Method and device for encoding and decoding a signal

Abstract

The invention relates to a method for decoding a data stream representative of a zone (Zodj) of a signal (I(Pdn)) segmented into a plurality of zones, the decoding method comprising: - decoding at least one feature map (FMüj) representative of at least the zone of the signal, comprising: ° determining (F22) an enlarged characteristic zone (ZCDj) in the at least one feature map (FMüj), corresponding to an enlarged zone of the signal comprising the zone of the signal; and ° decoding (F23) the data associated with the points of the enlarged characteristic zone (ZCI>); and - synthesising (F24-F29, F30) the zone of the signal (Zodj) on the basis of the decoded data.

Description

Description TITLE: Method and device for encoding and decoding a signal. Prior art

[0001] The invention relates to the general field of coding one-dimensional or multidimensional signals. It relates more particularly to the compression of digital images or videos.

[0002] Digital videos are generally encoded using source code to compress them and reduce the resources required for transmission and / or storage. Numerous encoding standards exist, such as those of the ITU / MPEG organizations (H.264 / AVC, H.265 / HEVC, H.266 / VVC, etc.) and their extensions (MVC, SVC, 3D-HEVC, etc.). In these approaches, image encoding is typically performed by predicting pixels using previously encoded and decoded pixels present in the image being encoded—in which case it is called "intra-prediction"—or previously encoded images—in which case it is called "inter-prediction."

[0003] In addition to these traditional approaches, approaches based on artificial intelligence, and in particular neural networks, are tending to develop.

[0004] Some neural network approaches use an input signal, such as an image, to train a synthetic neural network on characteristic vectors associated with a position in a sample of the input signal to be encoded. These characteristic vectors are constructed from feature maps, which may be at the resolution of the input signal or at a lower resolution. During training, or construction, the neural network parameters and the values ​​of the points on the feature maps (hereafter simply called feature map values ​​or characteristic values) are updated based on a performance metric, such as data rate-distortion.Once training is complete, meaning the performance measurement is satisfactory, the actual encoding of the synthetic neural network parameters and the feature map values ​​can be performed and stored or transmitted to the decoder. The encoded signal is then decoded by applying the synthetic neural network to the feature maps.

[0005] One drawback of classical and previously described neural approaches is that they do not allow the creation of independently decodable areas in the coded signal, which limits not only the ability to interact with the semantic content of the coded signal but also the possibility of limiting the memory required at the decoder level or parallelizing the decoding of this coded signal.

[0006] Therefore, there is a need for a solution that improves upon the classical and neural approaches described above. Summary of the invention

[0007] The invention relates to a method for encoding an area of ​​a signal, segmented into a plurality of areas, said encoding method comprising: - obtaining (E32) at least one characteristic map (FMi) representative of said signal (I(P n )), - the determination (E35) of an extended characteristic zone (1ZCD') in said at least one characteristic map (FMi), corresponding to an extended zone of said signal including said signal zone, - the coding (E36) of the value of the points of said extended characteristic zone (ZCD>).

[0008] The invention also relates to a method for decoding a data stream representative of an area (Zod j ) of a signal (I(Pd nsegmented into a plurality of zones, said decoding process comprising: the decoding of at least one characteristic map (FMDj) representative of at least said zone of the signal, comprising: • the determination (F22) of an extended characteristic zone (ECZ) j ) in said at least one feature map (FMDj), corresponding to an expanded signal area including said signal area, and • the decoding (F23) of the data associated with the points of said extended characteristic zone 1ZCD'). and the synthesis (F24-F29, F30) of said signal zone (Zod j ) from the decoded data.

[0009] For the purposes of this invention, encoding, or "coding," refers to the operation of representing one or more values ​​or a set of samples in a compact form carried by a data stream, for example, a digital binary stream. Decoding refers to the operation of processing a data stream, for example, a digital binary stream, to recover one or more decoded values ​​or samples.

[0010] A "sample" of the signal refers to a value taken from the signal. Sampling the signal produces a sequence of discrete values ​​called samples. In the case of an image signal, the sample is called a pixel, which can be, for example, a color pixel traditionally represented by a triplet of values, such as (R, G, B) or (Y, U, V). The position of the sample is identified by its x-coordinate (abscissa) and y-coordinate (ordinate) in the image.

[0011] The term "signal" refers to a signal with one (audio, sound), two (image), or more than two (stereoscopic, multiscopic, depth-mapped images, video, etc.) dimensions. Depending on its dimensionality, the sample has one, two, or more coordinates within the signal. In the case of an image signal, the sample's position is identified by its x-coordinate (abscissa) and y-coordinate (ordinate).

[0012] By "signal area," we mean a set of signal samples. For example, in the case of an audio signal, an area might correspond to a succession of units representing silences, noises, phonemes, words, etc. Similarly, in the case of an image signal, an area might correspond to a set of pixels, grouped according to criteria such as intensity, spatial distribution, etc. For example, a segmentation operation can identify two different areas within the image signal, one corresponding to the background and the other to the foreground.

[0013] Feature maps are abstract representations of signals containing a plurality of variable scalar data, discrete or non-discrete, also called values, such as real or integer numbers, signed or unsigned. These maps are also commonly referred to as latent representations, latent domain representations, or transformed domain representations.

[0014] The term "first rule," or sometimes "expansion rule," refers to a procedure that transforms an area by increasing its size (one-dimensional area), surface area (two-dimensional area), volume (two-dimensional area), or space (higher-dimensional area). A first rule is, for example, the Minkowski sum, which defines the morphological expansion of an area by adding a structuring element or a number of pixels to the region's boundary. Such a first rule may be known to the decoder or derived from a second rule. Therefore, it is generally not restricted to the decoder.

[0015] By "second rule," or sometimes "erosion rule," we mean a procedure that transforms an area by reducing its size (one-dimensional area), surface area (two-dimensional area), volume (two-dimensional area), or space (higher-dimensional area). A second rule is, for example, the morphological erosion of an area by a structuring element, or a number of pixels to be subtracted from the area's edge. Such a second rule may be known to the decoder. Therefore, it is not always transmitted to the decoder. In particular, such a second rule may apply a process inverse to that implemented by the first rule. It can in particular, it can be found by the decoder from the first rule, if the first rule is known to the decoder.

[0016] By "characteristic area" we mean a part of a characteristic map corresponding to an area of ​​the signal, defined in particular by its shape, size, etc.

[0017] The "enlarged signal area" refers to a portion of the signal obtained by applying the first rule to the signal area under consideration. The enlarged area thus encompasses the signal area. Such an enlarged area is also called an expanded area or a dilated area.

[0018] An "extended characteristic area" refers to a portion of a characteristic map corresponding to an expanded area of ​​the signal. Such an extended characteristic area is also called a magnified characteristic area or a dilated characteristic area.

[0019] Generally, the steps of an encoding or decoding process should not be interpreted as being linked to a notion of temporal succession. In other words, the steps can be carried out in a different order than that indicated in the independent encoding or decoding claim, or even in parallel.

[0020] The coding method according to the invention encodes a region of a signal from a representation of that signal in the form of characteristic maps by identifying in these characteristic maps the set of values ​​necessary to subsequently decode said region. In other words, the encoding of this set of values, called the extended characteristic region, makes it possible to obtain a coded representation of the region that can subsequently be decoded independently of any other part of the signal.

[0021] Thus, such a coding / decoding process improves the ability to interact with the semantic content of the coded signal and offers the possibility of parallelizing the decoding process of the entire signal.

[0022] The encoding and / or decoding process may further include one or more of the following optional features, taken individually or in any technically possible combination.

[0023] According to a particular embodiment, the encoding and / or decoding process further includes obtaining at least one segmentation information identifying said signal area or said enlarged signal area.

[0024] Segmentation information refers to any information that identifies one or more areas requiring specific encoding or synthesis for decoding. Such segmentation information can identify an area within the signal, or similarly, within a feature map representing the signal. For example, in the case of an image signal, segmentation information might indicate a division of the image into a plurality of areas. For instance, the signal areas might be The image can be divided into regular, identical-sized and shaped areas. Specifically, the image can be divided into rectangular areas with horizontal and vertical boundaries, or into areas corresponding to samples (or blocks) processed in a lexicographical order. In this case, the segmentation information can indicate the beginning and / or end of the set, for example, by using one or more flags. As another example, an area can be of arbitrary size and shape. In this case, the segmentation information can indicate the shape and size of the area, for example, in the form of a segmentation mask. Such a division of the signal into a plurality of areas can be known to the decoder (especially if areas of identical shape and size are considered). Therefore, the segmentation information is not always transmitted to the decoder.

[0025] According to a particular embodiment, the enlarged area of ​​the signal is obtained by applying a first rule to said area of ​​the signal or to at least one segmentation information identifying said area of ​​the signal.

[0026] According to a particular embodiment, the coding and / or decoding process includes obtaining said first rule.

[0027] According to a particular embodiment, said at least one segmentation information identifying said expanded area of ​​the signal is obtained by applying said first rule to said at least one segmentation information identifying said area of ​​the signal.

[0028] In one particular embodiment, the encoding and / or decoding process includes obtaining a second rule. In particular, the second rule applies a substantially inverse process to that implemented by the first rule.

[0029] According to a particular embodiment, the coding, resp. decoding process further includes the coding, resp. decoding, of said second rule.

[0030] According to a particular embodiment, the encoding, or decoding, process further comprises the encoding, or decoding, of at least one segmentation piece of information identifying the signal area or the extended signal area (i.e., within the domain of the signal (I(Pd) n )).

[0031] According to a particular embodiment of the decoding process, said synthesis takes into account the second rule allowing to obtain at least one segmentation information identifying said area of ​​the signal from said at least one segmentation information identifying said enlarged area of ​​the signal.

[0032] According to one embodiment of the decoding process, the synthesis involves determining at least one segmentation information identifying said area of ​​the signal by applying a second rule to said at least one segmentation information identifying said enlarged area of ​​the signal.

[0033] According to one embodiment of the decoding process, the synthesis comprises: processing said enlarged characteristic area to obtain an intermediate signal, and applying said at least one segmentation information identifying said area of ​​the signal to said intermediate signal to obtain samples of said area of ​​the signal.

[0034] According to one embodiment of the decoding process, the synthesis includes processing said enlarged characteristic area to obtain only samples of said area of ​​the signal.

[0035] According to one embodiment of the decoding process, the synthesis includes processing said enlarged characteristic area by taking into account said at least one segmentation information identifying said area of ​​the signal to obtain samples of said area of ​​the signal.

[0036] According to one embodiment of the decoding process, the second rule is morphological erosion by a structuring element.

[0037] According to a particular embodiment, said at least one segmentation information identifying said enlarged area of ​​the signal is a first mask and said at least one segmentation information identifying said area of ​​the signal is a second mask.

[0038] The expanded characteristic area can thus be determined by using a mask (first mask) encompassing the area of ​​the signal in the signal domain whose minimum area depends predictably on the type of encoding and / or decoding used to encode and / or decode the characteristic maps.

[0039] According to one embodiment of the decoding process, the second rule is defined by a maximum distance, the second mask comprising the points of the first mask located at a distance from an edge of the first mask in at least one direction inwards of the first mask greater than said maximum distance.

[0040] According to one embodiment of the decoding process, the second rule is defined by a number of parameters equal to twice the dimension of said signal, each direction of one of the dimensions of said signal being associated with a (possibly) different parameter, said second mask being obtained by removing from said first mask, in at least one direction of a dimension of said signal, a number of points equal to said parameter associated with said at least one direction of a dimension of said signal.

[0041] According to one embodiment of the decoding process, at least one feature map is representative of the signal in a latent domain, and the decoding process further comprises decoding the parameters of a neural network and in which the synthesis includes, for at least one sample of the signal, associated with a position in the signal: the construction of a characteristic vector from said decoded data as a function of said position of said current sample, and the processing of said characteristic vector by the neural network defined by the decoded parameters to provide a vector representative of the sample.

[0042] A "feature vector" constructed from feature maps based on a position is defined as a vector consisting of one or more elements, or data points, preferably discrete. These data points are constructed from feature maps at a position determined by the position of the sample being processed within the signal. This characteristic vector is applied to the input of the neural network. For example, in the case of a one-dimensional audio signal, such a vector can be constructed from a plurality of values ​​taken from each of the feature maps at the same coordinate as the sample to be encoded or in a neighborhood thereof. In the case of an image, such a vector can be constructed from a plurality of values ​​taken from each of the feature maps at the same x- and y-coordinates as the sample to be encoded (or decoded).Once these values ​​are taken from the feature maps, they can be processed to form the feature vector, before entering the neural network, for example by quantization, filtering, interpolation, etc.

[0043] By "neural network", we mean a neural network such as a convolutional neural network, a multilayer perceptron, an LSTM (for "Long Short Term Memory"), etc. The neural network is defined for example by a plurality of layers of artificial neurons and by a set of activation, weighting and addition functions (for example, a layer can calculate y = f (Ax+b), where y and b are vectors of dimension N, x a vector of dimension M, A is a matrix of dimension MxN, and f is the activation function).

[0044] By "neural network parameter" we mean one of the values ​​that characterizes the neural network, for example a weight associated with one of the neurons (filter coefficient, weighting, bias, value affecting the functioning of non-linearity, etc.)

[0045] "Neural network processing" refers to the application of a function expressed by a neural network to the input characteristic vector to produce an output vector representative of the sample to be encoded (or decoded). This output vector may contain one or more data points representative of the sample.

[0046] By "construction," we mean a step that aims to build the parameters representative of the image, before their actual encoding. The construction substeps can be repeated as many times as necessary to obtain an acceptable performance measurement.

[0047] According to one embodiment of the decoding process, said at least one characteristic map is representative of said signal in a transformed domain, for example associated with a direct discrete cosine transform, a Fourier transform, or a wavelet transform, and the synthesis includes the application to said decoded data of an inverse transform associated with said (direct) transform.

[0048] Correspondingly, the invention also relates to a corresponding coding device.

[0049] The invention also relates to a corresponding decoding device.

[0050] The characteristics and advantages of the coding or decoding process apply in the same way to the coding or decoding device according to the invention and vice versa.

[0051] The invention also relates to a computer program on a recording medium, this program being capable of being implemented in a computer or an encoding or decoding device according to the invention. This program includes instructions adapted to the implementation of the corresponding method. This program may use any programming language and may be in the form of source code, object code, or code intermediate between source and object code, such as in a partially compiled form, or in any other desirable form.

[0052] The invention also relates to a computer-readable information or recording medium containing the aforementioned computer program instructions. The information or recording medium can be any entity or device capable of storing programs. For example, the medium may include a storage means, such as a ROM (e.g., a CD-ROM or a microelectronic circuit ROM), or a magnetic recording means, such as a floppy disk or hard drive, a DNA sequence, or flash memory. Furthermore, the information or recording medium may be a transmissible medium, such as an electrical or optical signal, which can be transmitted via an electrical or optical cable, by radio link, wireless optical link, or by other means.

[0053] The program according to the invention can in particular be downloaded onto an Internet-type network.

[0054] Alternatively, each information or recording medium can be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of a process according to the invention. Brief description of the figures

[0055] The invention will be better understood with the aid of the following description, given solely by way of example and made with reference to the accompanying drawings in which: Figure 1 schematically represents an encoding device according to a first embodiment of the invention; Figure 2 schematically represents a decoding device according to the first embodiment of the invention; Figure 3 illustrates an example of a synthetic artificial neural network used in the context of the invention; Figure 4 schematically represents an example of the determination of an enlarged characteristic area by the encoding device of Figure 1 or by the decoding device of Figure 2; Figure 5 is a flowchart representing an example of an encoding method that can be implemented by the encoding device of Figure 1; Figure 6 illustrates an encoding method used by the encoding device of Figure 1.Figure 7 is a flowchart representing an example of a decoding process that can be implemented by the decoding device of Figure 2; Figure 8 illustrates a decoding process used by the decoding device of Figure 2; Figure 9 is a flowchart representing a method for encoding feature cards that can be implemented by the encoding device of Figure 1 and by the encoding process of Figure 5; Figure 10 is a flowchart representing a method for decoding feature cards that can be implemented by the decoding device of Figure 2 and by the decoding process of Figure 7; Figure 11 schematically represents a transcoding device according to a first embodiment of the invention. Detailed description of the invention

[0056] Figure 1 schematically represents, according to a first embodiment, an ENC coding device for at least one region of a signal (I(P n )) segmented into a plurality of zones. In the example described here, the signal I(P n The system is segmented into a set of J zones, and all zones are coded independently. Alternatively, only one zone or only a few zones can be coded independently.

[0057] In one embodiment, this ENC coding device includes a GEN module for generating feature maps, an SE module for transformation, and an XTR module. data extraction, a TT processing and quantification module, an NNSYN module corresponding to a synthetic artificial neural network, an NNC neural network coding module capable of coding the synthetic neural network, an FMC feature map coding module, an EVAL coding performance evaluation module, and an MAJ update module.

[0058] In one embodiment, the ENC coding device also includes an FMD decoding module for feature maps, a SEG module for obtaining zone segmentation, an ORD module for obtaining a first rule, an ORE module for obtaining a second rule, a CRE module for encoding the second rule, a DMZ module for obtaining an expanded zone mask, a CDMZ module for encoding this expanded zone mask, a DZCD module for determining expanded feature zones in the feature maps, and a CE module for encoding the value of the points of the expanded feature zones.

[0059] The ENC coding device can be implemented by means of an electronic device comprising a processor and a memory, not shown; each of the modules mentioned above can then be realized by the cooperation of the processor and computer program instructions stored in the aforementioned memory and designed to perform the functionalities of the module concerned, in particular as described below, when these instructions are executed by the processor.

[0060] The ENC coding device in Figure 1 receives as input a succession of samples to be coded, denoted P n , for example a temporal sequence of sound samples, or a set of image data denoted I(P n ). In this second case, the image signal I(P n ) can represent a two-dimensional image, or a plurality of two-dimensional images (video, color components, stereoscopic components, multiscopic components, etc.). P ndenotes a sample n of the input signal comprising N samples. In one embodiment, the signal is a color image signal represented by means of at least one two-dimensional representation, such as a pixel matrix, each pixel comprising a red (R), green (G), blue (B) component, or, alternatively, a luminance (Y) component and at least one chrominance (U, V) component. The location of each pixel is defined by its x and y coordinates in the image. In one embodiment, the image is a greyscale representation using a two-dimensional representation, such as a pixel matrix, with each pixel having a greyscale component, or luminance.

[0061] The GEN feature map generation module is configured to generate a plurality of M feature maps denoted FMi, forming for example a first group of feature maps.

[0062] In one embodiment, the SE module performs a transformation of the first group of FM characteristic maps; to generate a second group of FMSi characteristic maps at the same resolution as the input signal.

[0063] The optional SE module can perform quantification of the data extracted from this set of M FMi cards, or from the Z vector nconstructed from this data. Recall that quantizing a value refers to mapping that value to a member of a discrete set of possible code symbols. For example, the set of possible code symbols may consist of integers, and the quantization system performs a simple rounding of a real value to an integer. In another example, quantization involves multiplication by a given value followed by rounding. The SE module then performs a transformation of the values ​​in at least one of the feature maps, for example, oversampling, interpolation, filtering, etc. After the transformation, a transformed feature map from the second group has the same resolution as the images in the input sequence.Advantageously, according to this method, the feature maps that are coded can be of lower resolution than the images to be coded, while the maps of the second group, which are used to construct the feature vectors, are at the same resolution as the image sequence, which facilitates the extraction of values.

[0064] In one embodiment, the SE module is absent; in this case, the values ​​that will be used to construct the characteristic vector are extracted from the first group of characteristic maps.

[0065] The XTR module performs value extraction from the FM (and / or FMSi) feature maps for a current sample P n to be encoded, depending on its coordinates in the input signal. For example, if we want to encode sample P n at coordinates (x n , y nFrom an input image, the XTR module performs an extraction of values ​​in the maps at positions imposed by the coordinates (x n , y n ).

[0066] In one embodiment, the extracted values ​​constitute the vector Z n . Z n is an L-tuple, meaning it contains L elements, or data Zi. For example, in one embodiment, L=M, meaning that only one value is extracted for each feature map FMi. In another embodiment, L>M, meaning that multiple values ​​are extracted for at least one feature map FMi. The dimension L of the vector depends on the topology of the synthesis neural network NNSYN and more. particularly the topology of the input layer of this NNSYN synthetic neural network. The vector Z n index n refers to the characteristic vector of pixel P' n .

[0067] In one embodiment, the optional TT module processes the extracted values ​​to generate the vector Z n The TT module can perform quantification of the data extracted from the feature set. The processing may include other operations, such as filtering, scaling, etc. In particular, if the SE module is not used and if the feature maps in the first group have lower resolutions than the images in the sequence, the TT module can take into account the coordinates of the values ​​in the lower-resolution maps.

[0068] It should be noted that at least one of the SE or TT modules advantageously performs a quantification of the characteristic maps.

[0069] The NNSYN module is a synthetic neural network defined by K parameters Wk, capable of processing the vector Z n, or L-Tuple, as input, to generate as output a second vector representative of the sample P n to code.

[0070] An example of a synthetic neural network is presented later with reference to Figure 3.

[0071] The NNC module performs the encoding of the synthetic neural network, specifically its parameters Wk. During the training, or construction, process of the encoding—that is, until the performance evaluation stage is satisfactory—the NNC module performs a coding simulation, followed by decoding, for the evaluation module. Subsequently, it performs the actual encoding of the synthetic neural network parameters Wk. The encoded parameters are denoted Wck. As is known, the coding simulation can be identical to the actual encoding, or it can approximate it.

[0072] The FMC module performs the encoding of the FMi maps, that is, the values ​​of the feature maps in the first group (excluding the maps in the second group, which may optionally result from oversampling by the SE module). During the encoding training, or construction, process—that is, until the performance evaluation step is satisfactory—the FMC module performs a coding simulation, followed by decoding, for the evaluation module. Subsequently, it performs the actual encoding of the FM map values. The encoded maps are denoted FMc. As is known, the coding simulation can be identical to the actual encoding, or it can approximate it. The encoding module quantifies, if necessary, the latent representation of the values ​​in the first group maps using a quantifier to generate an ordered collection of quantized values.Then the coding module compresses the quantized data, using coding that takes into account the neighborhood of a value to be coded in the feature map.

[0073] The EVAL module performs an evaluation and minimization of coding performance. The evaluation function is, for example, of the rate-distortion type. Minimization can be performed by gradient descent, or any other method within the grasp of a person skilled in the art.

[0074] The MAJ module performs an update of the values ​​of the FM cards; to be encoded and / or the parameters of the Wk synthesis neural network, according to the results of the performance function.

[0075] When the evaluation function is minimized, or a predefined number of iterations have been performed, the coded values ​​FMc and Wck constitute the compressed representation of the input signal I(P n). At this stage, it is not possible to decode a specific area of ​​the coded signal from this compressed representation without decoding the entire coded signal.

[0076] To achieve this functionality, the FMD module decodes the FMc coded values. The cards decoded by the FMD module, numbering M, are labeled FMdi.

[0077] The SEG module performs a segmentation into J (greater than or equal to 2) zones Zo> of the sequence of samples to be coded P n This segmentation operation allows the samples to be grouped into J different homogeneous sets according to one or more predefined criteria. For example, in the case of a temporal sequence of sound samples, the segmentation operation makes it possible to obtain sequences of units corresponding to silences, noises, phonemes, words, etc. Similarly, in the case of an image signal I(P n), the segmentation operation allows the pixels P to be grouped together n of the image signal into J homogeneous zones according to criteria, notably intensity or spatial. For example, the segmentation operation can make it possible to identify in the image signal I(P n Two different zones (J=2) correspond, one to the background of the image and the other to the foreground. In the embodiments described below, segmentation is represented, for example, as a set of J masks MZo', each allowing the identification of a zone Zo', each mask being associated with a value different from that associated with the other masks. In other embodiments, simpler segmentation information can be considered, such as an indicator identifying the first and last samples of a signal zone, or the first sample of a zone and the size of the zone, or the shape of the zone, etc.

[0078] The ORD module obtains a first rule Rd, which allows the construction of an expanded zone from the signal zone. In other words, the first rule is intended to be applied to the zone or region of the signal I(P n ), or the corresponding mask, to obtain an expanded area or region in the signal domain, or a mask corresponding. Subsequently, the term "zone" or "region" is used interchangeably to designate a part of the signal in the signal domain, or a part of the characteristic maps in the transformed or latent domain.

[0079] In the following, we consider for example that this first rule is applied by the DMZ module to each mask MZo' to obtain an expanded mask MZD'.

[0080] Optionally, these expanded MZD' masks (defined in the signal domain (I(Pd n)) are encoded by the CDMZ module. More generally, at least one segmentation piece of information identifying the expanded area of ​​the signal, or even the signal area, is encoded by the CDMZ module.

[0081] This encoding can be achieved by independently encoding each expanded MZD' mask, or, in yet another variant, by independently encoding the contours of the J expanded MZD' masks, or, in yet another variant, by encoding a size or position data for an expanded area, for example, a rectangle, or a number of blocks. The encoding of expanded MZD> masks is denoted MZD>c.

[0082] For each expanded mask MZD', an expanded characteristic zone ZCD> of the FMdi maps is determined by the DZCD module. This expanded characteristic zone ZCD> includes all the points of the FMdi maps co-located (possibly up to a scale factor) with the points of an expanded mask MZD'.

[0083] The ORE module derives a second rule, Re, for example, from the first rule, Rd. This second rule allows the signal region, or an approximation thereof, to be recovered from the expanded region. For example, the second rule allows the MZo> mask, or an approximation thereof, to be obtained from the expanded MZD> mask.

[0084] Optionally, the CRE module performs the encoding of the second rule Re obtained by the ORE module in a coded form denoted Rec.

[0085] The CE module performs the entropic coding for each zone Zo' of the value of the points of the extended characteristic zones ZCD' in the form of coded EcJ data different for each zone Zo>.

[0086] In a first example of implementation, the CE module only codes the values ​​of the points in the extended characteristic zone ZCD' to the exclusion of any other point in the FMdi characteristic maps.

[0087] Alternatively, the CE module creates a secondary FM2d feature map for each extended feature area ZCD. The value of a point on the secondary FM2d feature map is equal to the value of the point on the FMdi map if that point belongs to the extended feature area ZCD, and to a predetermined value, for example zero, otherwise. The CE module then codes the entire FM 2d characteristic map to obtain the different EcJ coded data for each Zo' zone.

[0088] Thus, the coded values ​​Wck, EcJ, Rec and MZD'c constitute the compressed representation of the Zo> region of the input signal I(P n These coded values ​​are represented as binary streams Bl, B2' and B3'. Thus, according to this example, the binary stream B1 includes the coded values ​​Wck, the binary stream B2' the coded values ​​EcJ and optionally the binary stream B3' the coded values ​​Rec and MZD'c.

[0089] Thus, thanks to the invention, each Zo' zone is coded independently of the other zones, which allows their subsequent decoding to be carried out independently and possibly in parallel.

[0090] It should be noted that the ENC coding device of an area of ​​an I(P) signal n ) described above includes the successive application of a signal coding device I(P n ) in its entirety in order to produce a B0 stream comprising the FMc-coded values; and a transcoding device for this globally coded signal into a code for one or more area(s) of the signal I(P n ), the said zone(s) being decodable independently of each other.

[0091] According to this embodiment, the transcoding device includes the FMD decoding module for feature maps, the SEG module for obtaining zone segmentation, the ORD module for obtaining a first rule, the ORE module for obtaining a second rule, the CRE module for encoding the second rule, the DMZ module for obtaining an extended zone mask, the CDMZ module for encoding this extended zone mask, the DZCD module for determining extended feature zones in the feature maps, and the CE module for encoding the value of the points of the extended feature zones.

[0092] Figure 2 schematically represents a DEC decoding device for a data stream representative of a Zod' zone of a signal segmented into several zones, said Zod' signal zone comprising a plurality of samples Pd n to decode.

[0093] The DEC decoding device in Figure 2 receives input data streams Bl, B2', and optionally B3'. Stream Bl includes the encoded parameters Wck of the synthesis neural network NNSYN'. Stream B2' includes a set of encoded data EcJ corresponding to the M feature maps FMd! representative of the area to be decoded Zod', and stream B3' includes a coded form Rec of a second rule and the encoded parameters of an expanded mask MZD'c of the signal area Zod'.

[0094] In one embodiment, this decoding device DEC comprises an NND neural network decoding module capable of decoding the synthesis neural network NNSYN', a DRE module capable of decoding a second rule Re, a DDMZ module capable of decoding segmentation information enabling the identification of the expanded signal area, for example an expanded mask MZD' of the area in the signal domain or one or more flags, an EMZ module for obtaining segmentation information enabling the identification of the signal area, for example an MZo' mask of the area in the signal domain or one or more flags, a DZCD module capable of determining an expanded feature area ZCD' in the M feature maps, a DE module for decoding the feature maps from the encoded data group EcJ, an XTR' data extraction module, an SE' inverse transformation module, and a TT' inverse processing and quantization module.an NNSYN' module corresponding to a synthetic neural network and an optional EXTR module for extracting samples belonging to the Zod'IP zone,,).

[0095] According to one embodiment, the DEC decoder produces at output a decoded area of ​​the image Zod'lPd,,) comprising a plurality of decoded samples Pd n .

[0096] The DRE module obtains a second rule, for example by decoding Rec-encoded parameters if they are present in the data stream. Alternatively, the second rule may be known to the decoder, or obtained by the decoder from the first rule.

[0097] The DDMZ module optionally decodes segmentation information to identify the expanded area, for example the expanded MZD> mask of the signal area, from the MZD'c data of the B3 stream JIf the segmentation information is not decoded, it can be inferred by the decoder (for example, because it is static, or because the decoder knows the size of the rectangular areas corresponding to the tiles or the number of blocks in the slices, etc., before or after widening the signal area).

[0098] Based on the second rule Re and the segmentation information used to identify the expanded area (e.g., the expanded mask MZD'), the EMZ module determines the segmentation information used to identify the signal area (e.g., the mask MZod' of the signal area Zo>). Specifically, the mask MZod' can be identical to the mask MZo' if the second rule is the inverse function of the first rule used by the ENC encoder; otherwise, it will be different.

[0099] The DZCD module determines the extended characteristic zone ZCD' in the FMdJ characteristic maps. This characteristic zone ZCD' includes the set of points from the FMdi maps co-located (possibly up to a scale factor) with the points of the expanded mask MZD>.

[0100] The DE module performs the decoding of the point values ​​of the extended characteristic zone ZCD> from the coded EcJ data in order to obtain the FMdJ characteristic maps.

[0101] According to embodiments as described for the encoder: FMdJ cards are decoded independently of the MZD expanded mask JThen, only the values ​​corresponding to points in the characteristic zone ZCD' are retained. FMdJ maps are decoded taking into account the expanded mask MZD' when the EcJ-encoded data only includes values ​​corresponding to points in the characteristic zone ZCD'.

[0102] The NND module decodes the parameters of the NNSYN' synthesis neural network. These parameters are denoted Wdk.

[0103] In one embodiment, the SE' module transforms the first set of decoded FMdJ feature maps to generate a second set of feature maps with the same resolution as the signal to be decoded, denoted FMS'J. Optionally, the SE' module performs inverse quantization corresponding to the quantization performed by the encoder. Inverse quantization is not necessary if the encoder's quantizer Q simply rounded the input values. Inverse quantization is also unnecessary if the neural network is capable of handling quantization of its input data. Otherwise, the decoder can perform the inverse operation of quantizer Q. The SE' module then transforms the feature map values, including, for example, oversampling, interpolation, filtering, etc., similar to the transformation performed by the encoder.At the end of the transformation, a transformed feature map of the second group has the same resolution as the images of the sequence to be decoded.

[0104] In one embodiment, the SE' module is absent; in this case, the values ​​that will be used to construct the characteristic vector are extracted from the first group of characteristic maps.

[0105] The XTR' module is identical to the XTR module in Figure 1. It performs an extraction of values ​​from the M FMdJ (and / or FMS'J) feature maps, depending on one of the embodiments described previously, for a sample Pd n to be decoded according to its coordinates in the signal to be decoded.

[0106] In one embodiment, the extracted values ​​constitute the vector Zd n . Zd n is an L-tuple, that is to say it contains L elements, or zdi data.

[0107] In one embodiment, the optional TT' module processes the extracted values ​​to generate the vector Zd n The TT module can perform inverse quantization of the data extracted from the feature set. The processing can include other operations, such as filtering, scaling, etc., similar to those performed by the encoder.

[0108] The NNSYN' module is a so-called synthetic neural network, defined by K parameters Wdk, capable of processing the vector Zd nThe input is an L-tuple, or L-tuple, to generate as output a second vector representing a decoded sample, generally a vector containing A elements. In one embodiment, K=3 and the output vector is the triplet (R, G, B) of the decoded sample. The NNSYN' module has the same structure as the NNSYN module, and its parameters are either identical if the encoding of its parameters Wk is lossless, or different if the encoding is lossy.

[0109] The processing of decoded FMdJ maps described above generates an intermediate image L(P n ) which contains samples that may not belong to the area to be decoded Zod(Pd n ). To obtain this area to be decoded, the EXTR module transforms the intermediate image L(P n ) into a signal corresponding to the Zod zone j (P n ) of the image I(Pd n ). For this purpose, the EXTR module stores samples of the intermediate image L(P n) belonging to the MZod' mask (provided to the EXTR module by the EMZ module) while the others are suppressed, the resulting signal then corresponding to the Zod'(P) zone ri ) of the image I(Pd n ).

[0110] Alternatively, the NNSYN' module only processes Zd vectors n input capable of generating at output a second vector representative of a sample Pd n to decode. This selection of Zd vectors n is performed by the NNSYN' module, taking into account the MZod' mask of the area to be decoded. In this variant, the intermediate image L(P n ) corresponding directly to the area to be decoded Zod>(P n ) without the DEC decoder including the EXTR module.

[0111] The DEC decoding device can be successively implemented to decode all Zod' zones in order to reconstruct all the decoded samples of the signal, i.e., of the I(Pd) image n ).

[0112] Alternatively, multiple DEC decoding devices can be implemented in parallel, so as to decode the different areas of the image simultaneously. Once all areas are decoded, the image can be reconstructed by combining the different areas.

[0113] The DEC decoding device can be implemented by means of an electronic device comprising a processor and memory, not shown; each of the modules The aforementioned functions can then be achieved through the cooperation of the processor and computer program instructions stored in the aforementioned memory and designed to perform the functionalities of the module in question, particularly as described below, when these instructions are executed by the processor. The DEC device can also include multiple processors, each dedicated to the parallel decoding of image areas.

[0114] Figure 3 illustrates an example of a synthetic artificial neural network used for encoding and decoding in embodiments of the invention.

[0115] The synthetic artificial neural network used for encoding, NNSYN, and the synthetic artificial neural network used for decoding, NNSYN', are defined by an identical structure, including for example a plurality of layers of artificial neurons, and by a set of weights and activation functions associated respectively with the artificial neurons of the network concerned.

[0116] The synthetic neural network is, in one embodiment, an MLP, or Multi-Layer Perceptron, followed by one or more convolutional neural network(s) Ci, ... C n , each of the convolutional neural networks being associated with a filtering mask of predefined size, for example size 3x3.

[0117] MLP consists of an input layer adapted to the input format (the L-tuple), optionally one or more hidden layers, and an output layer providing an intermediate output vector Vs n For example, a vector containing A' elements. The intermediate output vector is then successively processed by the convolutional neural network(s) to provide an output vector also containing A elements.

[0118] Thus, a vector representation of a current sample (a vector Z) n or Zd n (derived from the EMi / EMSi feature maps at the encoder or FMdJ / FMS'J at the decoder) is applied to the input (i.e., the input layer) of the NNSYN or NNSYN' synthetic artificial neural network, which produces the output vector. In one embodiment, A is equal to 3 and the output vector is the triplet (R, G, B) of pixel P' nencoded then decoded or pixel Pd n decoded by the decoder.

[0119] The concatenation of all these reconstructed pixels in an image constitutes the reconstructed Zod signal J (Pd n ), according to an example, an image I containing only the Zod area J .

[0120] At the encoder, the NNSYN synthetic artificial neural network is trained on the image, so as to minimize the differences between the input representation of the current image I(P n ) and its representation I(P' n ) at the output, while also minimizing the amount of data to be encoded. The EVAL module performs a performance measurement in this regard.

[0121] A "performance measure" is defined as a measurement between at least one value of a sample to be encoded and a decoded value of that sample. The measure might assess, for example, distortion or perceptual error. It can be performed on one or more samples (e.g., a running sample, the current image, etc.). The measure may also include a measurement of throughput, particularly associated with the encoding of the synthetic neural network and / or the encoding of the feature maps in the first group. The measure may be a combined measurement of throughput and distortion through their weighting. As is well known in the state of the art, the value of this measure is generally minimized until a target value is reached.

[0122] Once the encoder training is complete, the network parameters are encoded, either losslessly, in which case the NNSYN' neural network is identical to NNSYN, or lossily, in which case the NNSYN' network may be slightly different from NNSYN.

[0123] With reference to Figure 4, we will now describe an example of determining an expanded characteristic zone (ACZ) 1 associated with a Zo zone 1 in the case of an image signal I(P n ) segmented into three zones Zo 1 Zo 2 Zo 3 identified by their MZo mask 1 , MZo 2 and MZo 3 Figure 4 thus presents two steps (1)-(2) of determining the enlarged characteristic zone corresponding to the Zo zone 1 .

[0124] In the example described, the expanded characteristic area is determined in four EMdi, EMdi, EMd, and ElVkU input maps of the NNSYN / NNSYN' synthesis network. Furthermore, this example assumes that the largest convolutional kernel used by the convolutional neural networks of the NNSYN / NNSYN' synthesis network is a 3x3 convolutional kernel.

[0125] Thus, in the example presented, the decoding of a sample Pd ni typically requires access not only to the point values ​​of the EMdi, EMdi, EMd3, and EMcL maps co-located with this sample Pd ni but also to the point values ​​of the EMdi, EMdi, EMd3, and EMd4 maps co-located with the Pd samples n 2 belonging to the neighborhood used during filtering, from this point Pd n i. In other words, when the sample Pd ni is on the edge of the Zo zone 1Its decoding requires access to points on EMdi, EMdi, EMd3, and EMcL maps co-located with a Pd sample. n 2 belonging to a Zo zone 2 different from the Zo zone 1 The set of points on the EMdi, EMd2, EMd3, and EMd4 maps necessary to decode the Zo zone 1 defines the extended characteristic zone ZCD 1 of these EMdi, EMd2, EMd3, EMd4 cards.

[0126] This determination includes, in a step (1), obtaining an expanded MZD mask 1 by applying a first rule to the MZo mask 1 , or, equivalently, obtaining an expanded area from an area of ​​the signal.

[0127] In the example described here, the first rule is defined by a structuring element ES, namely a 3x3 square corresponding to a convolution kernel used within the NNSYN / NNSYN' synthesis neural network. In this example, the expanded mask MZD 1is thus obtained by a morphological dilation of the MZol mask by this structuring square element of size 3x3.

[0128] Alternatively, the first rule can be defined in the case of an image I(P n ) by four values ​​indicating the number of points in the signal I(P n )to be added for each point on the edge of the MZo mask 1 (or the area of ​​the signal under consideration) in both horizontal and vertical directions. In the example given, all four values ​​are equal to one.

[0129] In another variant, the structuring element can be defined by a scale factor indicating proportionality to the size of the Zo zone 1 the number of points of the signal I(P n ) to be added to the edge of the MZo mask 1 to obtain the expanded MZD mask 1 .

[0130] During a step (2), the enlarged characteristic zone ZCD 1From these FMdi, FMdg, FMdg, FMdg maps, this expanded characteristic zone ZCD is obtained. 1 includes all points of the FMdi, FMdg, FMdg, FMd4 maps co-located with samples belonging to the expanded MZD mask 1 of the Zo zone 1 .

[0131] It should be noted that, according to the first rule chosen, the extended characteristic zone ZCD 1 FMdi, FMdg, and FMdg maps. FMdg may include, depending on the shape of the Zo zone outline, 1 more points than the total number of points on the FMdi, FMdg, FMdg, FMdg maps strictly necessary to decode the Zo zone 1 .

[0132] When the FMdi, FMdg, FMdg, FMdg feature maps have the same resolution as the I(Pn) image, as in the example shown, the boundary of this expanded feature area ZCD 1is in each FMdi, FMdg, FMdg, FMdg feature map identical to the boundary of the expanded MZD mask 1 .

[0133] In the previous example, the FMdi, FMdg, FMdg, FMdg feature maps have the same resolution as the I(Pn) image, but it should be noted that the determination of the expanded feature area associated with the Zo area 1 When the FMdi, FMdg, FMdg, FMdg feature maps have different resolutions, the I(Pn) image is determined in a similar way. Simply, the determination of the boundary of the expanded feature area takes into account the resolution transition function of the I(Pn) image. n ) to that of the FMdi, FMdg, FMdg, FMdg feature maps. For example, if the FMdi, FMdg, FMdg, FMdg feature maps have a lower resolution, the location of the boundary of the expanded feature area lies between the points whose samples are co-located in the image I(Pn ) is located on either side of the Zo zone border 1 .

[0134] Figure 5 is a logic diagram representing an example of a method for encoding at least one area of ​​a signal (I(P n )) which can be implemented by the coding device in Figure 1, for example when the NNSYN synthesis neural network is the one shown in Figure 3.

[0135] According to this embodiment, the signal is a two-dimensional image; each sample to be coded is therefore a pixel P n coordinates (x n , y n ).

[0136] The encoding process takes place in several phases.

[0137] In the first phase, called the construction phase, a learning process is carried out in order to determine, for an input signal I(P nThe values ​​of the FM maps and Wk parameters are used to optimize an overall cost function. Learning is performed, for example, by gradient descent, followed by updating the parameters of the NNSYN synthesis neural network and the values ​​of the FM feature maps. As is known in the prior art, the cost function can be rate-distortion, rate-distortion, or perceptual. To measure the throughput R, it is necessary to simulate the encoding of the FM maps and then measure the associated encoding throughput (the size of the B0 stream). In one embodiment, the encoding of the Wk parameters is not simulated because their influence is less significant than that of the feature maps. In another embodiment, the encoding of the Wk parameters is also simulated, and the associated throughput (the size of the B1 stream) is measured.To measure the distortion D, it is necessary to simulate the encoding and then the decoding of at least part of the image I, to obtain at least one pixel P'. n resulting from a simulation of encoding and then decoding, then measuring the difference between this part of the image I(P n ) as input and a corresponding part of the image I(P' n ) encoded then decoded.

[0138] Then, in a second phase, called the encoding phase, the Wk parameters, the FM dj maps, optionally the segmentation information (e.g., the expanded MZD> masks), and the second Re rule are encoded to produce the encoded values ​​Wck, EcJ, MZD'c, and Rec before transmission or storage. These constitute the compressed representation of the Zo' region of the input signal I(P n ).

[0139] Naturally, the first phase and the second phase correspond to two independent processes that can be carried out on different devices.

[0140] We will now describe the steps of a process for encoding a region Zo> of an input signal I(P n ) according to the first embodiment of the invention.

[0141] During an E20 step, a signal I(P n ) to be coded, comprising a plurality of N samples P n is provided as input to the process.

[0142] During step E21, the M FM maps of the first group and the Wk parameters of the synthesis neural network are initialized. Subsequently, the Wk parameters of the network NNSYN synthesis neurons and FM map values ​​must be optimized during the construction phase.

[0143] According to one embodiment, the FM cards have the same resolution as the input signal I(P n ) and therefore each have the same number of values ​​N as there are samples P n to code.

[0144] According to one embodiment, the FM cards have a resolution less than or equal to that of the input signal I(P n ) and therefore include, for at least one of them, a number N' of values ​​to be coded less than N. According to a variant, the first FMi card is at the resolution of the images and each subsequent card is at half the resolution of the previous one.

[0145] According to one embodiment, several FMi cards have the same resolution, lower than that of the input signal I(P n ).

[0146] In one embodiment, the FMi maps are transformed to provide a second group of transformed FMSi feature maps. In this embodiment, the feature vectors are preferably extracted from the transformed maps of the second group, and not directly from the maps of the first group. Therefore, in this embodiment, the feature vectors are indirectly extracted from the maps of the first group. The maps of the second group are not coded; they are used solely for constructing the feature vectors.

[0147] According to one embodiment, FMi cards are initialized with predefined constant values.

[0148] According to another embodiment, the feature maps are initialized by a set of random real values.

[0149] The FMi feature maps of the first group are subsequently updated, or refined, during an E22 step, by the encoder's MAJ update module during its learning.

[0150] During step E23, the FMi cards of the first group are encoded by the FMC module of the encoder. During the build phase, this operation is a coding simulation. During the actual encoding phase, this operation is the actual encoding, and the encoded values ​​constitute the B0 stream. The simulation may be identical to the actual encoding, but it may also be different (for example, simplified). For this encoding, a technique is used to predict a feature card value based on its neighborhood, as will be described, for example, in support of Figure 9. These parameters are subsequently updated, or refined, during the build phase, in later iterations of the process.

[0151] In one embodiment, the FMi cards are coded in order (FMi, FM2, ..., FM4), and the point values ​​of each card are in a predefined order, for example lexicographic. Each card undergoes entropic coding. The entropic coding produces a compressed stream B0 whose throughput is subsequently measured during a step E29.

[0152] In a step E24, according to one embodiment, the M cards of the first group FM; are transformed by the SE module to generate cards of the second group FMSi at the resolution of the images of the input sequence.

[0153] According to one embodiment, M FMSi cards are generated.

[0154] According to one embodiment, each FM card; is transformed into an FMSi card.

[0155] In one embodiment, at least one FM1 map has a lower resolution than the images in the sequence to be encoded, and the transformation operation includes oversampling so that the transformed FM1 map has the same number of samples as the images in the sequence. This oversampling consists of adding values ​​to the FM1 maps to achieve the resolution of the input sequence images. It can be simple (by nearest neighbor replication) or involve interpolation (linear, polynomial, filtered, etc.).

[0156] Apart from step E25, values ​​are extracted by the XTR module from the FM cards; or possibly transformed FMSi cards. This extraction is performed based on the coordinates (x n , y n ) of sample P n of the input signal. It can also be done according to the resolution of the card in question.

[0157] According to one embodiment, the vector Z n This characteristic results directly from this extraction.

[0158] The samples to be coded are, for example, processed in sequential order, from n=1 to n=N.

[0159] According to one embodiment, during step E26, the vector Z n The characteristic is constructed by the TT module from the values ​​extracted from the FM maps; or FMSi for each sample P n coordinates (x n , y n ) of the input signal. This processing may involve quantizing the values ​​extracted from the FM cards; or from the Z vector n constituted, if necessary. This processing may include other operations, such as filtering, scaling, the application of any function, preferably monotonic, etc.

[0160] In one embodiment, Z n It contains as many values ​​as there are FMi or FMSi cards in the input. In this case, E=M.

[0161] In one embodiment, Z n is an E-tuple (zi, Z2, ..., Zj), consisting of the values ​​of the FM or FMSi cards located at the coordinates (x n , y n ) of the current pixel P n as will be illustrated in Figure 6. In another embodiment Zd n is an E-tuple (zi, Z2, ..., Zj), consisting of the values ​​located in a neighborhood around these coordinates

[0162] In one embodiment, Z n is an E-tuple constructed from values ​​taken from FMi maps at coordinates that may differ between maps. By For example, if the FM maps are at different resolutions because they have been undersampled, the coordinates are adapted (by scaling) to match the resolution of each map.

[0163] In one embodiment, Z nis an L-tuple constructed from values ​​taken from the FM maps; by applying processing to one or more values ​​from the maps, for example, filtering out values ​​close to the target value in a map. For example, for a current sample P n In an FMi card that has the same resolution as the input signal, we can extract the values ​​located at the coordinates (x n , y n ), (x n -l, y n ), (x n , y n -l) and (x n -l, y n -l) and apply processing to these values ​​(filtering, averaging, interpolation, etc.) to obtain the final value (zQ of element i of the vector Z) n relating to this FM card; According to another example, in an FM card; which is at half the resolution of the input signal, we can consider the values ​​located at the coordinates (x n / 2, y n / 2), (x n / 2-l, y n / 2), (x n / 2, yn / 2-l) and (Xn / 2-1, y n / 2-l) and apply processing to these values ​​(filtering, averaging, interpolation, etc.) to obtain the final value (zQ of element i of the vector Z) n relating to this FM card;

[0164] During step E27, the vector Z n is processed by the NNSYN synthesis neural network to generate as output a vector representative of the sample P n to code, according to one embodiment, the triplet (R, G, B) of the sample P' n (sample P) n (encoded then decoded).

[0165] The structure and Wk parameters of the synthesis neural network are initialized, for example, during the first iteration of this step. These parameters are subsequently updated, or refined, during the construction phase, in later iterations of the process.

[0166] According to one embodiment, the parameters of the synthetic neural network are initialized by predefined values ​​known to give a satisfactory result (for example, following training on a corpus of images).

[0167] According to another embodiment, the parameters of the synthesis neural network are initialized by a set of random values.

[0168] During step E28, the Wk parameters of the NNSYN synthetic neural network are quantized and encoded. During the construction phase, this operation is a coding simulation. During the encoding phase, this operation is the actual encoding, and the encoded values ​​constitute the B1 stream. The simulation can be identical to the actual encoding, but it can also be different (for example, simplified). Any known technique can be used for this purpose, for example, the neural network coding standard proposed by the MPEG-7 Part 17 standard, also called NNR (Neural Network Representation). Note that in this case, the amount of degradation that the encoding introduces to the Wk weights must be chosen.

[0169] During an E29 step, a performance measure is evaluated.

[0170] To this end, the coding simulation rates associated with the feature maps of the first group (simulation of the B0 flow by coding the FMi maps) and optionally with the parameters of the neural network (simulation of the B1 flow by coding the Wk parameters) are measured.

[0171] In one embodiment, the cost function is of the rate-distortion type, denoted (D+L*R), where D, for example, is the root mean square error measured between the input signal and the decoded signal (or the error measured on a subset of the signal samples). In another example, D is calculated from a perceptual function such as SSIM (for Structural SIMUarity) or MSSSIM (for Multi-scale Structural SIMUarity). In one embodiment, R is the simulated rate of the B0 stream. In another embodiment, R is the total rate used to encode this image, i.e., the sum of the simulated rates of B0 and B1. L is a parameter that controls the rate-distortion trade-off. Other cost functions are possible.

[0172] As long as the cost function has not reached its minimum, or a predefined number of iterations has not been reached, the performance measurement is not satisfactory, and the process is repeated from step E22. This minimization can be performed by a mechanism known as gradient descent with parameter updates during step E22 for the feature map values ​​and E23, E27 for the network parameters.

[0173] During step E30, if the cost function has reached its minimum, or if the desired number of iterations has been reached, training stops. If a coded version corresponding to the last simulation of the synthesis neural network parameters Wk and the feature maps FM; is available, the streams B0 and B1 can be constructed from it. According to another embodiment, the actual coding of the updated synthesis neural network parameters Wk and the values ​​of the feature maps FM; is performed at this step to produce the encoded parameters Wcket FMc; which constitute the streams B0 and B1.

[0174] At this stage, it is not possible to decode a specific area of ​​the coded signal from the B0 and B1 streams without decoding the entire coded signal.

[0175] To achieve this functionality, according to one embodiment, the coding process performs the steps described below. These steps correspond to a transcoding, or re-encoding, of the B0 stream into a new zone-decodable stream.

[0176] During an E31 step, the FMc coded values ​​are decoded by the FMD module to generate the FMdi maps.

[0177] During an E32 step, the segmentation of the signal I(P n ) in J (greater than or equal to 2) zones Zo> is performed by the SEG module.

[0178] According to a first example of segmentation, the image is divided into J regular zones, for example, of identical size and shape (except possibly at the edges of the image). For example, the image can be divided into rectangular zones with horizontal and vertical boundaries. Such a division corresponds to the concept of "tiles" implemented by encoding standards such as H.264 / AVC, H.265 / HEVC, and H.266 / VVC. The segmentation information for a given area of ​​the image allows these rectangular zones to be identified, for example, by specifying a vertical and horizontal size, the coordinates of a rectangle, etc.

[0179] In a second example of segmentation, the image is divided into J zones corresponding to samples (or blocks) scanned in a lexicographical order. This division corresponds to the concept of "slices" implemented by encoding standards such as H.264 / AVC, H.265 / HEVC, and H.266 / VVC. The segmentation information for a given area of ​​the image allows these areas to be identified, for example, by indicating the number of blocks.

[0180] In a third segmentation example, the image is divided into semantic zones, such as a background and a foreground. This can be done manually by an operator. Segmentation can also be automatic or semi-automatic, depending on the segmentation algorithm used. It's worth noting that there are no restrictions related to the type of segmentation algorithm used. Segmentation information allows these zones to be identified, for example, through a segmentation mask.

[0181] Advantageously, such a division allows, during the subsequent decoding of the signal I(Pd n ) to parallelize the decoding of the zones to be decoded Zod J by distributing the decoding load evenly across each of the decoders and / or decoder processors.

[0182] At the end of step E32, this segmentation is represented as segmentation information allowing the identification of at least one area of ​​the image. We consider below a set of J masks MZo', each allowing the identification of an area Zo>.

[0183] For a given Zo' zone, the coding process then performs steps E33 to E39.

[0184] During an E33 step, a first rule Rd is obtained by the ORD module.

[0185] During an E34 step, segmentation information is obtained to identify an expanded area encompassing the signal area. For example, an expanded MZD maskJ is obtained by applying the first rule Rd to the mask MZo>.

[0186] According to one embodiment, the first rule Rd can be defined by a structuring element ES defined as a square whose dimension is a function of the size of the convolutional kernels used by the convolutional neural networks of the synthetic neural network NNSYN. The expanded mask MZD> is for example obtained by a morphological dilation of the mask MZo> by the structuring element ES.

[0187] According to another embodiment, the first rule is defined by a number of numerical values ​​equal to twice the dimension of the signal. Thus, there are four numerical values ​​for a 2-dimensional signal such that an image I(P n) corresponding to a dilation in each of the two vertical and horizontal directions. The expanded mask MZD' is obtained by adding a number of points to the boundary of the mask MZo' in the two directions associated with each dimension of the signal. In other words, for a 2-dimensional signal such as an image I(P n pixels are added in both horizontal directions and both vertical directions.

[0188] According to another embodiment, the first rule is defined by a scale factor. The expanded mask MZD' is obtained by adding to the points of the mask MZo' a number of points at the edge of the mask MZo' proportionally to the size of the mask MZo>.

[0189] During an E35 step, an expanded characteristic zone ZCD' of the EMdi maps comprising all points of the EMdi maps co-located with samples belonging to the expanded mask MZD' is obtained.

[0190] During step E36, the CE module performs the encoding for each zone Zo> of the point values ​​of the extended characteristic zones ZCD' in the form of coded data EcJ. The coded EcJ values ​​constitute the flux B2' associated with the zone Zo'. Note that this encoding is entropic in nature, and preferably lossless.

[0191] During an E37 step, the expanded MZD' mask of the Zo' zone is encoded as an MZD'c stream.

[0192] During step E38, a second rule Re is obtained. In one embodiment, this second rule Re is derived from the first rule Rd. This second rule allows, for example, obtaining the mask MZo' or an approximation thereof from the expanded mask MZD'.

[0193] For example, according to an embodiment as described to the coder, the second rule Re is defined by a structuring element obtained by a morphological erosion corresponding to the morphological dilation element defined previously.

[0194] According to another embodiment, the second rule Re is defined by a number of numerical values ​​equal to twice the dimension of the signal. Thus, there are four numerical values ​​for a 2-dimensional signal such that an image I(P n ) corresponding to erosion in each of the two vertical and horizontal directions. The MZo> mask is obtained by subtracting a number of points from the boundary of the expanded MZD> mask in both directions associated with each dimension of the signal corresponding to the points that were added when applying the first rule.

[0195] According to another embodiment, the second rule is defined by a scaling factor inverse to that used in the encoder. The mask MZo' is obtained by subtracting a number of points from the edge of the expanded mask MZD' proportionally to the size of the mask MZD'.

[0196] During an E39 step, the second rule is coded in a form denoted Rec

[0197] The coded values ​​MZD'c and Rec constitute the B3' flow associated with the Zo' zone.

[0198] Thus, the coded values ​​Rec, EcJ, MZD'c and Wck constitute, according to the illustrated embodiment, the compressed representation of the Zo' region of the input signal I(P n ).

[0199] It should be noted that steps E31 to E39 can be analyzed as steps in a transcoding process of a stream B0 representing a coded image (obtained during a preliminary step EP) into two streams B2' and optionally B3' representing the encoding of a Zo' region of the coded image. The preliminary step EP can correspond to the encoding steps (E20-E30) described previously, or to any other encoding process resulting in a coded representation of a set of feature maps (and a synthetic neural network).

[0200] Figure 6 illustrates a method for encoding an I(P) signal. n ) used in an embodiment of the invention.

[0201] In this embodiment, there are 4 FM cards generated. In a preferred mode, there are 7.

[0202] The first FMi card has the same resolution as the I(P) image n), and therefore comprises WxH points, where W represents the image width in pixels, and H its height. The second map, FM2, has half the resolution (in each dimension) of the FM1 map. Each additional map has half the resolution of the previous map. This structure reduces the number of points in the feature maps, which facilitates coding and learning while minimizing coding costs.

[0203] The FM2 map is oversampled by the SE module by a factor of 2 in each dimension, according to a procedure illustrated in Figure 6. The FM3 map is oversampled by a factor of 4 in each dimension, and the FM4 map by a factor of 8 in each dimension. The FMi map is not affected by the oversampling. (FMSi=FMi).

[0204] The resulting FMSi maps have the same resolution as the I(P) image. n), and therefore each have WxH values, where W represents the width of the image in pixels and H its height (N=WxH).

[0205] Other types of structure are possible, for example one can use a different reduction rate than half between cards (a quarter, or a third, etc.).

[0206] In this embodiment, the vector Z n is a 4-tuple (Z1...Z4) consisting of the values ​​extracted from the FMSi maps located at the coordinates (x n , y n ) of the current pixel P n The vector Z n consisting of the extracted (quantified) values ​​from the FMSi maps is processed by the NNSYN synthesis neural network to generate a second output vector; in this example, the output vector is the triplet (R, G, B) of pixel P'. n encoded then decoded. The triplet is inserted into the decoded image I(P' n ) at positions (x n , y n ) of the color components (R', G', B').

[0207] In another embodiment, not shown, the vector Z n is extracted directly from the FMi layers, with positions recalculated according to map size, then the extracted values ​​are possibly processed and quantified after extraction.

[0208] Figure 7 is a flowchart representing an example of a process for decoding a zone (Zod) j ) of a signal (I(P n )) which can be implemented by the decoding device in Figure 2 when the synthesis neural network NNSYN' is, for example, the one shown in Figure 3.

[0209] During an F20 step, the Bl, B2' and optionally B3' streams are extracted from the encoded stream. They contain respectively the Wck parameters, the EcJ coded representations of the points of the extended characteristic zones ZCD' and the MZD'c and Rec coded values.

[0210] In the embodiment illustrated in Figure 7, during step F21, the second rule and the segmentation information used to identify the expanded signal area (e.g., the expanded mask MZD' of the Zod' area) are generated by decoding the encoded values ​​Rec and MZD'c. Then, segmentation information used to identify the signal area (e.g., a mask MZod') is generated by applying the second rule to the segmentation information used to identify the expanded signal area (e.g., the expanded mask MZD'). In the case where the segmentation information used to identify the signal area or the expanded signal area is not encoded in the B3 stream J It can be obtained from decoder parameters (known size or position of slices or tiles, etc.). The first and / or second rule may be known to the decoder beforehand, or determined by the decoder.

[0211] During an F22 step, FMdJ maps are optionally initialized (for example with mile values) and an expanded characteristic area ZCD' comprising all points of the FMdJ maps co-located (possibly taking into account a scale factor) with samples belonging to the expanded mask MZD' of the Zod' area is obtained.

[0212] During step F23, the M FMdJ maps are generated by decoding the EcJ values ​​of the points in the extended characteristic zones ZCD'. In one embodiment, the FMdJ maps are decoded in the order (FMdi, FMdi, . . . FMd4), and the values ​​of each map in a predefined order, for example lexicographic.

[0213] According to embodiments such as those described for the encoder: the FMdJ cards have the same resolution as the I(P) signal n ) coded, that is to say they contain N=WxH values. FMdJ cards have a resolution less than or equal to that of the I(P) signaln ) coded. Several FMdJ cards have the same resolution, lower than the resolution of the I(P) signal n ) coded.

[0214] During a step F24, according to one embodiment, the M cards of the first group FMdJ are transformed by the module SE' to generate cards of the second group FMS'J at the resolution of the input images.

[0215] This step is similar to step E24, which was described for the encoder in support of Figure 5, and the embodiments apply. In particular: according to one embodiment, M FMS'J cards are generated; according to another embodiment, each FMdJ card is transformed into a card In one embodiment, at least one FMdJ map has a resolution lower than that of the images in the image to be encoded, and the transformation operation includes oversampling so that the transformed FMS'J map has the same number of samples as the input image. Oversampling consists of adding values ​​to the FMS'J maps to achieve the resolution of the input image. It can be simple (by nearest neighbor replication) or involve interpolation (linear, polynomial, filtered, etc.).

[0216] The transformation may optionally include inverse quantization of the extracted values, if necessary. However, inverse quantization is not mandatory.

[0217] During step F25, values ​​are extracted by the XTR' module from the FMdJ or possibly transformed FMS'J maps. This extraction is performed based on the coordinates (x n , y n) of a sample P n of the input signal. It can also be performed depending on the resolution of the card in question. This step is similar to step E25 which was described for The encoder shown in Figure 5, and the embodiments described above, apply. In particular:

[0218] According to one embodiment, the vector Zd n This characteristic results directly from this extraction.

[0219] In one embodiment, Zd n is an L-tuple (zi, Z2,..., Zj), consisting of the values ​​of the FMdJ or FMS'J maps located at the coordinates (x n , y n ) of a current pixel Pd n , as will be illustrated in Figure 8. In another embodiment Zd n is an L-tuple (zi, Z2,. . ., zj), consisting of the values ​​located in a neighborhood around these coordinates.

[0220] For example, the samples to be decoded are processed in sequential order.

[0221] According to one embodiment, during a step F26, a vector Zd n is constructed by the TT' module from the values ​​extracted from the FMd'J maps of the first group or the FM S' maps of the second group, for each sample Pd n coordinates (x n , y n ) to decode, based on the coordinates (x n , y n This step is similar to step E26, which was described for the encoder in support of Figure 5, and the described embodiments apply. The extraction may involve inverse quantization of the extracted values ​​or of the vector Zd n constituted, if necessary.

[0222] During step F27, the Wdk parameters of the NNSYN' synthesis neural network are generated by decoding the Wck values ​​of the Bl stream. Any known decoding technique corresponding to the encoding technique used by the encoder can be used for this purpose. The NNSYN' synthesis neural network is similar to the NNSYN synthesis network; that is, it has the same structure and parameters, except for the encoding, which can be performed with or without loss.

[0223] According to one embodiment, the B2 stream is decoded before the Bl stream, in order to have the NNSYN' synthesis neural network available before starting to decode the samples.

[0224] During step F28, the vector Zd n is processed by the NNSYN' synthesis neural network to generate as output a second vector representative of the sample Pd nto decode, according to one embodiment, a triplet that is injected into the intermediate image Ii(Pd n ) at positions (x n , y n ) of the color components (Rd, Gd, Bd). This step is similar to step E27 which was described for the encoder in support of Figure 5.

[0225] During an F29 step, the intermediate image L(P n ) is transformed into a signal corresponding to the Zod zone J of the image I(Pd n ) by the EXTR module.

[0226] Alternatively, steps F28 and F29 described above are replaced by a single step F30. During step F30, the NNSYN' module only processes the Zd vectors. n input capable of generating at output a second vector representative of a sample Pd n This needs decoding. selection of the Zd vectors n is performed by the NNSYN' module taking into account the MZod> mask of the area to be decoded.

[0227] Figure 8 illustrates a method for decoding an I(Pd) signal. n ) used in an embodiment of the invention.

[0228] In this embodiment, there are 4 FMdJ decoded cards. In a preferred mode, there are 7.

[0229] In this embodiment, the first map FMd^ has the same resolution as image I, and therefore comprises WxH points, where W represents the image width in pixels, and H its height. The second map FMd2 has half the resolution (in each dimension) of map FMd^. Each additional map has half the resolution of the preceding map. This structure reduces the number of variables in the feature maps, which facilitates decoding while minimizing the encoding cost.

[0230] The FMd2 map is oversampled by a factor of 2 in each dimension, using any oversampling method available to a person skilled in the art. The FMd map J 3est is oversampled by a factor of 4 in each dimension, and the FMd^ map by a factor of 8 in each dimension.

[0231] The FMS'J cards have the same resolution as the image to be decoded, and therefore include WxH values, where W represents the width of the image in pixels, and H its height.

[0232] In this embodiment, the vector Zd n is a 4-tuple (Z1...Z4) consisting of the values ​​of the FMS'J cards located at the coordinates (x n , y n ) of the current pixel Pd n The vector Zd n is optionally dequantized and then processed by the NNSYN' synthesis neural network (and optionally by the EXTR module) to generate the output triplet (R, G, B) representative of the Pd sample nto be decoded. The triplet (R, G, B) is inserted into the decoded Zod image j (Pd n ) at the coordinates (x n , y n ) in the color components (Rd, Gd, Bd).

[0233] Figure 9 is a flowchart representing a method for coding feature maps which can be implemented by the coding device of Figure 1 and by the coding process of Figure 5.

[0234] These steps are substeps of step E23 described earlier, supported by Figure 5. Their purpose is to encode a current value V n from a point in an expanded characteristic area ZCD> of a feature map FMdJ of the first group being processed using neighborhood values.

[0235] During a substep E231, a neighborhood vector (C n ) is established, including values ​​close to the value V nThese neighboring values ​​can be located on the same map. or / and in a map different from the plurality M of FMdJ maps. This neighborhood vector consists of a number C of values, or data, corresponding to neighborhood values ​​(for example, C=10). These values ​​must be known to the encoder and the decoder; therefore, they must lie in a causal neighborhood of the value V n .

[0236] According to a first embodiment, these values ​​are used to determine the context of an entropy encoder to encode the current value during an E233 step. This encoder can be a CAB AC (Context-Adaptive Binary Arithmetic Coding) encoder. This type of encoder is well known to those skilled in the art. It is notably used in the H.265 / HEVC video compression standard. It is an arithmetic encoder with lossless compression. It decomposes all non-binary symbols into binary symbols. Then, for each bit, the encoder selects the most suitable probability model and uses a context to optimize the probability estimation. This context can be defined by information from neighboring elements. Adaptive or non-adaptive arithmetic coding is then applied to compress the resulting data. As is known to those skilled in the art, there are several ways to use the neighborhood vector to generate context information.For example, one can count the number of neighboring values ​​that are not zero and associate a context with each number. Alternatively, one can perform comparisons between several neighboring values ​​and associate a given context with a configuration of order between the neighboring values, for example by ranking the neighboring values ​​in ascending order and associating a context with each possible order.

[0237] In a second embodiment, the neighborhood is used to predict, in step E232, the current value from an autoregressive model. Recall that an autoregressive model predicts a sample from a series based on its past values. In this embodiment, the past values ​​are constituted by the context, and the difference between the predicted variable and the actual value is quantified and then encoded entropically in step E233.

[0238] At the end of the process, the current coded value Vc nThe FMdJ card currently being processed is coded.

[0239] Figure 10 is a flowchart representing a method for decoding feature cards that can be implemented by the decoding device in Figure 2 and by the decoding process in Figure 7.

[0240] These steps are sub-steps of step F23 described previously in support of figure 7. They aim to decode a current value of a point of an extended characteristic area ZCD> of a feature map FMdJ of the first group being processed using neighborhood values.

[0241] During a substep F231, a neighborhood vector (Cd n ) is established, including values ​​close to the value Vd nThis step is similar to step E231 described earlier, and the same embodiments apply. This neighborhood vector consists of a number C of values, or data, corresponding to neighborhood values ​​(e.g., C=10) located in the same map and / or in a different map from the plurality M of maps of a point in an expanded characteristic area ZCD J of an FMdj characteristic map. These values ​​are located in a causal neighborhood of the Vd value. n are known to the decoder.

[0242] According to a first embodiment, these values ​​are used to determine the context of an entropy decoder for decoding the current value during an F233 step. This decoding is similar to that used in the encoder, for example, CABAC. The use of the neighborhood to produce context information is similar to that chosen for the encoder. For example, one can count the number of non-zero neighboring values ​​and associate a context with each number. Alternatively, one can perform comparisons between several neighboring values ​​and associate a given context with an ordering configuration among the neighboring values, for example, by sorting the neighboring values ​​in ascending order and associating a context with each possible order.

[0243] In a second embodiment, the neighborhood is used to predict the current value from an autoregressive model during step F232. In this mode, the past values ​​are constituted by the context, and the difference between the predicted variable and the actual value is quantified and then entropically coded during step F233.

[0244] At the end of the process, the current decoded value Vd n The FMdi card being processed is decoded.

[0245] Figure 11 schematically represents a TRANSC transcoding device for a coded representation of a signal in the form of an independent J-stream B2> and optionally B3 J in a single stream B0.

[0246] The TRANSC transcoding device in Figure 11 receives as input the J data streams B2 J and optionally B3 J corresponding to the J zones Zo> segmenting the signal I(P n ).

[0247] In one embodiment, this TRANSC transcoding device includes a DDMZ module capable of decoding segmentation information to identify an expanded area of ​​the signal, for example an expanded MZD mask J , a DZCD module capable of determining an extended characteristic zone ZCD J in the M feature cards, a DE module for decoding feature cards from the encoded data group Ec, a COMP module for composing FMdj feature cards and an FMC module for encoding FMdj feature cards.

[0248] With the exception of the COMP module, the other modules of the TRANSC transcoding device correspond to the modules of the decoding device in Figure 2, referenced in a similar way, and are not described again here.

[0249] The COMP module receives as input the J sets of FMdj characteristic maps associated with the Zo> zones of the I signal, as well as the associated extended characteristic zones (these extended characteristic zones are provided by the DZCD module). From this data, the COMP module will construct a set of FMdj characteristic maps by assigning, to each point in a region of an FMdj characteristic map co-located with an extended characteristic zone, the value of the corresponding point in the FMdj map.

[0250] The cards are then coded by the FMC module to produce the FMci data that makes up the B0 flow.

[0251] It should also be noted that the invention is not limited to the embodiments described above. Indeed, it will be apparent to those skilled in the art that various modifications can be made to the embodiments described above, in light of the information just disclosed to them.

[0252] For example, NNSYN / NNSYN' synthetic neural networks can be recurrent neural networks.

[0253] In another example, synthetic neural networks can consist of one or more convolutional neural networks, followed by a machine learning program (MLP), and then followed by one or more convolutional neural networks, each convolutional neural network being associated with a convolutional kernel of predefined size, for example, 3x3. In these examples, obtaining the Z vectors n / Zd n is adapted to the topology of NNSYN / NNSYN' synthesis neural networks, meaning that multiple values ​​are extracted from at least one FM feature map; In other words, Z nis an L-tuple with L>M, constructed from values ​​taken from the FM maps; at coordinates close to those of the current pixel, for example on the support of the kernel of the convolution filter of one of the convolutional neural networks.

[0254] For example, the encoding and decoding processes described earlier can use a wavelet transform or a discrete cosine transform instead of a neural coding method. In this case, without loss of generality, the NNSYN encoding neural network can be replaced by a direct transform followed by a quantizer, and the synthesis-to-decoder network can take the form, for example, of an (inverse) quantizer and an inverse transform followed by filtering (at the block contours). The transform is, for example, applied to a block of pixels. The inverse transform is for example applied to a characteristic vector of values ​​from the characteristic value maps corresponding to a block.

[0255] Other implementation methods of the coding process are presented below.

[0256] In one embodiment, the method of encoding a zone (Zo>) of a signal (I(P n )), said coding area, said coding area comprising a plurality of samples (P n ) to be coded, involves the following steps: - obtaining (E32) a group of at least one characteristic map (FMi) representative of said signal (I(P n )), - obtaining (E32) a mask (MZo>), called the first mask, in said signal (I(P n )) of said zone to be coded (Zo>), - obtaining E(33) a dilation rule (Rd) of said first mask (MZo>), - obtaining (E34) a dilated mask (MZD j), said second mask, by application of said dilation rule (Rd) to said first mask (MZo>), - the determination (E35) of a characteristic dilated zone (CDZ) j ) in said group of at least one feature card (FMi), depending on said second mask (MZI), - the coding (E36) of the value of the points of said dilated characteristic zone (ZCD) j ), - the determination of an erosion rule (Re) of said second mask (MZD j ), And - the coding (E37, E39) of said second mask (MZD>) and of said erosion rule (Re).

[0257] Other implementation methods of the decoding process are presented below.

[0258] In one embodiment, the process of decoding a zone (Zod j ) of a signal (I(Pd n )), said decoding area, said decoding area comprising a plurality of samples (Pd n) to decode, said decoding process comprising the following steps: - the decoding (F21) of a first mask (MZD j ) representative of an area within the domain of said signal (I(Pd n )), - the decoding of a group of at least one feature map (FMDj) representative of at least said area to be decoded, said decoding comprising: the determination (F22) of an expanded feature area (ZCD) j ) in a group of at least one feature map (FMDj) based on said first mask (MZD i ), and the decoding (F23) of the value of the points of said dilated characteristic zone (ZCDi), the decoding (F21) of an erosion rule (Re), and - the synthesis (F24-F29, F30) of said zone to be decoded (Zod>) from the points of said dilated characteristic zone (ZCD) j ) and the erosion rule (Re) to obtain the decoded signal (Zod j ).

[0259] The decoding process (and symmetrically the encoding process) may further include one or more of the following optional features, taken individually or in any technically possible combination.

[0260] Thus, according to a particular embodiment, the synthesis (F24-F29) includes the determination (F21) of a mask (MZod) j ) of the area to be decoded in the domain of said signal, called second mask, by application of said erosion rule (Re) said first mask (MZD j ).

[0261] According to a particular embodiment, the synthesis (F24-F29) comprises: - the processing (F28) of said dilated characteristic zone IZCD') to obtain an intermediate signal (Ii(P n )), And - F application (F29) of said second mask (MZod) j ) intermediate signal audit (Ii(P n )) to obtain said plurality of samples (Pd n) of said zone to be decoded (Zod j ).

[0262] According to a particular embodiment, the synthesis includes the treatment (F24-F27, F30) of said dilated characteristic zone (DCD) j ) taking into account said second mask (MZod>) to obtain the decoded signal (Zod>) corresponding to the second mask (MZodJ).

[0263] According to a particular embodiment, said erosion rule (Re) is a morphological erosion by a structuring element (ES).

[0264] According to a particular embodiment, said erosion rule (Re) is defined by a maximum distance, the second mask (MZod>) comprising the points of the first mask (MZD j ) located at a distance from one edge of the first mask (MZD j ) in at least one direction towards the inside of the first mask (MZD j ) greater than said maximum distance.

[0265] According to a particular embodiment, said erosion rule (Re) is defined by a number of parameters equal to twice the dimension of said signal (I(Pd n )), each direction of one of the dimensions of said signal being associated with a parameter, said second mask (MZod J ) being obtained by removing from said first mask (MZI), in at least one direction of a dimension of said signal, a number of points equal to said parameter associated with said at least one direction of a dimension of said signal.

[0266] According to a particular embodiment, the points of said dilated characteristic area are representative of said area to be decoded in a latent domain, in which said decoding process further comprises the decoding of the parameters (Wdk) of a neural network (NNSYN'), called a synthesis neural network, and in which the synthesis comprises, for at least one sample, called the current sample (Pd n), of the signal to be decoded, associated with a position (x n , y n ) in the signal to be decoded: - the construction of a characteristic vector (Zd n ) from said characteristic values ​​as a function of said position (xn, y n ) of said current sample, and - the processing of said characteristic vector (Zd n ) by the synthetic neural network (NNSYN') defined by the decoded parameters (Wdk) to provide a decoded value of the current sample (Pd n ).

[0267] According to a particular embodiment, said characteristic values ​​are representative of said area to be decoded in a transformed domain, for example associated with a direct discrete cosine transform or a wavelet transform and in which the synthesis includes the application of said characteristic values ​​of an inverse transform associated with said direct transform.

[0268] In the detailed presentation of the invention given above, the terms used shall not be interpreted as limiting the invention to the embodiments set forth in this description, but shall be interpreted as including all equivalents which can be foreseen by a person skilled in the art by applying their general knowledge to the implementation of the teaching which has just been disclosed to them.

Claims

Demands [1] Method for encoding a zone (Zo>) of a signal (I(P n segmented into a plurality of zones, said coding process comprising: - obtaining (E32) at least one characteristic map (FMi) representative of said signal (I(P n )), - the determination (E35) of an extended characteristic area IZCD') in said at least one characteristic map (FMi), corresponding to an extended area of ​​said signal including said signal area, - the coding (E36) of the value of the points of said extended characteristic zone (ZCD'). [2] Method for decoding a data stream representative of an area (Zod>) of a signal (I(Pd n segmented into a plurality of zones, said decoding process comprising: the decoding of at least one characteristic map (FMDj) representative of at least said zone of the signal, comprising: • the determination (F22) of an extended characteristic zone (ECZ) j ) in said at least one feature map (FMDj), corresponding to an expanded signal area including said signal area, and • the decoding (F23) of the data associated with the points of said extended characteristic zone (ZCD'), and the synthesis (F24-F29, F30) of said signal zone (Zod j ) from the decoded data. [3] Decoding method according to claim 2, further comprising obtaining at least one segmentation information identifying said enlarged area of ​​the signal. [4] A decoding method according to the preceding claim, wherein the synthesis (F24-F29) comprises the determination (F21) of at least one segmentation information identifying said signal zone (MZod) j ), by applying a second rule (Re) to said at least one segmentation information identifying said expanded signal area (MZD j). [5] Decoding method according to the preceding claim wherein the synthesis (F24-F29) comprises: the processing (F28) of said enlarged characteristic zone (CDZ) j ) to obtain an intermediate signal (L(P n )), and the application (F29) of said at least one segmentation information identifying said signal zone (MZod>) said intermediate signal (Ii(P n )) to obtain the samples (Pd n ) of said signal zone (Zod j ). [6] Decoding method according to claim 4 wherein the synthesis comprises the processing (F24-F27, F30) of said extended characteristic zone (ECZ) j taking into account said at least one segmentation information identifying said signal zone (MZod) j ) to obtain the samples (Pd n ) of said signal zone (Zod>). [7] Decoding method according to any one of claims 4 to 6 wherein said second rule (Re) is a morphological erosion by a structuring element (ES). [8] A decoding method according to any one of claims 4 to 6, wherein said at least one segmentation information identifying said enlarged signal area is a first mask (MZI) and said at least one segmentation information identifying said signal area is a second mask (MZod) j ). [9] Decoding method according to claim 8 wherein said second rule (Re) is defined by a maximum distance, the second mask (MZod j ) including the points of the first mask (MZI) located at a distance from an edge of the first mask (MZD j ) in at least one direction towards the inside of the first mask (MZI) greater than said maximum distance. [10] A decoding method according to claim 8, wherein said second rule (Re) is defined by a number of parameters equal to twice the dimension of said signal (I(Pd n )), each direction of one of the dimensions of said signal being associated with a parameter, said second mask (MZod j ) being obtained by removing said first mask (MZD j ), in at least one direction of a dimension of said signal, a number of points equal to said parameter associated with said at least one direction of a dimension of said signal. [11] A decoding method according to any one of claims 2 to 10, wherein said at least one characteristic map is representative of said signal (I(Pd n )) in a latent domain, wherein said decoding method further comprises decoding the parameters (Wdk) of a synthetic neural network (NNSYN'), and wherein the synthesis comprises, for at least one sample (Pd n) of the signal, associated with a position (x n , y n ) in the signal: the construction of a characteristic vector (Zd n ) from said data decoded according to said position (xn, y n ) of said current sample, and the processing of said characteristic vector (Zd n ) by the neural network (NNSYN') defined by the decoded parameters (Wdk) to provide a representative sample vector (Pd n ). [12] A decoding method according to any one of claims 2 to 10, wherein said at least one characteristic map is representative of said signal (I(Pd n )) in a transformed domain, for example associated with a direct discrete cosine transform, a Fourier transform, or a wavelet transform, and in which the synthesis includes the application to said decoded data of an inverse transform associated with said direct transform. [13] Encoding device for a zone (Zo>) of a signal (I(P n )) segmented into a plurality of zones, characterized in that said coding device is configured to: - obtain (E32) at least one characteristic chart (FMi) representative of said signal (KPn)), - determine (E35) an extended characteristic zone (ZCD j ) in said at least one feature map (FMi), corresponding to an expanded signal area including said signal area, - the coding (E36) of data associated with the points of said extended characteristic zone (ZCD j ). [14] Device for decoding a data stream representative of an area (Zod j ) of a signal (I(Pd nsegmented into a plurality of zones, characterized in that said decoding device is configured to: decode at least one characteristic map (FMDj) representative of at least said signal zone, comprising: • determine (F22) an extended characteristic area 1ZCD') in said at least one characteristic map (FMDj), corresponding to an extended signal area including said signal area, and • decode (F23) data associated with points in said extended characteristic zone (ECZ) i ), and the synthesis (F24-F29, F30) of said signal zone (Zod j ) from the decoded data. [15] Computer program comprising instructions for carrying out the steps of an encoding process according to any one of claims 1 to 2 or of a decoding process according to any one of claims 3 to 12 when said program is executed by a computer.

Images