Patent Application
20240430428
Embodiments of the present disclosure relates generally to visual data processing techniques, and more particularly, to neural network-based visual data coding.
The past decade has witnessed the rapid development of deep learning in a variety of areas, especially in computer vision and image processing. Neural network was invented originally with the interdisciplinary research of neuroscience and mathematics. It has shown strong capabilities in the context of non-linear transform and classification. Neural network-based image/video compression technology has gained significant progress during the past half decade. It is reported that the latest neural network-based image compression algorithm achieves comparable rate-distortion (R-D) performance with Versatile Video Coding (VVC). With the performance of neural image compression continually being improved, neural network-based video compression has become an actively developing research area. However, coding efficiency and/or coding quality of neural network-based visual data coding is generally expected to be further improved.
Embodiments of the present disclosure provide a solution for visual data processing.
In a first aspect, a method for visual data processing is proposed. The method comprises: obtaining, for a conversion between visual data and a bitstream of the visual data, a first representation of the visual data, the first representation being obtained by quantizing a second representation of the visual data, the second representation being generated based on applying a first neural network to the visual data; adjusting a plurality of sets of first samples of the first representation with different parameters; and performing the conversion based on the plurality of sets of adjusted first samples.
According to the method in accordance with the first aspect of the present disclosure, samples of a first representation (such as quantized latent representation or quantized residual latent representation) of the visual data are divided into a plurality of sets, and the plurality of sets of first samples are adjusted with different parameters. Thereby, the proposed method makes it possible to quantize the samples with different parameters, so as to maximizing the quality of the reconstructed visual data while at the same time reducing the number of bits to be transmitted. As such, the proposed method can advantageously improve the coding quality and coding efficiency.
In a second aspect, an apparatus for visual data processing is proposed. The apparatus comprises a processor and a non-transitory memory with instructions thereon. The instructions upon execution by the processor, cause the processor to perform a method in accordance with the first aspect of the present disclosure.
In a third aspect, a non-transitory computer-readable storage medium is proposed. The non-transitory computer-readable storage medium stores instructions that cause a processor to perform a method in accordance with the first aspect of the present disclosure.
In a fourth aspect, another non-transitory computer-readable recording medium is proposed. The non-transitory computer-readable recording medium stores a bitstream of visual data which is generated by a method performed by an apparatus for visual data processing. The method comprises: obtaining a first representation of the visual data, the first representation being obtained by quantizing a second representation of the visual data, the second representation being generated based on applying a first neural network to the visual data; adjusting a plurality of sets of first samples of the first representation with different parameters; and generating the bitstream based on the plurality of sets of adjusted first samples.
In a fifth aspect, a method for storing a bitstream of visual data is proposed. The method comprises: obtaining a first representation of the visual data, the first representation being obtained by quantizing a second representation of the visual data, the second representation being generated based on applying a first neural network to the visual data; adjusting a plurality of sets of first samples of the first representation with different parameters; generating the bitstream based on the plurality of sets of adjusted first samples; and storing the bitstream in a non-transitory computer-readable recording medium.
This Summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.
Through the following detailed description with reference to the accompanying drawings, the above and other objectives, features, and advantages of example embodiments of the present disclosure will become more apparent. In the example embodiments of the present disclosure, the same reference numerals usually refer to the same components.
Throughout the drawings, the same or similar reference numerals usually refer to the same or similar elements.
Principle of the present disclosure will now be described with reference to some embodiments. It is to be understood that these embodiments are described only for the purpose of illustration and help those skilled in the art to understand and implement the present disclosure, without suggesting any limitation as to the scope of the disclosure. The disclosure described herein can be implemented in various manners other than the ones described below.
In the following description and claims, unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skills in the art to which this disclosure belongs.
References in the present disclosure to “one embodiment,” “an embodiment,” “an example embodiment,” and the like indicate that the embodiment described may include a particular feature, structure, or characteristic, but it is not necessary that every embodiment includes the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an example embodiment, it is submitted that it is within the knowledge of one skilled in the art to affect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.
It shall be understood that although the terms “first” and “second” etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and similarly, a second element could be termed a first element, without departing from the scope of example embodiments. As used herein, the term “and/or” includes any and all combinations of one or more of the listed terms.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises”, “comprising”, “has”, “having”, “includes” and/or “including”, when used herein, specify the presence of stated features, elements, and/or components etc., but do not preclude the presence or addition of one or more other features, elements, components and/or combinations thereof.
The visual data source 112 may include a source such as a visual data capture device. Examples of the visual data capture device include, but are not limited to, an interface to receive visual data from a visual data provider, a computer graphics system for generating visual data, and/or a combination thereof.
The visual data may comprise one or more pictures of a video or one or more images. The visual data encoder 114 encodes the visual data from the visual data source 112 to generate a bitstream. The bitstream may include a sequence of bits that form a coded representation of the visual data. The bitstream may include coded pictures and associated visual data. The coded picture is a coded representation of a picture. The associated visual data may include sequence parameter sets, picture parameter sets, and other syntax structures. The I/O interface 116 may include a modulator/demodulator and/or a transmitter. The encoded visual data may be transmitted directly to destination device 120 via the I/O interface 116 through the network 130A. The encoded visual data may also be stored onto a storage medium/server 130B for access by destination device 120.
The destination device 120 may include an I/O interface 126, a visual data decoder 124, and a display device 122. The I/O interface 126 may include a receiver and/or a modem. The I/O interface 126 may acquire encoded visual data from the source device 110 or the storage medium/server 130B. The visual data decoder 124 may decode the encoded visual data. The display device 122 may display the decoded visual data to a user. The display device 122 may be integrated with the destination device 120, or may be external to the destination device 120 which is configured to interface with an external display device.
The visual data encoder 114 and the visual data decoder 124 may operate according to a visual data coding standard, such as video coding standard or still picture coding standard and other current and/or further standards.
Some exemplary embodiments of the present disclosure will be described in detailed hereinafter. It should be understood that section headings are used in the present document to facilitate ease of understanding and do not limit the embodiments disclosed in a section to only that section. Furthermore, while certain embodiments are described with reference to Versatile Video Coding or other specific visual data codecs, the disclosed techniques are applicable to other coding technologies also. Furthermore, while some embodiments describe coding steps in detail, it will be understood that corresponding steps decoding that undo the coding will be implemented by a decoder. Furthermore, the term visual data processing encompasses visual data coding or compression, visual data decoding or decompression and visual data transcoding in which visual data are represented from one compressed format into another compressed format or at a different compressed bitrate.
A neural network based image and video compression method comprising quantization and dequantization, wherein the samples of the latent representation are divided into at least 2 subsets and samples in each subset are quantized/dequantized using different quantization step sizes.
The past decade has witnessed the rapid development of deep learning in a variety of areas, especially in computer vision and image processing. Inspired from the great success of deep learning technology to computer vision areas, many researchers have shifted their attention from conventional image/video compression techniques to neural image/video compression technologies. Neural network was invented originally with the interdisciplinary research of neuroscience and mathematics. It has shown strong capabilities in the context of non-linear transform and classification. Neural network-based image/video compression technology has gained significant progress during the past half decade. It is reported that the latest neural network-based image compression algorithm achieves comparable R-D performance with Versatile Video Coding (VVC), the latest video coding standard developed by Joint Video Experts Team (JVET) with experts from MPEG and VCEG. With the performance of neural image compression continually being improved, neural network-based video compression has become an actively developing research area. However, neural network-based video coding still remains in its infancy due to the inherent difficulty of the problem.
2.1. Image/Video compression
Image/video compression usually refers to the computing technology that compresses image/video into binary code to facilitate storage and transmission. The binary codes may or may not support losslessly reconstructing the original image/video, termed lossless compression and lossy compression. Most of the efforts are devoted to lossy compression since lossless reconstruction is not necessary in most scenarios. Usually the performance of image/video compression algorithms is evaluated from two aspects, i.e. compression ratio and reconstruction quality. Compression ratio is directly related to the number of binary codes, the less the better; Reconstruction quality is measured by comparing the reconstructed image/video with the original image/video, the higher the better.
Image/video compression techniques can be divided into two branches, the classical video coding methods and the neural-network-based video compression methods. Classical video coding schemes adopt transform-based solutions, in which researchers have exploited statistical dependency in the latent variables (e.g., DCT or wavelet coefficients) by carefully hand-engineering entropy codes modeling the dependencies in the quantized regime. Neural network-based video compression is in two flavors, neural network-based coding tools and end-to-end neural network-based video compression. The former is embedded into existing classical video codecs as coding tools and only serves as part of the framework, while the latter is a separate framework developed based on neural networks without depending on classical video codecs.
In the last three decades, a series of classical video coding standards have been developed to accommodate the increasing visual content. The international standardization organizations ISO/IEC has two expert groups namely Joint Photographic Experts Group (JPEG) and Moving Picture Experts Group (MPEG), and ITU-T also has its own Video Coding Experts Group (VCEG) which is for standardization of image/video coding technology. The influential video coding standards published by these organizations include JPEG, JPEG 2000, H.262, H.264/AVC and H.265/HEVC. After H.265/HEVC, the Joint Video Experts Team (JVET) formed by MPEG and VCEG has been working on a new video coding standard Versatile Video Coding (VVC). The first version of VVC was released in July 2020. An average of 50% bitrate reduction is reported by VVC under the same visual quality compared with HEVC.
Neural network-based image/video compression is not a new invention since there were a number of researchers working on neural network-based image coding. But the network architectures were relatively shallow, and the performance was not satisfactory. Benefit from the abundance of data and the support of powerful computing resources, neural network-based methods are better exploited in a variety of applications. At present, neural network-based image/video compression has shown promising improvements, confirmed its feasibility. Nevertheless, this technology is still far from mature and a lot of challenges need to be addressed.
2.2. Neural networks
Neural networks, also known as artificial neural networks (ANN), are the computational models used in machine learning technology which are usually composed of multiple processing layers and each layer is composed of multiple simple but non-linear basic computational units. One benefit of such deep networks is believed to be the capacity for processing data with multiple levels of abstraction and converting data into different kinds of representations. Note that these representations are not manually designed; instead, the deep network including the processing layers is learned from massive data using a general machine learning procedure. Deep learning eliminates the necessity of handcrafted representations, and thus is regarded useful especially for processing natively unstructured data, such as acoustic and visual signal, whilst processing such data has been a longstanding difficulty in the artificial intelligence field.
2.3. Neural networks for image compression
Existing neural networks for image compression methods can be classified in two categories, i.e., pixel probability modeling and auto-encoder. The former one belongs to the predictive coding strategy, while the latter one is the transform-based solution. Sometimes, these two methods are combined together in literature.
According to Shannon's information theory, the optimal method for lossless coding can reach the minimal coding rate—log2 p(x) where p(x) is the probability of symbol x. A number of lossless coding methods were developed in literature and among them arithmetic coding is believed to be among the optimal ones. Given a probability distribution p(x), arithmetic coding ensures that the coding rate to be as close as possible to its theoretical limit—log2p(x) without considering the rounding error. Therefore, the remaining problem is to how to determine the probability, which is however very challenging for natural image/video due to the curse of dimensionality.
Following the predictive coding strategy, one way to model p(x) is to predict pixel probabilities one by one in a raster scan order based on previous observations, where x is an image.
where m and n are the height and width of the image, respectively. The previous observation is also known as the context of the current pixel. When the image is large, it can be difficult to estimate the conditional probability, thereby a simplified method is to limit the range of its context.
where k is a pre-defined constant controlling the range of the context.
It should be noted that the condition may also take the sample values of other color components into consideration. For example, when coding the RGB color component, R sample is dependent on previously coded pixels (including R/G/B samples), the current G sample may be coded according to previously coded pixels and the current R sample, while for coding the current B sample, the previously coded pixels and the current R and G samples may also be taken into consideration. Neural networks were originally introduced for computer vision tasks and have been proven to be effective in regression and classification problems. Therefore, it has been proposed using neural networks to estimate the probability of p(xi) given its context x1, x2, . . . , xi-1. In [7], the pixel probability is proposed for binary images, i.e., xiϵ{−1, +1}. The neural autoregressive distribution estimator (NADE) is designed for pixel probability modeling, where is a feed-forward network with a single hidden layer. A similar work is presented, where the feed-forward network also has connections skipping the hidden layer, and the parameters are also shared. Experiments are performed on the binarized MNIST dataset. In an existing design, NADE is extended to a real-valued model RNADE, where the probability p(xi|x1, . . . , xi-1) is derived with a mixture of Gaussians. Their feed-forward network also has a single hidden layer, but the hidden layer is with rescaling to avoid saturation and uses rectified linear unit (ReLU) instead of sigmoid. In an existing design, NADE and RNADE are improved by using reorganizing the order of the pixels and with deeper neural networks. Designing advanced neural networks plays an important role in improving pixel probability modeling. In an existing design, multi-dimensional long short-term memory (LSTM) is proposed, which is working together with mixtures of conditional Gaussian scale mixtures for probability modeling. LSTM is a special kind of recurrent neural networks (RNNs) and is proven to be good at modeling sequential data. The spatial variant of LSTM is used for images later in an existing design. Several different neural networks are studied, including RNNs and CNNs namely PixelRNN and PixelCNN, respectively. In PixelRNN, two variants of LSTM, called row LSTM and diagonal BILSTM are proposed, where the latter is specifically designed for images. PixelRNN incorporates residual connections to help train deep neural networks with up to 12 layers. In PixelCNN, masked convolutions are used to suit for the shape of the context. Comparing with previous works, PixelRNN and PixelCNN are more dedicated to natural images: they consider pixels as discrete values (e.g., 0, 1, . . . , 255) and predict a multinomial distribution over the discrete values; they deal with color images in RGB color space; they work well on large-scale image dataset ImageNet. In an existing design, Gated PixelCNN is proposed to improve the PixelCNN, and achieves comparable performance with PixelRNN but with much less complexity. In an existing design, PixelCNN++ is proposed with the following improvements upon PixelCNN: a discretized logistic mixture likelihood is used rather than a 256-way multinomial distribution; down-sampling is used to capture structures at multiple resolutions; additional short-cut connections are introduced to speed up training; dropout is adopted for regularization; RGB is combined for one pixel. In an existing design, PixelSNAIL is proposed, in which casual convolutions are combined with self-attention.
Most of the above methods directly model the probability distribution in the pixel domain. Some researchers also attempt to model the probability distribution as a conditional one upon explicit or latent representations. That being said, it may be estimated that
where h is the additional condition and p(x)=p(h)p(x|h), meaning the modeling is split into an unconditional one and a conditional one. The additional condition can be image label information or high-level representations.
Auto-encoder originates from the well-known work proposed by Hinton and Salakhutdinov. The method is trained for dimensionality reduction and consists of two parts: encoding and decoding. The encoding part converts the high-dimension input signal to low-dimension representations, typically with reduced spatial size but a greater number of channels. The decoding part attempts to recover the high-dimension input from the low-dimension representation. Auto-encoder enables automated learning of representations and eliminates the need of hand-crafted features, which is also believed to be one of the most important advantages of neural networks.
It is intuitive to apply auto-encoder network to lossy image compression. It only need to encode the learned latent representation from the well-trained neural networks. However, it is not trivial to adapt auto-encoder to image compression since the original auto-encoder is not optimized for compression thereby not efficient by directly using a trained auto-encoder. In addition, there exist other major challenges: First, the low-dimension representation should be quantized before being encoded, but the quantization is not differentiable, which is required in backpropagation while training the neural networks. Second, the objective under compression scenario is different since both the distortion and the rate need to be take into consideration. Estimating the rate is challenging. Third, a practical image coding scheme needs to support variable rate, scalability, encoding/decoding speed, interoperability. In response to these challenges, a number of researchers have been actively contributing to this area.
The prototype auto-encoder for image compression is in =D+λR, where D is the distortion between x and {circumflex over (x)}, R is the rate calculated or estimated from the quantized representation ŷ, and λ is the Lagrange multiplier. It should be noted that D can be calculated in either pixel domain or perceptual domain. All existing research works follow this prototype and the difference might only be the network structure or loss function.
In terms of network structure, RNNs and CNNs are the most widely used architectures. In the RNNs relevant category, Toderici et al. propose a general framework for variable rate image compression using RNN. They use binary quantization to generate codes and do not consider rate during training. The framework indeed provides a scalable coding functionality, where RNN with convolutional and deconvolution layers is reported to perform decently. Toderici et al. then proposed an improved version by upgrading the encoder with a neural network similar to PixelRNN to compress the binary codes. The performance is reportedly better than JPEG on Kodak image dataset using MS-SSIM evaluation metric. Johnston et al. further improve the RNN-based solution by introducing hidden-state priming. In addition, an SSIM-weighted loss function is also designed, and spatially adaptive bitrates mechanism is enabled. They achieve better results than BPG on Kodak image dataset using MS-SSIM as evaluation metric. Covell et al. support spatially adaptive bitrates by training stop-code tolerant RNNs.
Ballé et al. proposes a general framework for rate-distortion optimized image compression. The use multiary quantization to generate integer codes and consider the rate during training, i.e. the loss is the joint rate-distortion cost, which can be MSE or others. They add random uniform noise to stimulate the quantization during training and use the differential entropy of the noisy codes as a proxy for the rate. They use generalized divisive normalization (GDN) as the network structure, which consists of a linear mapping followed by a nonlinear parametric normalization. The effectiveness of GDN on image coding is verified in an existing design. Ballé et al. then propose an improved version, where they use 3 convolutional layers each followed by a down-sampling layer and a GDN layer as the forward transform. Accordingly, they use 3 layers of inverse GDN each followed by an up-sampling layer and convolution layer to stimulate the inverse transform. In addition, an arithmetic coding method is devised to compress the integer codes. The performance is reportedly better than JPEG and JPEG 2000 on Kodak dataset in terms of MSE. Furthermore, Ballé et al. improve the method by devising a scale hyper-prior into the auto-encoder. They transform the latent representation y with a subnet ha to z=ha(y) and z will be quantized and transmitted as side information. Accordingly, the inverse transform is implemented with a subnet hs attempting to decode from the quantized side information {circumflex over (z)} to the standard deviation of the quantized ŷ, which will be further used during the arithmetic coding of ŷ. On the Kodak image set, their method is slightly worse than BPG in terms of PSNR. D. Minnen et al. further exploit the structures in the residue space by introducing an autoregressive model to estimate both the standard deviation and the mean. In an existing design, Z. Cheng et al. use Gaussian mixture model to further remove redundancy in the residue. The reported performance is on par with VVC on the Kodak image set using PSNR as evaluation metric.
In the transform coding approach to image compression, the encoder subnetwork (section 2.3.2) transforms the image vector x using a parametric analysis transform ga(x, Øg) into a latent representation y, which is then quantized to form ŷ. Because ŷ is discrete-valued, it can be losslessly compressed using entropy coding techniques such as arithmetic coding and transmitted as a sequence of bits.
As evident from the middle left and middle right image of
In
When the hyper encoder and hyper decoder are added to the image compression network, the spatial redundancies of the quantized latent {grave over (y)} are reduced. The rightmost image in
Although the hyper prior model improves the modelling of the probability distribution of the quantized latent ŷ, additional improvement can be obtained by utilizing an autoregressive model that predicts quantized latents from their causal context (Context Model).
The term auto-regressive means that the output of a process is later used as input to it. For example the context model subnetwork generates one sample of a latent, which is later used as input to obtain the next sample.
An existing design utilizes a joint architecture where both hyper prior model subnetwork (hyper encoder and hyper decoder) and a context model subnetwork are utilized. The hyper prior and the context model are combined to learn a probabilistic model over quantized latents ŷ, which is then used for entropy coding. As depicted in
Typically the latent samples are modeled as gaussian distribution or gaussian mixture models (not limited to). In an existing design and according to the
The
The Entropy Parameters subnetwork generates the probability distribution estimations, that are used to encode the quantized latent ŷ. The information that is generated by the Entropy Parameters typically include a mean μ and scale (or variance) σ parameters, that are together used to obtain a gaussian probability distribution. A gaussian distribution of a random variable x is defined as
wherein the parameter μ is the mean or expectation of the distribution (and also its median and mode), while the parameter σ is its standard deviation (or variance, or scale). In order to define a gaussian distribution, the mean and the variance need to be determined. In an existing design the entropy parameters module are used to estimate the mean and the variance values.
The subnetwork hyper decoder generates part of the information that is used by the entropy parameters subnetwork, the other part of the information is generated by the autoregressive module called context module. The context module generates information about the probability distribution of a sample of the quantized latent, using the samples that are already encoded by the arithmetic encoding (AE) module. The quantized latent ŷ is typically a matrix composed of many samples. The samples can be indicated using indices, such as ŷ[i,j,k] or ŷ[i,j] depending on the dimensions of the matrix ŷ. The samples ŷ[i,j] are encoded by AE one by one, typically using a raster scan order. In a raster scan order the rows of a matrix are processed from top to bottom, wherein the samples in a row are processed from left to right. In such a scenario (wherein the raster scan order is used by the AE to encode the samples into bitstream), the context module generates the information pertaining to a sample ŷ[i,j], using the samples encoded before, in raster scan order. The information generated by the context module and the hyper decoder are combined by the entropy parameters module to generate the probability distributions that are used to encode the quantized latent ŷ into bitstream (bits1). Finally the first and the second bitstream are transmitted to the decoder as result of the encoding process.
It is noted that the other names can be used for the modules described above.
In the above description, the all of the elements in
The
After obtaining of {circumflex over (z)}, it is processed by the hyper decoder, whose output is fed to entropy parameters module. The three subnetworks, context, hyper decoder and entropy parameters that are employed in the decoder are identical to the ones in the encoder. Therefore the exact same probability distributions can be obtained in the decoder (as in encoder), which is essential for reconstructing the quantized latent ŷ without any loss. As a result the identical version of the quantized latent ŷ that was obtained in the encoder can be obtained in the decoder.
After the probability distributions (e.g. the mean and variance parameters) are obtained by the entropy parameters subnetwork, the arithmetic decoding module decodes the samples of the quantized latent one by one from the bitstream bits1. From a practical standpoint, autoregressive model (the context model) is inherently serial, and therefore cannot be sped up using techniques such as parallelization. Finally the fully reconstructed quantized latent ŷ is input to the synthesis transform (denoted as decoder in
In the above description, the all of the elements in
Similar to conventional video coding technologies, neural image compression serves as the foundation of intra compression in neural network-based video compression, thus development of neural network-based video compression technology comes later than neural network-based image compression but needs far more efforts to solve the challenges due to its complexity. Starting from 2017, a few researchers have been working on neural network-based video compression schemes. Compared with image compression, video compression needs efficient methods to remove inter-picture redundancy. Inter-picture prediction is then a crucial step in these works. Motion estimation and compensation is widely adopted but is not implemented by trained neural networks until recently.
Studies on neural network-based video compression can be divided into two categories according to the targeted scenarios: random access and the low-latency. In random access case, it requires the decoding can be started from any point of the sequence, typically divides the entire sequence into multiple individual segments and each segment can be decoded independently. In low-latency case, it aims at reducing decoding time thereby usually merely temporally previous frames can be used as reference frames to decode subsequent frames.
Chen et al. are the first to propose a video compression scheme with trained neural networks. They first split the video sequence frames into blocks and each block will choose one from two available modes, either intra coding or inter coding. If intra coding is selected, there is an associated auto-encoder to compress the block. If inter coding is selected, motion estimation and compensation are performed with tradition methods and a trained neural network will be used for residue compression. The outputs of auto-encoders are directly quantized and coded by the Huffman method. Chen et al. propose another neural network-based video coding scheme with PixelMotionCNN. The frames are compressed in the temporal order, and each frame is split into blocks which are compressed in the raster scan order. Each frame will firstly be extrapolated with the preceding two reconstructed frames. When a block is to be compressed, the extrapolated frame along with the context of the current block are fed into the PixelMotionCNN to derive a latent representation. Then the residues are compressed by the variable rate image scheme. This scheme performs on par with H.264.
Lu et al. propose the real-sense end-to-end neural network-based video compression framework, in which all the modules are implemented with neural networks. The scheme accepts current frame and the prior reconstructed frame as inputs and optical flow will be derived with a pre-trained neural network as the motion information. The motion information will be warped with the reference frame followed by a neural network generating the motion compensated frame. The residues and the motion information are compressed with two separate neural auto-encoders. The whole framework is trained with a single rate-distortion loss function. It achieves better performance than H.264.
Rippel et al. propose an advanced neural network-based video compression scheme. It inherits and extends traditional video coding schemes with neural networks with the following major features: 1) using only one auto-encoder to compress motion information and residues; 2) motion compensation with multiple frames and multiple optical flows; 3) an on-line state is learned and propagated through the following frames over time. This scheme achieves better performance in MS-SSIM than HEVC reference software.
J. Lin et al. propose an extended end-to-end neural network-based video compression framework. In this solution, multiple frames are used as references. It is thereby able to provide more accurate prediction of current frame by using multiple reference frames and associated motion information. In addition, motion field prediction is deployed to remove motion redundancy along temporal channel. Postprocessing networks are also introduced in this work to remove reconstruction artifacts from previous processes. The performance is better than H.265 by a noticeable margin in terms of both PSNR and MS-SSIM.
Eirikur et al. propose scale-space flow to replace commonly used optical flow by adding a scale parameter. It is reportedly achieving better performance than H.264.
Z. Hu et al. propose a multi-resolution representation for optical flows. Concretely, the motion estimation network produces multiple optical flows with different resolutions and let the network to learn which one to choose under the loss function. The performance is slightly improved compared with H.265.
Wu et al. propose a neural network-based video compression scheme with frame interpolation. The key frames are first compressed with a neural image compressor and the remaining frames are compressed in a hierarchical order. They perform motion compensation in the perceptual domain, i.e. deriving the feature maps at multiple spatial scales of the original frame and using motion to warp the feature maps, which will be used for the image compressor. The method is reportedly on par with H.264.
Djelouah et al. propose a method for interpolation-based video compression, wherein the interpolation model combines motion information compression and image synthesis, and the same auto-encoder is used for image and residual.
Amirhossein et al. propose a neural network-based video compression method based on variational auto-encoders with a deterministic encoder. Concretely, the model consists of an auto-encoder and an auto-regressive prior. Different from previous methods, this method accepts a group of pictures (GOP) as inputs and incorporates a 3D autoregressive prior by taking into account of the temporal correlation while coding the laten representations. It provides comparative performance as H.265.
Almost all the natural image/video is in digital format. A grayscale digital image can be represented by xϵm×n, where
is the set of values of a pixel, m is the image height and n is the image width. For example,
= {0, 1, 2, . . . , 255} is a common setting and in this case |
|=256=28, thus the pixel can be represented by an 8-bit integer. An uncompressed grayscale digital image has 8 bits-per-pixel (bpp), while compressed bits are definitely less.
A color image is typically represented in multiple channels to record the color information. For example, in the RGB color space an image can be denoted by xϵm×n×3 with three separate channels storing Red, Green and Blue information. Similar to the 8-bit grayscale image, an uncompressed 8-bit RGB image has 24 bpp. Digital images/videos can be represented in different color spaces. The neural network-based video compression schemes are mostly developed in RGB color space while the traditional codecs typically use YUV color space to represent the video sequences. In YUV color space, an image is decomposed into three channels, namely Y, Cb and Cr, where Y is the luminance component and Cb/Cr are the chroma components. The benefits come from that Cb and Cr are typically down sampled to achieve pre-compression since human vision system is less sensitive to chroma components.
A color video sequence is composed of multiple color images, called frames, to record scenes at different timestamps. For example, in the RGB color space, a color video can be denoted by X={x0, x1, . . . , xt, . . . , xT-1} where T is the number of frames in this video sequence, xϵm×n. If m=1080, n=1920, |
|=28, and the video has 50 frames-per-second (fps), then the data rate of this uncompressed video is 1920×1080×8×3×50=2,488,320,000 bits-per-second (bps), about 2.32 Gbps, which needs a lot storage thereby definitely needs to be compressed before transmission over the internet.
Usually the lossless methods can achieve compression ratio of about 1.5 to 3 for natural images, which is clearly below requirement. Therefore, lossy compression is developed to achieve further compression ratio, but at the cost of incurred distortion. The distortion can be measured by calculating the average squared difference between the original image and the reconstructed image, i.e., mean-squared-error (MSE). For a grayscale image, MSE can be calculated with the following equation.
Accordingly, the quality of the reconstructed image compared with the original image can be measured by peak signal-to-noise ratio (PSNR):
where max () is the maximal value in
, e.g., 255 for 8-bit grayscale images. There are other quality evaluation metrics such as structural similarity (SSIM) and multi-scale SSIM (MS-SSIM).
To compare different lossless compression schemes, it is sufficient to compare either the compression ratio given the resulting rate or vice versa. However, to compare different lossy compression methods, it has to take into account both the rate and reconstructed quality. For example, to calculate the relative rates at several different quality levels, and then to average the rates, is a commonly adopted method; the average relative rate is known as Bjontegaard's delta-rate (BD-rate). There are other important aspects to evaluate image/video coding schemes, including encoding/decoding complexity, scalability, robustness, and so on.
The terms quantization and entropy coding can be defined as follows:
Quantization, in mathematics and digital signal processing, is the process of mapping input values from a large set (often a continuous set) to output values in a (countable) smaller set, often with a finite number of elements. Rounding and truncation are typical examples of quantization processes. Quantization is involved to some degree in nearly all digital signal processing, as the process of representing a signal in digital form ordinarily involves rounding. Quantization also forms the core of essentially all lossy compression algorithms.
The equation of quantization can be as follows:
Wherein the ŷ is the quantized value, y is the sample to be quantized and A is the quantization step size. The round function converts the input to a nearest integer value. The floor function is the function that takes as input a real number x, and gives as output the greatest integer less than or equal to x, denoted floor(x). Similarly, the ceiling function maps x to the least integer greater than or equal to x, denoted ceil(x). For example according to the equations above, if the quantization step size is increased, the quantized value ŷ will get smaller and less bits will be needed to encode the quantized sample. In other words the increased quantization step size results in a coarser quantization. On the other hand a smaller Δ (which is usually a positive number) results in a finer quantization and higher number of bits to encode ŷ into the bitstream. An entropy coding is a general lossless data compression method that encodes symbols by using an amount of bits inversely proportional to the probability of the symbols. A term used to denote some compression algorithms, which works based on the probability distribution of source symbols. Huffman coding and arithmetic coding are two typical algorithms of this kind. In order to perform entropy coding, the statistical properties (e.g. probability distribution) of the symbols must be known.
In
Two slightly different implementation alternatives exist for the encoder, based on how the mean value μ is used. In the first implementation the entropy coding module receives the mean value and the other statistics (like the variance σ) as input, and uses them to encode the quantized latent symbols ŷ into bitstream. In the second implementation alternative the mean value is first subtracted from latent symbols y to obtain the residual samples w. Then the residual samples are quantized by the quantization module to obtain quantized residual samples ŵ. The entropy coding module converts the quantized residual samples into bitstream using the statistical information generated by the estimation module. The statistical information might include for example variance σ.
According to the second alternative, the estimation module estimates the mean values μ and other statistical properties (e.g. variance σ). The mean, variance and optionally other statistical information are used by the entropy coder (denoted by EC) to decode the bitstream into quantized residual samples quantized latent samples ŷ. ŷ is finally transformed into reconstructed picture by a synthesis transform. The second alternative decoder implementation is depicted in the bottom flowchart in
In the above a sample of quantized latent is denoted by ŷ. It is noted that the sample is not necessarily a scalar value, it might be a vector o and might contain multiple elements. In the rest of the application a sample can be denoted by ŷ[i, j] or ŷ[: , i, j], or ŷ[c, i, j]. In the latter, the “:” is used to denote that there is a third dimension and is used to stress that the sample has multiple elements.
In the state of the art, the quantization is applied to the latent samples y or the residual samples w using a predefined quantization step size that is constant for all samples. However not all samples are equally important, some samples might have more impact on the quality of the reconstructed image than others. In the state of the art method, it is not possible to apply different quantization precision to different samples of the latent samples or residual samples.
The detailed solutions below should be considered as examples to explain general concepts. These solutions should not be interpreted in a narrow way. Furthermore, these solutions can be combined in any manner.
The quantization step size determined how precisely the y or w is going to the quantized, and in turn how much bits are going to be spent to encode the quantized values ŷ and ŵ.
Some of the individual samples of y or w might be more important and some of the individual samples might be less important. For example, quantization of a subset of samples of y with a smaller quantization step size (finer quantization) might improve the quality of the reconstructed output picture more than others. On the other hand, applying a coarser quantization to another subset of the samples might have a very small impact on the reconstructed output picture. For those samples applying a coarser quantization can help reducing the bitrate required to transmit the quantized samples. Since the goal of the image compression is maximizing the quality of the output image while at the same time reducing the number of bits to be transmitted, it is important to selectively quantize the samples of y or w with different quantization step sizes, which is the goal of the solution.
According to the core of the solution, the samples of y and or w are divided into at least 2 subsets. And different quantization step sizes are applied for quantization of the samples of each subset.
According to the solution:
Wherein the first and second quantization step sizes are denoted as Δ1 and Δ2, and the function f( ) multiplies samples belonging to first subset with Δ1 and samples belonging to second subset with Δ2. The function Q( ) corresponds to the application of quantization.
According to the solution:
ŷ=f(ŷs, m1, m2)
or
ŵ=f(ŵs,m1, m2)
Wherein the first and second multipliers are denoted as m1 and m2, and the function f( ) multiplies samples belonging to first subset with m1 and samples belonging to second subset with m2.
The core solution can be explained with the
The
As described in section 4.2, the solution can be applied to different architectural designs, wherein in one design the quantized latent samples ŷs are included in the bitstream, and in the other design the quantized residual latent samples ŵs are included in the bitstream. For the sake of simplicity in explanation, in this section latent samples (ŷs, ŷ and y) are going to be used to indicate also the residual latent samples (ŵs, ŵ and w).
According to the solution:
Wherein the first and second quantization step sizes are denoted as Δ1 and Δ2, and the function f( ) multiplies samples belonging to first subset with Δ1 and samples belonging to second subset with Δ2.
σs=g(σ, m3, m4)
Wherein the first and second multipliers are denoted as m3 and m4, and the function g( ) multiplies samples belonging to first subset with m3 and samples belonging to second subset with m4.
According to the solution:
σs=g(σ,m3, m4)
Wherein the first and second multipliers are denoted as m3 and m4, and the function g( ) multiplies samples belonging to first subset with m3 and samples belonging to second subset with m4.
ŷ=f(ŷs,m1,m2)
or
ŵ=f(ŵs, m1, m2).
Additionally, the samples of the statistical parameters (μ and/or σ) are also divided into at least 2 subsets. In one example, the samples of the statistical parameters corresponding to the samples of latent y that are included in the first subset are included in a first subset. In other words, if a sample y[i, j] is included in a first subset of latent samples, the statistical parameter σ[i, j] is also included in a first subset of statistical parameter samples.
Afterwards the samples of statistical parameters belonging to first subset are multiplied by a third multiplier, whereas samples of statistical parameters belonging to first subset are multiplied by a fourth multiplier. As a result, the scaled statistical parameters σs are obtained.
Finally, the quantized latent samples ŷ are encoded in a bitstream by an entropy coder using the σs. The determination of which samples belongs to which subset can be based on the value of the statistical parameters. For example the following rule can be applied to determine that a sample belongs to a first subset:
Afterwards the samples of statistical parameters belonging to first subset are multiplied by a third multiplier, whereas samples of statistical parameters belonging to first subset are multiplied by a fourth multiplier. As a result, the scaled statistical parameters σs are obtained.
The quantized latent samples ŷ are decoded from a bitstream by an entropy decoder using the σs. The samples of quantized latent ŷ (or residual latent ŵ) are divided into at least two subsets, and samples of the first subset is multiplied by a first multiplier and the samples of the second subset are multiplied by a second multiplier. One of the multipliers might be equal to 1, whereas the other multiplier is different from 1.
Finally, a synthesis transform is applied to quantized latent ŷ (or residual latent ŵ) to obtain the reconstructed picture.
The determination of which samples belongs to which subset can be based on the value of the statistical parameters. For example the following rule can be applied to determine that a sample belongs to a first subset:
In the
If the statistical properties estimated by the estimation module are not a good estimation of the latent samples, the performance of the entropy coding process would deteriorate. Therefore, compensating for the change in latent samples (by multiplying corresponding statistical parameters) increases the performance of the estimation module.
According to the solution the grouping process (into subsets) of the samples are based on the estimated statistical parameters. This is particularly useful since the statistical parameters indicate how “unclear” the value of a sample is. For example, the
The goal of the image compression is maximizing the quality of the output image while at the same time reducing the number of bits to be transmitted. Not all parts of the image are equally important, the solution allows adjusting the precision of compression in selected parts of the image by adjusting the quantization parameter selectively for the samples of the transformed image (latent representation). The solution additionally allows grouping the samples according to their statistical properties, which allows representing samples that are “unclear” with more precision (therefore with more bits) and vice versa. As a result, the overall quality of the reconstructed image is improved without increasing the bitrate a lot.
The step of obtaining a statistical parameter σs using an estimation module comprises:
More details of the embodiments of the present disclosure will be described below which are related to neural network-based visual data coding. As used herein, the term “visual data” may refer to an image, a picture in a video, or any other visual data suitable to be coded.
As discussed above, in the existing design, the quantizing process is applied to samples of the latent representation y or the residual latent representation w of the visual data by using a predefined quantization step size that is constant for all samples. However not all samples are equally important, some samples might have more impact on the quality of the reconstructed visual data than others. In the existing design, it is not possible to apply different quantization precision to different samples of the latent representation y or the residual latent representation w of the visual data.
To solve the above problems and some other problems not mentioned, visual data processing solutions as described below are disclosed.
As shown in
The first statistical value may be generated by a second neural network 1522. For example, the second neural network 1522 may be referred to as an estimation model. It should be noted that a model may also be referred to as a module in the present application. That is, the second neural network may also be referred to as a first module, and the estimation model may also be referred to as an estimation module. The input of the second neural network 1522, may comprise the latent representation, and the output of the second neural network 1522 comprise at least one statistical value. In one example, the second neural network 1522 may comprise a neural network-based subnetwork. In another example, the second neural network 1522 may comprise a first subnetwork for generating the first statistical value and a second subnetwork for generating a second statistical value. By way of example rather than limitation, the second statistical value may be a variance (denoted as σ in
At the scaling block 1524, samples of the second statistical value (also referred to as “third samples” hereinafter) may be divided into a plurality of sets to obtain a plurality of sets of third samples. In one example, this dividing process may be performed at a sample-level. In other words, the determination of the division is made for each of the third samples. Alternatively, the dividing process may be performed at a block-level. More specifically, the third samples of the second statistical value may be grouped into a plurality of blocks. Each of the plurality of blocks may have a predetermined size of N by M, such as 8×8. This predetermined size may be indicated in the bitstream. The third samples may be divided based on the plurality of blocks. In other words, the determination of the division is made for each of the plurality of blocks, and samples in the same block are divided into the same set.
In some embodiments, a single third sample of the second statistical value may be divided into one of the plurality of sets based on a sample of the first statistical value corresponding to the third sample and the single third sample itself. By way of example rather than limitation, a value of the sample of the first statistical value and a value of the single third sample itself may be compared with a first threshold and a second threshold, respectively. If the value of the sample of the first statistical value is smaller than the first threshold and the value of the single third sample is smaller than the second threshold, this third sample may be divided into a first set. Otherwise, this third sample may be divided into a second set different from the first set.
In another example, an index of the single third sample may be compared with a third threshold during the division process. If the index is smaller than the third threshold, this third sample may be divided into a first set. Otherwise, this third sample may be divided into a second set different from the first set. An index of a sample may indicate a channel number of the sample, a feature map identifier of the sample, a spatial coordinate of the sample, or the like. The above-mentioned comparison may be performed for each of the third samples, so as to dividing the third samples.
It should be understood that a single third sample may be divided based on any other suitable parameters or values, e.g., at least one sample of the at least one statistical value corresponding to the single third sample, a probability distribution determined based on the at least one sample, a probability value determined based on the at least one sample, a threshold, or a function of the at least one sample. This is described in detail at the above section 4.3. In some embodiments, the threshold used may be indicated in the bitstream or may be predetermined.
In some alternative embodiments, if the dividing process is performed at a block-level, third samples in a block may be divided into one of the plurality sets based on a comparison between a threshold and each of the third samples, or each of probability values determined based on the respective third sample. Alternatively, third samples in a block may be divided into one of the plurality sets based on a comparison between a threshold and an average of the third samples. In a further example, third samples in a block may be divided into one of the plurality sets based on a comparison between a threshold and a maximum or a minimum of the third samples. The above-mentioned comparison may be performed for each of the plurality of blocks, so as to dividing the third samples.
It should be understood that third samples in a block may be divided based on any other suitable parameters or values, e.g., a set of samples of the at least one reference statistical value corresponding to the third samples in the block, probability distributions determined based on the set of samples, probability values determined based on the set of samples, a threshold, a function of the set of samples, indices of the third samples in the third block, or a metric determined based on one of: the set of samples, the probability distributions, the probability values or the indices. The metric may be an average, a maximum or a minimum.
At the scaling block 1524, the obtained plurality of sets of third samples may be further adjusted with different parameters to generate an adjusted second statistical value, such as an adjusted variance (denoted as σs in
As shown in
At the scaling block 1516, samples of the residual latent representation (also referred to as “second samples” hereinafter) may be divided into a plurality of sets to obtain a plurality of sets of second samples. The second samples of the residual latent representation may also be divided at a sample-level or a block-level in a manner similar to the dividing process described with regard to the scaling block 1524. It should be noted that in addition to the reference statistical value (such as the mean μ and/or the variance σ), the second samples of the residual latent representation may also be divided based on the target statistical value (such as the adjusted variance σs) in a manner similar to the reference statistical value. In one non-limiting example, the second samples may be divided in the same manner as the third samples. In other words, if a third sample is divided into a first set of third samples, a second sample corresponding to the third sample may also be divided into a first set of second samples.
Furthermore, the obtained plurality of sets of second samples may be further adjusted with different parameters to generate a plurality of sets of adjusted second samples (denoted as ws in
At quantizing block 1518, the plurality of sets of adjusted second samples may be quantized to obtain samples of the quantized residual latent representation (denoted as ŵs in
At entropy encoder 1520, an entropy encoding process may be performed on the output of the quantizing block 1518 (e.g., the quantized samples of the adjusted residual latent representation) based on the target statistical value (e.g., the adjusted variance σs) to generate at least a part of the bitstream. The entropy encoding process performed by the entropy encoder 1520 may be an arithmetic encoding process, a Huffman encoding process, or the like. Additionally or alternatively, either the reference statistical value or the target statistical value may also be encoded into the bitstream.
In view of the above, samples of a residual latent representation of the visual data are divided into a plurality of sets, and the plurality of sets of samples are adjusted with different parameters. Thereby, it is possible to quantize the samples with different parameters, so as to maximizing the quality of the reconstructed visual data while at the same time reducing the number of bits to be transmitted. As such, the proposed method can advantageously improve the coding quality and coding efficiency.
Although an example visual data encoding process is described above with respect to
As shown in
At scaling block 1620, samples of the statistical value may be divided and adjusted in a manner similar to the dividing and adjusting process described with regard to the scaling block 1524, and thus this will not be described in detail for conciseness.
At the entropy decoder 1618, an entropy decoding process may be performed on the bitstream based on the at least on target statistical value, so as to obtain the samples of the quantized residual latent representation (denoted as ŵs in
At scaling block 1616, samples of the quantized residual latent representation (also referred to as “first samples” hereinafter) may be divided into a plurality of sets to obtain a plurality of sets of first samples. The first samples of the quantized residual latent representation may also be divided at a sample-level or a block-level in a manner similar to the dividing process described with regard to the scaling block 1524. It should be noted that in addition to the reference statistical value (such as the mean μ and/or the variance σ), the first samples of the residual latent representation may also be divided based on the target statistical value (such as the adjusted variance σs) in a manner similar to the reference statistical value. In one non-limiting example, the first samples may be divided in the same manner as the third samples. In other words, if a third sample is divided into a first set of third samples, a first sample corresponding to the third sample may also be divided into a first set of first samples.
Furthermore, the obtained plurality of sets of first samples may be further adjusted with different parameters to generate a plurality of sets of adjusted first samples (denoted as ŵ in
It should be understood that in addition to scaling, the plurality of sets of first samples may be adjusted in any suitable manner, such as by adding different parameters. The scope of the present disclosure is not limited in this respect.
At block 1614, the quantized latent representation (denoted as ŷ in
In view of the above, samples of a quantized residual latent representation of the visual data are divided into a plurality of sets, and the plurality of sets of samples are adjusted with different parameters. Thereby, it is possible to quantize the samples with different parameters, so as to maximizing the quality of the reconstructed visual data while at the same time reducing the number of bits to be transmitted. As such, the proposed method can advantageously improve the coding quality and coding efficiency.
Although an example visual data decoding process is described above with respect to
It should be understood that the above illustrations and/or examples are described merely for purpose of description. The scope of the present disclosure is not limited in this respect. The embodiments of the present disclosure should be considered as examples to explain the general concepts and should not be interpreted in a narrow way. Furthermore, these embodiments can be applied individually or combined in any manner.
As shown in
At block 1704, a plurality of sets of first samples of the first representation with different parameters are adjusted. By way of example rather than limitation, a first set of first samples among the plurality of sets of first samples may be adjusted by scaling the first set of first samples with a first parameter. A second set of first samples among the plurality of sets of first samples may be adjusted by scaling the second set of first samples with a second parameter different from the first parameter. It should be understood that in addition to scaling, the plurality of sets of first samples may be adjusted in any suitable manner, such as by adding different parameters. The scope of the present disclosure is not limited in this respect.
At block 1706, the conversion is performed based on the plurality of sets of adjusted first samples. In one example, the conversion may include encoding the visual data into the bitstream. Alternatively or additionally, the conversion may include decoding the visual data from the bitstream. It should be understood that the above illustrations are described merely for purpose of description. The scope of the present disclosure is not limited in this respect.
In view of the foregoing, samples of a first representation (such as quantized latent representation or quantized residual latent representation) of the visual data are divided into a plurality of sets, and the plurality of sets of first samples are adjusted with different parameters. Thereby, the proposed method makes it possible to quantize the samples with different parameters, so as to maximizing the quality of the reconstructed visual data while at the same time reducing the number of bits to be transmitted. As such, the proposed method can advantageously improve the coding quality and coding efficiency.
In some embodiments, at 1704, the plurality of sets of first samples may be determined from the first representation based on at least one of the following: indices of samples of the first representation, or at least one reference statistical value associated with the second representation. By way of example rather than limitation, the least one reference statistical value may be generated by using a second neural network. It should be understood that the least one reference statistical value may be generated in any other suitable manner, such as by using a machine learning-based model. The scope of the present disclosure is not limited in this respect.
In some embodiments, for determining the plurality of sets of first samples from the first representation, a single first sample may be divided into one of the plurality of sets of first samples based on at least one of the following: at least one sample of the at least one reference statistical value, the at least one sample corresponding to the single first sample, a probability distribution determined based on the at least one sample, a probability value determined based on the at least one sample, a threshold, a function of the at least one sample, or an index of the single first sample.
In some alternative embodiments, for determining the plurality of sets of first samples from the first representation: first samples in a first block of the first representation may be divided into one of the plurality of sets of first samples based on at least one of the following: a set of samples of the at least one reference statistical value, the set of samples corresponding to the first samples in the first block, probability distributions determined based on the set of samples, probability values determined based on the set of samples, a threshold, a function of the set of samples, indices of the first samples in the first block, or a metric determined based on one of: the set of samples, the probability distributions, the probability values or the indices.
In some alternative embodiments, at least one target statistical value may be generated based on the at least one reference statistical value. The plurality of sets of first samples may be determined from the first representation based on the at least one target statistical value. In one example, for determining the plurality of sets of first samples from the first representation based on the at least one target statistical value, a single first sample may be divided into one of the plurality of sets of first samples based on at least one of the following: at least one sample of the at least one target statistical value, the at least one sample corresponding to the single first sample, a probability distribution determined based on the at least one sample, a probability value determined based on the at least one sample, or a function of the at least one sample.
Alternatively, for determining the plurality of sets of first samples from the first representation based on the at least one target statistical value, first samples in a first block of the first representation may be divided into one of the plurality of sets of first samples based on at least one of the following: a set of samples of the at least one target statistical value, the set of samples corresponding to the first samples in the first block, probability distributions determined based on the set of samples, probability values determined based on the set of samples, a function of the set of samples, or a metric determined based on one of: the set of samples, the probability distributions, or the probability values.
In some embodiments, the first representation may be obtained by performing an entropy decoding process on the bitstream based on the at least one reference statistical value. Alternatively, the first representation may be obtained by performing an entropy decoding process on the bitstream based on the at least one target statistical value.
In some embodiments, the second representation may be a latent representation of the visual data. At 1706, the visual data may be reconstructed by performing a synthesis transform on the plurality of sets of adjusted first samples. For example, the synthesis transform may be performed by using a neural network.
In some embodiments, the second representation may be a residual latent representation of the visual data. At 1706, a quantized latent representation of the visual data may be generated based on the plurality of sets of adjusted first samples and a first reference statistical value of the at least one reference statistical value. The visual data may be reconstructed by performing a synthesis transform on the quantized latent representation.
In some embodiments, at 1706, the bitstream may be generated by performing an entropy encoding process on the plurality of sets of first samples based on the at least one reference statistical value. Alternatively, the bitstream may be generated by performing an entropy encoding process on the plurality of sets of first samples based on the at least one target statistical value.
In some embodiments, at 1702, a plurality of sets of second samples may be determined from the second representation based on at least one of the following: indices of samples of the second representation, or at least one reference statistical value associated with the second representation. The plurality of sets of second samples may be adjusted with different parameters. Moreover, the plurality of sets of adjusted second samples may be quantized to obtain the plurality of sets of first samples.
In some embodiments, for determining the plurality of sets of second samples from the second representation, a single second sample may be divided into one of the plurality of sets of second samples based on at least one of the following: at least one sample of the at least one reference statistical value, the at least one sample corresponding to the single second sample, a probability distribution determined based on the at least one sample, a probability value determined based on the at least one sample, a threshold, a function of the at least one sample, or an index of the single second sample.
In some embodiments, for determining the plurality of sets of second samples from the second representation, second samples in a second block of the second representation may be divided into one of the plurality of sets of second samples based on at least one of the following: a set of samples of the at least one reference statistical value, the set of samples corresponding to the second samples in the second block, probability distributions determined based on the set of samples, probability values determined based on the set of samples, a threshold, a function of the set of samples, indices of the second samples in the second block, or a metric determined based on one of: the set of samples, the probability distributions, the probability values or the indices.
In some embodiments, a first set of second samples among the plurality of sets of second samples may be adjusted by scaling the first set of second samples with a third parameter. Furthermore, a second set of second samples among the plurality of sets of second samples may be adjusted by scaling the second set of second samples with a fourth parameter different from the third parameter.
In some embodiments, the plurality of sets of adjusted second samples may be quantized by using a rounding function. Alternatively, the plurality of sets of adjusted second samples may be quantized by using a floor function. In some further embodiments, the plurality of sets of adjusted second samples may be quantized by using a ceiling function. It should be understood that the plurality of sets of adjusted second samples may also be quantized in any other suitable manner. The scope of the present disclosure is not limited in this respect.
In some embodiments, the at least one target statistical value may comprise a first target statistical value, the at least one reference statistical value may comprise a second reference statistical value corresponding to the first target statistical value. For generating the at least one target statistical value, a plurality of sets of third samples may be determined from the second reference statistical value based on at least one of the following: indices of samples of the second reference statistical value, or the at least one reference statistical value. Moreover, the plurality of sets of third samples may be adjusted with different parameters to obtain the first target statistical parameter.
In some embodiments, for determining the plurality of sets of third samples from the second reference statistical value, a single third sample may be divided into one of the plurality of sets of third samples based on at least one of the following: at least one sample of the at least one reference statistical value, the at least one sample corresponding to the single third sample, a probability distribution determined based on the at least one sample, a probability value determined based on the at least one sample, a threshold, a function of the at least one sample, or an index of the single third sample.
In some alternative embodiments, for determining the plurality of sets of third samples from the second reference statistical value, third samples in a third block of the second reference statistical value may be divided into one of the plurality of sets of third samples based on at least one of the following: a set of samples of the at least one reference statistical value, the set of samples corresponding to the third samples in the third block, probability distributions determined based on the set of samples, probability values determined based on the set of samples, a threshold, a function of the set of samples, indices of the third samples in the third block, or a metric determined based on one of: the set of samples, the probability distributions, the probability values or the indices. By way of example rather than limitation, the threshold may be indicated in the bitstream. The metric may be an average a minimum or a maximum. An index of a sample may indicate one of the following: a channel number of the sample, a feature map identifier of the sample, or a spatial coordinate of the sample.
In some embodiments, the plurality of sets of first samples may comprise a first set of first samples, and third samples corresponding to the first set of first samples may be divided into a single set.
In some embodiments, a first set of third samples among the plurality of sets of third samples may be adjusted by scaling the first set of third samples with a fifth parameter. Furthermore, a second set of third samples among the plurality of sets of third samples may be adjusted by scaling the second set of third samples with a sixth parameter different from the fifth parameter.
In some embodiments, the second neural network may be an estimation model. In one example, the second neural network may comprise a neural network-based subnetwork. For example, an input of the second neural network may comprise the bitstream. Additionally or alternatively, the second neural network may comprise a first subnetwork for generating a first reference statistical value and a second subnetwork for generating a reference second statistical value. By way of example rather than limitation, the first reference statistical value may be a mean, and the second reference statistical value may be a variance. The first subnetwork may be a hyper decoder subnetwork, and the second subnetwork may be a hyper scale decoder subnetwork.
In some embodiments, at least one of the following parameters may be a scalar number: the first parameter, the second parameter, the third parameter, the fourth parameter, the fifth parameter, or the sixth parameter. Alternatively, at least one of the following parameters may be a vector: the first parameter, the second parameter, the third parameter, the fourth parameter, the fifth parameter, or the sixth parameter.
In some embodiments, at least one of the following parameters may be indicated in the bitstream: the first parameter, the second parameter, the third parameter, the fourth parameter, the fifth parameter, or the sixth parameter. Alternatively, at least one indication of at least one of the following parameters may be indicated in the bitstream: the first parameter, the second parameter, the third parameter, the fourth parameter, the fifth parameter, or the sixth parameter. In some further embodiments, at least one of the following parameters may be obtained based on a list and an index value: the first parameter, the second parameter, the third parameter, the fourth parameter, the fifth parameter, or the sixth parameter.
In some embodiments, the first parameter may be determined based on the third parameter. By way of example rather than limitation, the first parameter may be determined as a reciprocal of the third parameter. For example, one of the first parameter and the second parameter may be 1. Additionally or alternatively, one of the third parameter and the fourth parameter may be 1. Additionally or alternatively, one of the fifth parameter and the sixth parameter may be 1.
In some embodiments, the fifth parameter may be the same as one of the first parameter or the second parameter, and the sixth parameter may be the same as another one of the first parameter or the second parameter. In such a case, the plurality of sets of first samples and the plurality of sets of third samples may be adjusted in a same manner.
In some embodiments, the second representation may be a latent representation of the visual data. The second representation may be obtained by performing an analysis transform on the visual data by using the first neural network.
In some embodiments, the second representation may be a residual latent representation of the visual data. A latent representation of the visual data may be generated by performing an analysis transform on the visual data by using the first neural network. The second representation may be generated based on the latent representation and a first reference statistical value of the at least one reference statistical value. In some embodiments, the first reference statistical value may be a mean.
In some embodiments, the at least one reference statistical value may comprise at least one of a mean or a variance of a probability distribution. By way of example rather than limitation, the probability distribution may be a gaussian probability distribution or Laplace probability distribution.
In some embodiments, at least one of the following may be predetermined or indicated in the bitstream: a size of the first block, a size of the second block, or a size of the third block.
In some embodiments, at least one of the following may be indicated the bitstream: information on whether to apply the method, or information on how to apply the method. Alternatively, at least one of the following may be dependent on a color format and/or a color component of the visual data: information on whether to apply the method, or information on how to apply the method.
In some embodiments, a value included in the bitstream may be coded at one of the following: a sequence level, a picture level, a slice level, or a block level. Additionally or alternatively, a value included in the bitstream may be binarized before may be coded. Additionally or alternatively, a value included in the bitstream may be coded with at least one arithmetic coding context.
According to further embodiments of the present disclosure, a non-transitory computer-readable recording medium is provided. The non-transitory computer-readable recording medium stores a bitstream of visual data which is generated by a method performed by an apparatus for visual data processing. The method comprises: obtaining a first representation of the visual data, the first representation being obtained by quantizing a second representation of the visual data, the second representation being generated based on applying a first neural network to the visual data; adjusting a plurality of sets of first samples of the first representation with different parameters; and generating the bitstream based on the plurality of sets of adjusted first samples.
According to still further embodiments of the present disclosure, a method for storing a bitstream of visual data is provided. The method comprises: obtaining a first representation of the visual data, the first representation being obtained by quantizing a second representation of the visual data, the second representation being generated based on applying a first neural network to the visual data; adjusting a plurality of sets of first samples of the first representation with different parameters; generating the bitstream based on the plurality of sets of adjusted first samples; and storing the bitstream in a non-transitory computer-readable recording medium.
Implementations of the present disclosure can be described in view of the following clauses, the features of which can be combined in any reasonable manner.
Clause 1. A method for visual data processing, comprising: obtaining, for a conversion between visual data and a bitstream of the visual data, a first representation of the visual data, the first representation being obtained by quantizing a second representation of the visual data, the second representation being generated based on applying a first neural network to the visual data; adjusting a plurality of sets of first samples of the first representation with different parameters; and performing the conversion based on the plurality of sets of adjusted first samples.
Clause 2. The method of clause 1, wherein adjusting the plurality of sets of first samples comprises: determining the plurality of sets of first samples from the first representation based on at least one of the following: indices of samples of the first representation, or at least one reference statistical value associated with the second representation.
Clause 3. The method of clause 2, wherein the least one reference statistical value is generated by using a second neural network.
Clause 4. The method of any of clauses 2-3, wherein determining the plurality of sets of first samples from the first representation comprises: dividing a single first sample into one of the plurality of sets of first samples based on at least one of the following: at least one sample of the at least one reference statistical value, the at least one sample corresponding to the single first sample, a probability distribution determined based on the at least one sample, a probability value determined based on the at least one sample, a threshold, a function of the at least one sample, or an index of the single first sample.
Clause 5. The method of any of clauses 2-3, wherein determining the plurality of sets of first samples from the first representation comprises: dividing first samples in a first block of the first representation into one of the plurality of sets of first samples based on at least one of the following: a set of samples of the at least one reference statistical value, the set of samples corresponding to the first samples in the first block, probability distributions determined based on the set of samples, probability values determined based on the set of samples, a threshold, a function of the set of samples, indices of the first samples in the first block, or a metric determined based on one of: the set of samples, the probability distributions, the probability values or the indices.
Clause 6. The method of any of clauses 2-3, wherein determining the plurality of sets of first samples from the first representation comprises: generating at least one target statistical value based on the at least one reference statistical value; and determining the plurality of sets of first samples from the first representation based on the at least one target statistical value.
Clause 7. The method of clause 6, wherein determining the plurality of sets of first samples from the first representation based on the at least one target statistical value comprises: dividing a single first sample into one of the plurality of sets of first samples based on at least one of the following: at least one sample of the at least one target statistical value, the at least one sample corresponding to the single first sample, a probability distribution determined based on the at least one sample, a probability value determined based on the at least one sample, or a function of the at least one sample.
Clause 8. The method of clause 6, wherein determining the plurality of sets of first samples from the first representation based on the at least one target statistical value comprises: dividing first samples in a first block of the first representation into one of the plurality of sets of first samples based on at least one of the following: a set of samples of the at least one target statistical value, the set of samples corresponding to the first samples in the first block, probability distributions determined based on the set of samples, probability values determined based on the set of samples, a function of the set of samples, or a metric determined based on one of: the set of samples, the probability distributions, or the probability values.
Clause 9. The method of any of clauses 1-8, wherein adjusting the plurality of sets of first samples comprises: adjusting a first set of first samples among the plurality of sets of first samples by scaling the first set of first samples with a first parameter; and adjusting a second set of first samples among the plurality of sets of first samples by scaling the second set of first samples with a second parameter different from the first parameter.
Clause 10. The method of any of clauses 6-9, wherein the first representation is obtained by performing an entropy decoding process on the bitstream based on the at least one reference statistical value or the at least one target statistical value.
Clause 11. The method of any of clauses 1-10, wherein the second representation comprises a latent representation of the visual data or a residual latent presentation of the visual data.
Clause 12. The method of any of clauses 1-11, wherein the second representation is a latent representation of the visual data, and performing the conversion comprises: reconstructing the visual data by performing a synthesis transform on the plurality of sets of adjusted first samples.
Clause 13. The method of any of clauses 2-11, wherein the second representation is a residual latent representation of the visual data, and performing the conversion comprises: generating a quantized latent representation of the visual data based on the plurality of sets of adjusted first samples and a first reference statistical value of the at least one reference statistical value; and reconstructing the visual data by performing a synthesis transform on the quantized latent representation.
Clause 14. The method of any of clauses 6-11, wherein performing the conversion comprises: generating the bitstream by performing an entropy encoding process on the plurality of sets of first samples based on the at least one reference statistical value or the at least one target statistical value.
Clause 15. The method of any of clauses 1-14, wherein obtaining the plurality of sets of first samples comprises: determining a plurality of sets of second samples from the second representation based on at least one of the following: indices of samples of the second representation, or at least one reference statistical value associated with the second representation; adjusting the plurality of sets of second samples with different parameters; and quantizing the plurality of sets of adjusted second samples to obtain the plurality of sets of first samples.
Clause 16. The method of clause 15, wherein determining the plurality of sets of second samples from the second representation comprises: dividing a single second sample into one of the plurality of sets of second samples based on at least one of the following: at least one sample of the at least one reference statistical value, the at least one sample corresponding to the single second sample, a probability distribution determined based on the at least one sample, a probability value determined based on the at least one sample, a threshold, a function of the at least one sample, or an index of the single second sample.
Clause 17. The method of clause 15, wherein determining the plurality of sets of second samples from the second representation comprises: dividing second samples in a second block of the second representation into one of the plurality of sets of second samples based on at least one of the following: a set of samples of the at least one reference statistical value, the set of samples corresponding to the second samples in the second block, probability distributions determined based on the set of samples, probability values determined based on the set of samples, a threshold, a function of the set of samples, indices of the second samples in the second block, or a metric determined based on one of: the set of samples, the probability distributions, the probability values or the indices.
Clause 18. The method of any of clauses 15-17, wherein adjusting the plurality of sets of second samples comprises: adjusting a first set of second samples among the plurality of sets of second samples by scaling the first set of second samples with a third parameter; and adjusting a second set of second samples among the plurality of sets of second samples by scaling the second set of second samples with a fourth parameter different from the third parameter.
Clause 19. The method of any of clauses 15-18, wherein the plurality of sets of adjusted second samples are quantized by using one of the following: a rounding function, a floor function, or a ceiling function.
Clause 20. The method of any of clauses 6-19, wherein the at least one target statistical value comprises a first target statistical value, the at least one reference statistical value comprises a second reference statistical value corresponding to the first target statistical value, and generating the at least one target statistical value comprises: determining a plurality of sets of third samples from the second reference statistical value based on at least one of the following: indices of samples of the second reference statistical value, or the at least one reference statistical value; and adjusting the plurality of sets of third samples with different parameters to obtain the first target statistical parameter.
Clause 21. The method of clause 20, wherein determining the plurality of sets of third samples from the second reference statistical value comprises: dividing a single third sample into one of the plurality of sets of third samples based on at least one of the following: at least one sample of the at least one reference statistical value, the at least one sample corresponding to the single third sample, a probability distribution determined based on the at least one sample, a probability value determined based on the at least one sample, a threshold, a function of the at least one sample, or an index of the single third sample.
Clause 22. The method of clause 20, wherein determining the plurality of sets of third samples from the second reference statistical value comprises: dividing third samples in a third block of the second reference statistical value into one of the plurality of sets of third samples based on at least one of the following: a set of samples of the at least one reference statistical value, the set of samples corresponding to the third samples in the third block, probability distributions determined based on the set of samples, probability values determined based on the set of samples, a threshold, a function of the set of samples, indices of the third samples in the third block, or a metric determined based on one of: the set of samples, the probability distributions, the probability values or the indices.
Clause 23. The method of any of clauses 20-22, wherein the plurality of sets of first samples comprises a first set of first samples, and third samples corresponding to the first set of first samples are divided into a single set.
Clause 24. The method of any of clauses 20-23, wherein adjusting the plurality of sets of third samples comprises: adjusting a first set of third samples among the plurality of sets of third samples by scaling the first set of third samples with a fifth parameter; and adjusting a second set of third samples among the plurality of sets of third samples by scaling the second set of third samples with a sixth parameter different from the fifth parameter.
Clause 25. The method of clause 20-24, wherein the threshold is indicated in the bitstream.
Clause 26. The method of any of clauses 5-25, wherein the metric is an average a minimum or a maximum.
Clause 27. The method of any of clauses 5-26, wherein an index of a sample indicates one of the following: a channel number of the sample, a feature map identifier of the sample, or a spatial coordinate of the sample.
Clause 28. The method of any of clauses 3-27, wherein the second neural network is an estimation model.
Clause 29. The method of any of clauses 3-28, wherein the second neural network comprises a neural network-based subnetwork.
Clause 30. The method of any of clauses 3-29, wherein an input of the second neural network comprises the bitstream.
Clause 31. The method of any of clauses 3-28, wherein the second neural network comprises a first subnetwork for generating a first reference statistical value and a second subnetwork for generating a reference second statistical value.
Clause 32. The method of clause 31, wherein the first reference statistical value is a mean, and the second reference statistical value is a variance, the first subnetwork is a hyper decoder subnetwork, and the second subnetwork is a hyper scale decoder subnetwork.
Clause 33. The method of any clauses 9-32, wherein at least one of the following parameters is a scalar number: the first parameter, the second parameter, the third parameter, the fourth parameter, the fifth parameter, or the sixth parameter.
Clause 34. The method of any clauses 9-32, wherein at least one of the following parameters is a vector: the first parameter, the second parameter, the third parameter, the fourth parameter, the fifth parameter, or the sixth parameter.
Clause 35. The method of any clauses 9-34, wherein at least one of the following parameters is indicated in the bitstream: the first parameter, the second parameter, the third parameter, the fourth parameter, the fifth parameter, or the sixth parameter.
Clause 36. The method of any clauses 9-34, wherein at least one indication of at least one of the following parameters is indicated in the bitstream: the first parameter, the second parameter, the third parameter, the fourth parameter, the fifth parameter, or the sixth parameter.
Clause 37. The method of any clauses 9-34, wherein at least one of the following parameters is obtained based on a list and an index value: the first parameter, the second parameter, the third parameter, the fourth parameter, the fifth parameter, or the sixth parameter.
Clause 38. The method of any clauses 9-37, wherein the first parameter is determined based on the third parameter.
Clause 39. The method of clause 38, wherein the first parameter is determined as a reciprocal of the third parameter.
Clause 40. The method of any of clauses 9-39, wherein one of the first parameter and the second parameter is 1.
Clause 41. The method of any of clauses 18-40, wherein one of the third parameter and the fourth parameter is 1.
Clause 42. The method of any of clauses 24-41, wherein one of the fifth parameter and the sixth parameter is 1.
Clause 43. The method of any of clauses 24-42, wherein the fifth parameter is the same as one of the first parameter or the second parameter, and the sixth parameter is the same as another one of the first parameter or the second parameter.
Clause 44. The method of any of clauses 1-43, wherein the second representation is a latent representation of the visual data, and the second representation is obtained by performing an analysis transform on the visual data by using the first neural network.
Clause 45. The method of any of clauses 2-44, wherein the second representation is a residual latent representation of the visual data, and the second representation is obtained by: generating a latent representation of the visual data by performing an analysis transform on the visual data by using the first neural network; and generating the second representation based on the latent representation and a first reference statistical value of the at least one reference statistical value.
Clause 46. The method of any of clauses 13-45, wherein the first reference statistical value is a mean.
Clause 47. The method of any of clauses 2-13, wherein the at least one reference statistical value comprises at least one of a mean or a variance of a probability distribution.
Clause 48. The method of clause 47, wherein the probability distribution is a gaussian probability distribution.
Clause 49. The method of any of clauses 5-48, wherein at least one of the following is predetermined or indicated in the bitstream: a size of the first block, a size of the second block, or a size of the third block.
Clause 50. The method of any of clauses 1-49, wherein at least one of the following is indicated the bitstream: information on whether to apply the method, or information on how to apply the method.
Clause 51. The method of any of clauses 1-49, wherein at least one of the following is dependent on a color format and/or a color component of the visual data: information on whether to apply the method, or information on how to apply the method.
Clause 52. The method of any of clauses 1-51, wherein a value included in the bitstream is coded at one of the following: a sequence level, a picture level, a slice level, or a block level.
Clause 53. The method of any of clauses 1-52, wherein a value included in the bitstream is binarized before being coded.
Clause 54. The method of any of clauses 1-53, wherein a value included in the bitstream is coded with at least one arithmetic coding context.
Clause 55. The method of any of clauses 1-54, wherein the visual data comprise a picture of a video or an image.
Clause 56. The method of any of clauses 1-55, wherein the conversion includes encoding the visual data into the bitstream.
Clause 57. The method of any of clauses 1-55, wherein the conversion includes decoding the visual data from the bitstream.
Clause 58. An apparatus for visual data processing comprising a processor and a non-transitory memory with instructions thereon, wherein the instructions upon execution by the processor, cause the processor to perform a method in accordance with any of clauses 1-57.
Clause 59. A non-transitory computer-readable storage medium storing instructions that cause a processor to perform a method in accordance with any of clauses 1-57.
Clause 60. A non-transitory computer-readable recording medium storing a bitstream of visual data which is generated by a method performed by an apparatus for visual data processing, wherein the method comprises: obtaining a first representation of the visual data, the first representation being obtained by quantizing a second representation of the visual data, the second representation being generated based on applying a first neural network to the visual data; adjusting a plurality of sets of first samples of the first representation with different parameters; and generating the bitstream based on the plurality of sets of adjusted first samples.
Clause 61. A method for storing a bitstream of visual data, comprising: obtaining a first representation of the visual data, the first representation being obtained by quantizing a second representation of the visual data, the second representation being generated based on applying a first neural network to the visual data; adjusting a plurality of sets of first samples of the first representation with different parameters; generating the bitstream based on the plurality of sets of adjusted first samples; and storing the bitstream in a non-transitory computer-readable recording medium.
It would be appreciated that the computing device 1800 shown in
As shown in
In some embodiments, the computing device 1800 may be implemented as any user terminal or server terminal having the computing capability. The server terminal may be a server, a large-scale computing device or the like that is provided by a service provider. The user terminal may for example be any type of mobile terminal, fixed terminal, or portable terminal, including a mobile phone, station, unit, device, multimedia computer, multimedia tablet, Internet node, communicator, desktop computer, laptop computer, notebook computer, netbook computer, tablet computer, personal communication system (PCS) device, personal navigation device, personal digital assistant (PDA), audio/video player, digital camera/video camera, positioning device, television receiver, radio broadcast receiver, E-book device, gaming device, or any combination thereof, including the accessories and peripherals of these devices, or any combination thereof. It would be contemplated that the computing device 1800 can support any type of interface to a user (such as “wearable” circuitry and the like).
The processing unit 1810 may be a physical or virtual processor and can implement various processes based on programs stored in the memory 1820. In a multi-processor system, multiple processing units execute computer executable instructions in parallel so as to improve the parallel processing capability of the computing device 1800. The processing unit 1810 may also be referred to as a central processing unit (CPU), a microprocessor, a controller or a microcontroller.
The computing device 1800 typically includes various computer storage medium. Such medium can be any medium accessible by the computing device 1800, including, but not limited to, volatile and non-volatile medium, or detachable and non-detachable medium. The memory 1820 can be a volatile memory (for example, a register, cache, Random Access Memory (RAM)), a non-volatile memory (such as a Read-Only Memory (ROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), or a flash memory), or any combination thereof. The storage unit 1830 may be any detachable or non-detachable medium and may include a machine-readable medium such as a memory, flash memory drive, magnetic disk or another other media, which can be used for storing information and/or data and can be accessed in the computing device 1800.
The computing device 1800 may further include additional detachable/non-detachable, volatile/non-volatile memory medium. Although not shown in
The communication unit 1840 communicates with a further computing device via the communication medium. In addition, the functions of the components in the computing device 1800 can be implemented by a single computing cluster or multiple computing machines that can communicate via communication connections. Therefore, the computing device 1800 can operate in a networked environment using a logical connection with one or more other servers, networked personal computers (PCs) or further general network nodes.
The input device 1850 may be one or more of a variety of input devices, such as a mouse, keyboard, tracking ball, voice-input device, and the like. The output device 1860 may be one or more of a variety of output devices, such as a display, loudspeaker, printer, and the like. By means of the communication unit 1840, the computing device 1800 can further communicate with one or more external devices (not shown) such as the storage devices and display device, with one or more devices enabling the user to interact with the computing device 1800, or any devices (such as a network card, a modem and the like) enabling the computing device 1800 to communicate with one or more other computing devices, if required. Such communication can be performed via input/output (I/O) interfaces (not shown).
In some embodiments, instead of being integrated in a single device, some or all components of the computing device 1800 may also be arranged in cloud computing architecture. In the cloud computing architecture, the components may be provided remotely and work together to implement the functionalities described in the present disclosure. In some embodiments, cloud computing provides computing, software, data access and storage service, which will not require end users to be aware of the physical locations or configurations of the systems or hardware providing these services. In various embodiments, the cloud computing provides the services via a wide area network (such as Internet) using suitable protocols. For example, a cloud computing provider provides applications over the wide area network, which can be accessed through a web browser or any other computing components. The software or components of the cloud computing architecture and corresponding data may be stored on a server at a remote position. The computing resources in the cloud computing environment may be merged or distributed at locations in a remote data center. Cloud computing infrastructures may provide the services through a shared data center, though they behave as a single access point for the users. Therefore, the cloud computing architectures may be used to provide the components and functionalities described herein from a service provider at a remote location. Alternatively, they may be provided from a conventional server or installed directly or otherwise on a client device.
The computing device 1800 may be used to implement visual data encoding/decoding in embodiments of the present disclosure. The memory 1820 may include one or more visual data coding modules 1825 having one or more program instructions. These modules are accessible and executable by the processing unit 1810 to perform the functionalities of the various embodiments described herein.
In the example embodiments of performing visual data encoding, the input device 1850 may receive visual data as an input 1870 to be encoded. The visual data may be processed, for example, by the visual data coding module 1825, to generate an encoded bitstream. The encoded bitstream may be provided via the output device 1860 as an output 1880.
In the example embodiments of performing visual data decoding, the input device 1850 may receive an encoded bitstream as the input 1870. The encoded bitstream may be processed, for example, by the visual data coding module 1825, to generate decoded visual data. The decoded visual data may be provided via the output device 1860 as the output 1880.
While this disclosure has been particularly shown and described with references to preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present application as defined by the appended claims. Such variations are intended to be covered by the scope of this present application. As such, the foregoing description of embodiments of the present application is not intended to be limiting.
| Number | Date | Country | Kind |
|---|---|---|---|
| PCT/CN2022/079002 | Mar 2022 | WO | international |
This application is a continuation of International Application No. PCT/CN2023/079534, filed on Mar. 3, 2023, which claims priority to Chinese Application No. PCT/CN2022/079002 filed on Mar. 3, 2022. The entire contents of these applications are hereby incorporated by reference in their entireties.
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/CN2023/079534 | Mar 2023 | WO |
| Child | 18823527 | US |
