elements of the parameter to thereby determine the number of channels in the second substream for the component of the image coding format of the bitstream; and parse the second substream in dependence on the determined number of channels in the second substream for the component of the image coding format of the bitstream.

[0287] Fig. 17 shows a device 1700 for encoding for processing by a neural network based unit. The device comprises a receiving unit 17a configured to receive an input signal. The device comprises an encoding unit 17b configured to encode the input signal into a bitstream, the bitstream being formed of multiple substreams, a first substream of the multiple substreams comprising a parameter indicating a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream.

[0288] In a possible implementation, the encoding device may generate the parameter for inclusion in the first substream and encode the parameter.

[0289] The arithmetic coding of the present disclosure may be readily applied to encoding of feature maps of a neural network or in classic picture (still or video) encoding and decoding. The neural networks may be used for any purpose, in particular for encoding and decoding or pictures (still or moving), or encoding and decoding of picture-related data such as motion flow or motion vectors or other parameters. The neural network may also be used for computer vision applications such as classification of images, depth detection, segmentation map determination, object recognition of identification or the like.

[0290] The entropy decoding may be performed in parallel, for example by a multi-core decoder. In addition, only parts of the entropy decoding may be performed in parallel.

[0291] The input data channels may refer to channels obtained by processing some data by a neural network. For example, the input data may be feature channels such as output channels or latent representation channels of a neural network. In an exemplary implementation, the neural network is a deep neural network and / or a convolutional neural network or the like. The neural network may be trained to process pictures (still or moving). The processing may be for picture encoding and reconstruction or for computer vision such as object recognition, classification, segmentation, or the like. In general, the present disclosure is not limited to any particular kind of tasks or neural networks. Rather, the present disclosure is applicable for encoding any kind of data coming from a plurality of channels, which are to be generally understood as any sources of data. Moreover, the channels may be provided by a pre-processing of source data.

[0292] Some exemplary implementations in hardware and software

[0293] The corresponding system which may deploy the above-mentioned encoder-decoder processing chain is illustrated in Fig. 18. Fig. 18. is a schematic block diagram illustrating an example coding system, e.g. a video, image, audio, and / or other coding system (or short coding system) that may utilize techniques of this present application. Video encoder 20 (or short encoder 20) and video decoder 30 (or short decoder 30) of video coding system 10 represent examples of devices that may be configured to perform techniques in accordance with various examples described in the present application. For example, the video coding and decoding may employ neural network such which may be distributed and which may apply the above-mentioned bitstream parsing and / or bitstream generation to convey feature maps between the distributed computation nodes (two or more).

[0294] As shown in Fig. 18., the coding system 10 comprises a source device 12 configured to provide encoded picture data 21 e.g. to a destination device 14 for decoding the encoded picture data 13.

[0295] The source device 12 comprises an encoder 20, and may additionally, i.e. optionally, comprise a picture source 16, a preprocessor (or pre-processing unit) 18, e.g. a picture pre-processor 18, and a communication interface or communication unit 22.

[0296] The picture source 16 may comprise or be any kind of picture capturing device, for example a camera for capturing a real-world picture, and / or any kind of a picture generating device, for example a computer-graphics processor for generating a computer animated picture, or any kind of other device for obtaining and / or providing a real- world picture, a computer generated picture (e.g. a screen content, a virtual reality (VR) picture) and / or any combination thereof (e.g. an augmented reality (AR) picture). The picture source may be any kind of memory or storage storing any of the aforementioned pictures.

[0297] In distinction to the pre-processor 18 and the processing performed by the pre-processing unit 18, the picture or picture data 17 may also be referred to as raw picture or raw picture data 17.

[0298] Pre-processor 18 is configured to receive the (raw) picture data 17 and to perform pre-processing on the picture data 17 to obtain a pre-processed picture 19 or pre-processed picture data 19. Pre-processing performed by the pre-processor 18 may, e.g., comprise trimming, color format conversion (e.g. from RGB to YCbCr), color correction, or de-noising. It can be understood that the pre-processing unit 18 may be optional component. It is noted that the pre-processing may also employ a neural network (such as in any of Figs. 1 to 7) which uses the presence indicator signaling.

[0299] The video encoder 20 is configured to receive the pre-processed picture data 19 and provide encoded picture data 21.

[0300] Communication interface 22 of the source device 12 may be configured to receive the encoded picture data 21 and to transmit the encoded picture data 21 (or any further processed version thereof) over communication channel 13 to another device, e.g. the destination device 14 or any other device, for storage or direct reconstruction.

[0301] The destination device 14 comprises a decoder 30 (e.g. a video decoder 30), and may additionally, i.e. optionally, comprise a communication interface or communication unit 28, a post-processor 32 (or post-processing unit 32) and a display device 34.

[0302] The communication interface 28 of the destination device 14 is configured receive the encoded picture data 21 (or any further processed version thereof), e.g. directly from the source device 12 or from any other source, e.g. a storage device, e.g. an encoded picture data storage device, and provide the encoded picture data 21 to the decoder 30.

[0303] The communication interface 22 and the communication interface 28 may be configured to transmit or receive the encoded picture data 21 or encoded data 13 via a direct communication link between the source device 12 and the destination device 14, e.g. a direct wired or wireless connection, or via any kind of network, e.g. a wired or wireless network or any combination thereof, or any kind of private and public network, or any kind of combination thereof.

[0304] The communication interface 22 may be, e.g., configured to package the encoded picture data 21 into an appropriate format, e.g. packets, and / or process the encoded picture data using any kind of transmission encoding or processing for transmission over a communication link or communication network.

[0305] The communication interface 28, forming the counterpart of the communication interface 22, may be, e.g., configured to receive the transmitted data and process the transmission data using any kind of corresponding transmission decoding or processing and / or de-packaging to obtain the encoded picture data 21.

[0306] Both, communication interface 22 and communication interface 28 may be configured as unidirectional communication interfaces as indicated by the arrow for the communication channel 13 in Fig. 18 pointing from the source device 12 to the destination device 14, or bi-directional communication interfaces, and may be configured, e.g. to send and receive messages, e.g. to set up a connection, to acknowledge and exchange any other information related to the communication link and / or data transmission, e.g. encoded picture data transmission. The decoder 30 is configured to receive the encoded picture data 21 and provide decoded picture data 31 or a decoded picture 31.

[0307] The post-processor 32 of destination device 14 is configured to post-process the decoded picture data 31 (also called reconstructed picture data), e.g. the decoded picture 31, to obtain post-processed picture data 33, e.g. a post-processed picture 33. The post-processing performed by the post-processing unit 32 may comprise, e.g. color format conversion (e.g. from YCbCr to RGB), color correction, trimming, or re-sampling, or any other processing, e.g. for preparing the decoded picture data 31 for display, e.g. by display device 34. The display device 34 of the destination device 14 is configured to receive the post-processed picture data 33 for displaying the picture, e.g. to a user or viewer. The display device 34 may be or comprise any kind of display for representing the reconstructed picture, e.g. an integrated or external display or monitor. The displays may, e.g. comprise liquid crystal displays (LCD), organic light emitting diodes (OLED) displays, plasma displays, projectors , micro LED displays, liquid crystal on silicon (LCoS), digital light processor (DLP) or any kind of other display.

[0308] Although Fig. 18 depicts the source device 12 and the destination device 14 as separate devices, embodiments of devices may also comprise both or both functionalities, the source device 12 or corresponding functionality and the destination device 14 or corresponding functionality. In such embodiments the source device 12 or corresponding functionality and the destination device 14 or corresponding functionality may be implemented using the same hardware and / or software or by separate hardware and / or software or any combination thereof.

[0309] As will be apparent for the skilled person based on the description, the existence and (exact) split of functionalities of the different units or functionalities within the source device 12 and / or destination device 14 as shown in Fig. 18 may vary depending on the actual device and application.

[0310] The encoder 20 (e.g. a video encoder 20) or the decoder 30 (e.g. a video decoder 30) or both encoder 20 and decoder 30 may be implemented via processing circuitry, such as one or more microprocessors, digital signal processors (DSPs), application- specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), discrete logic, hardware, video coding dedicated or any combinations thereof. The encoder 20 may be implemented via processing circuitry 46 to embody the various modules including the neural network or its parts. The decoder 30 may be implemented via processing circuitry 46 to embody any coding system or subsystem described herein. The processing circuitry may be configured to perform the various operations as discussed later. If the techniques are implemented partially in software, a device may store instructions for the software in a suitable, non-transitory computer-readable storage medium and may execute the instructions in hardware using one or more processors to perform the techniques of this disclosure. Either of video encoder 20 and video decoder 30 may be integrated as part of a combined encoder / decoder (CODEC) in a single device, for example, as shown in Fig. 19.

[0311] Source device 12 and destination device 14 may comprise any of a wide range of devices, including any kind of handheld or stationary devices, e.g. notebook or laptop computers, mobile phones, smart phones, tablets or tablet computers, cameras, desktop computers, set-top boxes, televisions, display devices, digital media players, video gaming consoles, video streaming devices(such as content services servers or content delivery servers), broadcast receiver device, broadcast transmitter device, or the like and may use no or any kind of operating system. In some cases, the source device 12 and the destination device 14 may be equipped for wireless communication. Thus, the source device 12 and the destination device 14 may be wireless communication devices. In some cases, video coding system 10 illustrated in Fig. 18 is merely an example and the techniques of the present application may apply to video coding settings (e.g., video encoding or video decoding) that do not necessarily include any data communication between the encoding and decoding devices. In other examples, data is retrieved from a local memory, streamed over a network, or the like. A video encoding device may encode and store data to memory, and / or a video decoding device may retrieve and decode data from memory. In some examples, the encoding and decoding is performed by devices that do not communicate with one another, but simply encode data to memory and / or retrieve and decode data from memory.

[0312] Fig. 20 is a schematic diagram of a video coding device 8000 according to an embodiment of the disclosure. The video coding device 8000 is suitable for implementing the disclosed embodiments as described herein. In an embodiment, the video coding device 8000 may be a decoder such as video decoder 30 of Fig. 18 or an encoder such as video encoder 20 ofFig. 18.

[0313] The video coding device 8000 comprises ingress ports 8010 (or input ports 8010) and receiver units (Rx) 8020 for receiving data; a processor, logic unit, or central processing unit (CPU) 8030 to process the data; transmitter units (Tx) 8040 and egress ports 8050 (or output ports 8050) for transmitting the data; and a memory 8060 for storing the data. The video coding device 8000 may also comprise optical-to-electrical (OE) components and electrical-to-optical (EO) components coupled to the ingress ports 8010, the receiver units 8020, the transmitter units 8040, and the egress ports 8050 for egress or ingress of optical or electrical signals. The processor 8030 is implemented by hardware and software. The processor 8030 may be implemented as one or more CPU chips, cores (e.g., as a multi-core processor), FPGAs, ASICs, and DSPs. The processor 8030 is in communication with the ingress ports 8010, receiver units 8020, transmitter units 8040, egress ports 8050, and memory 8060. The processor 8030 comprises a neural network based codec 8070. The neural network based codec 8070 implements the disclosed embodiments described above. For instance, the neural network based codec 8070 implements, processes, prepares, or provides the various coding operations. The inclusion of the neural network based codec 8070 therefore provides a substantial improvement to the functionality of the video coding device 8000 and effects a transformation of the video coding device 8000 to a different state. Alternatively, the neural network based codec 8070 is implemented as instructions stored in the memory 8060 and executed by the processor 8030.

[0314] The memory 8060 may comprise one or more disks, tape drives, and solid-state drives and may be used as an over-flow data storage device, to store programs when such programs are selected for execution, and to store instructions and data that are read during program execution. The memory 8060 may be, for example, volatile and / or non-volatile and may be a read-only memory (ROM), random access memory (RAM), ternary content-addressable memory (TCAM), and / or static random-access memory (SRAM).

[0315] Fig. 21 is a simplified block diagram of an apparatus that may be used as either or both of the source device 12 and the destination device 14 from Fig. 18 according to an exemplary embodiment.

[0316] A processor 9002 in the apparatus 9000 can be a central processing unit. Alternatively, the processor 9002 can be any other type of device, or multiple devices, capable of manipulating or processing information now-existing or hereafter developed. Although the disclosed implementations can be practiced with a single processor as shown, e.g., the processor 9002, advantages in speed and efficiency can be achieved using more than one processor. A memory 9004 in the apparatus 9000 can be a read only memory (ROM) device or a random access memory (RAM) device in an implementation. Any other suitable type of storage device can be used as the memory 9004. The memory 9004 can include code and data 9006 that is accessed by the processor 9002 using a bus 9012. The memory 9004 can further include an operating system 9008 and application programs 9010, the application programs 9010 including at least one program that permits the processor 9002 to perform the methods described here. For example, the application programs 9010 can include applications 1 through N, which further include a video coding application that performs the methods described here.

[0317] The apparatus 9000 can also include one or more output devices, such as a display 9018. The display 9018 may be, in one example, a touch sensitive display that combines a display with a touch sensitive element that is operable to sense touch inputs. The display 9018 can be coupled to the processor 9002 via the bus 9012.

[0318] Although depicted here as a single bus, the bus 9012 of the apparatus 9000 can be composed of multiple buses. Further, a secondary storage can be directly coupled to the other components of the apparatus 9000 or can be accessed via a network and can comprise a single integrated unit such as a memory card or multiple units such as multiple memory cards. The apparatus 9000 can thus be implemented in a wide variety of configurations.

[0319] Fig. 22 is a block diagram of a video coding system 10000 according to an embodiment of the disclosure. A platform 10002 in the system 10000 can be could sever or local sever. Alternatively, the platform 10002 can be any other type of device, or multiple devices, capable of calculation, storing, transcoding, encryption, rendering, decoding or encoding. Although the disclosed implementations can be practiced with a single platform as shown, e.g., the platform 10002, advantages in speed and efficiency can be achieved using more than one platform. A content delivery network (CDN) 10004 in the system 10000 can be a group of geographically distributed servers. Alternatively, the CDN 10004 can be any other type of device, or multiple devices, capable of data buffering, scheduling, dissemination or speed up the delivery of web content by bringing it closer to where users are. Although the disclosed implementations can be practiced with a single CDN as shown, e.g., the CDN 10004, advantages in speed and efficiency can be achieved using more than one CDN. A terminal 10006 in the apparatus 10000 can be a mobile phone, computer, television, laptop, camera. Alternatively, the terminal 10006 can be any other type of device, or multiple devices, capable of displaying video or image.

[0320] The applicant hereby discloses in isolation each individual feature described herein and any combination of two or more such features, to the extent that such features or combinations are capable of being carried out based on the present specification as a whole in the light of the common general knowledge of a person skilled in the art, irrespective of whether such features or combinations of features solve any problems disclosed herein, and without limitation to the scope of the claims. The applicant indicates that aspects of the present invention may consist of any such individual feature or combination of features. In view of the foregoing description, it will be evident to a person skilled in the art that various modifications may be made within the scope of the invention.

Claims

CLAIMS1. A decoding method (1400), implemented by a decoder, the method comprising: receiving (1401) a bitstream including encoded data of an input signal, wherein the bitstream is formed of multiple substreams, a first substream of the multiple substreams comprising a parameter indicating a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream; parsing (1402) the first substream, the parsing of the first substream comprising parsing syntax elements of the parameter to thereby determine the number of channels in the second substream for the component of the image coding format of the bitstream; and parsing (1403) the second substream in dependence on the determined number of channels in the second substream for the component of the image coding format of the bitstream.

2. The decoding method as claimed in claim 1 , wherein the second substream is a residual substream for the component of the image coding format of the bitstream.

3. The decoding method as claimed in claim 2, wherein the component of the image coding format is indicated by a component value, wherein if the component value is 0, the parameter relates to a number of residual channels in a primary residual substream.

4. The decoding method as claimed in claim 2 or claim 3, wherein the component of the image coding format is indicated by a component value, wherein if the component value is 1 , the parameter relates to a number of residual channels in a secondary residual substream.

5. The decoding method as claimed in any preceding claim, wherein the parameter is signaled as 8 bits for a primary component of the image coding format and 7 bits for a secondary component of the image coding format of the bitstream.

6. The decoding method as claimed in any of claims 1 to 4, wherein the parameter is signaled as 8 bits for each of a primary component of the image coding format and a secondary component of the image coding forma of the bitstream.

7. The decoding method as claimed in any preceding claim, wherein the component of the image coding format is luma or chroma.

8. The decoding method as claimed in any preceding claim, wherein the parameter is specified in a field num_chs[comp] in the first substream, where comp is the component of the image coding format of the bitstream.

9. The decoding method as claimed in claim 8, wherein the maximum value of the field num_chs[comp] is 160 where comp is equal to 0 and 96 where comp is equal to 1.

10. The decoding method as claimed in any preceding claim, wherein the first substream is a picture header substream of the bitstream.

11. The decoding method as claimed in claim 10, wherein each of the multiple substreams comprises an Marker ID indicating a data type of the respective substream, wherein the marker ID for the picture header substream is PIH and wherein the parsing comprises parsing the marker ID of the picture header substream.3512. The decoding method as claimed in any preceding claim, wherein each of the multiple substreams comprises an Marker ID indicating a data type of the respective substream, wherein the marker ID for the second substream is SORc, where c indicates the component of the image coding format of the bitstream.

13. The decoding method as claimed in any preceding claim, wherein the method further comprises, subsequent to determining the number of channels in the second substream for the component of an image coding format of the bitstream, configuring the parsing process for the second substream.

14. A device (1600) for decoding data for picture or video processing from a bitstream, the device comprising: a receiving unit (16a) configured to receive (1401) a bitstream including encoded data of an input signal, wherein the bitstream is formed of multiple substreams, a first substream of the multiple substreams comprising a parameter indicating a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream; and a parsing unit (16b) configured to: parse (1402) the first substream, the parsing of the first substream comprising parsing syntax elements of the parameter to thereby determine the number of channels in the second substream for the component of the image coding format of the bitstream; and parse (1403) the second substream in dependence on the determined number of channels in the second substream for the component of the image coding format of the bitstream.

15. An encoding method (1500), implemented by an encoder, the method comprising: receiving (1501) an input signal; and encoding (1502) the input signal into a bitstream, the bitstream being formed of multiple substreams, a first substream of the multiple substreams comprising a parameter indicating a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream.

16. The encoding method as claimed in claim 15, where the encoding comprises generating the parameter.

17. The encoding method as claimed in claim 15 or claim 16, wherein the second substream is a residual substream for the component of the image coding format of the bitstream.

18. The encoding method as claimed in any of claims 15 to 17, wherein the component of the image coding format is indicated by a component value, wherein if the component value is 0, the parameter relates to a number of residual channels in a primary residual substream.

19. The encoding method as claimed in any of claims 15 to 17, wherein the component of the image coding format is indicated by a component value, wherein if the component value is 1, the parameter relates to a number of residual channels in a second residual substream.

20. The encoding method as claimed in any of claims 15 to 19, wherein the parameter is signaled as 8 bits for a primary component of the image coding format and 7 bits for a secondary component of the image coding format of the bitstream.

21. The encoding method as claimed in any of claims 15 to 19, wherein the parameter is signaled as 8 bits for each of a primary component of the image coding format and a secondary component of the image coding forma of the bitstream.3622. The encoding method as claimed in any of claims 15 to 21, wherein the component of the image coding format is luma or chroma.

23. The encoding method as claimed in any of claims 15 to 22, wherein the parameter is specified in a field num_chs[comp] in the first substream, where comp is the component of the image coding format of the bitstream.

24. The encoding method as claimed in claim 23, wherein the maximum value of the field num_chs[comp] is 160 where comp is equal to 0 and 96 where comp is equal to 1.

25. The encoding method as claimed in any of claims 15 to 24, the method comprising for each substream of the multiple substreams: generating an Marker ID indicating a data type encoded within the substream; inserting the Marker ID within the substream.

26. The encoding method as claimed in any of claims 15 to 25, wherein the first substream is a picture header substream of the bitstream.

27. The encoding method as claimed in claim 26 as dependent on claim 25, wherein the marker ID for the picture header substream is PIH.

28. The encoding method as claimed in claim 27 or claim 26 as dependent on claim 25, wherein the marker ID for the second substream is SORc, where c indicates the component of the image coding format of the bitstream.

29. A device (1700) for encoding data for picture or video processing to obtain a bitstream, the device comprising: a receiving unit (17a) configured to receive (1501) an input signal; and an encoding unit (17b) configured to encode (1502) the input signal into a bitstream bitstream, the bitstream being formed of multiple substreams, a first substream of the multiple substreams comprising a parameter indicating a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream.

30. A decoding apparatus comprising processing circuitry configured to execute steps of the method according to any of claims 1 to 13.

31. An encoding apparatus comprising processing circuitry configured to execute steps of the method according to any of claims 15 to 28.

32. A decoder comprising one or more processors and a non- transitory computer-readable storage medium coupled to the one or more processors, wherein the storage medium stores programming for execution by the one or more processors, wherein the programming, when executed by the one or more processors, configures the decoder to carry out the method according to any of claims 1 to 13.

33. A non-transitory storage medium comprising a bitstream encoded by the method of any of claims 15 to 28.

34. A computer program stored on a non-transitory medium and including code instructions, which, when executed on one or more processor, causes the one or more processor to execute the method according to any of claims 1 to 13, or 15 to 28.

35. A system for delivering a bitstream, the system including at least one storage medium configured to store at least one bitstream generated by the encoding method described in any of claims 15 to 28.

36. A system for delivering a bitstream, the system comprising: at least one storage medium configured to store at least one bitstream generated by the method of any one of the claims 15 to 28; and a video streaming device configured to obtain the bitstream from one of the at least one storage medium and send the bitstream to a terminal device, wherein the video streaming device comprises a content server or a content delivery server.

37. The system according to claim 36, further comprising one or more processor configured to perform encryption processing on at least one bitstream to obtain at least one encrypted bitstream, the at least one storage medium configured to store the encrypted bitstream; or the one or more processor configured to convert the bitstream in a first format into a bitstream in a second format, the at least one storage medium configured to store the bitstream in the second format.

38. The system according to claim 36 or 37, further comprising: a receiver configured to receive a first operation request; wherein the one or more processor is configured to determine a target bitstream in the at least one storage medium in response to the first operation request; and a transmitter configured to send the target bitstream to a terminal-side apparatus.

39. The system according to claim 38, wherein the one or more processor is further configured to encapsulate the bitstream to obtain a transport stream in a first format, wherein the transmitter is further configured to: send the transport stream in the first format to a terminal-side apparatus for display; or send the transport stream in the first format to storage space for storage.

40. A bitstream (900) including encoded data of an input signal, wherein the bitstream is formed of multiple substreams, a first substream of the multiple substreams comprising a parameter indicating a number of channels in a second substream of the multiple substreams for a component of an image coding format of the bitstream.

Images