Method and device for encoding both a hdr picture and a sdr picture obtained from said hdr picture using color mapping functions

0001The present disclosure generally relates to picture / video coding and decoding.

0002This section is intended to introduce the reader to various aspects of the technology that may be relevant to the various aspects of this Principle described and / or claimed below. This discussion may be helpful in providing the reader with background information to facilitate a better understanding of the various aspects of this principle. Therefore, it should be understood that these statements should be read from this perspective, not as an endorsement of the prior art.

0003In the following, a picture contains an array of one or several samples (pixel values) of a particular picture / video format, which format includes all information about the pixel values of the picture (or video) and, for example, a picture. It specifies all the information that can be used by the display and / or any other device for visualizing and / or decoding (or video). The picture is at least one component of the shape of the array of the first sample, usually the luma (or luminance) component, and in some cases at least one of the shapes of the array of at least one other sample. Contains one other component, usually a color component. Alternatively, equivalently, the same information can also be represented by a set of arrays of color samples, such as the traditional three-color RGB representation.

0004The pixel value is represented by a vector of C values, where C is the number of components. Each value of the vector is represented by a few bits, which defines the maximum dynamic range of the pixel values.

0005A standard dynamic range picture (SDR (Standard-Dynamic-Range) picture) is a color picture represented by a limited dynamic range, which is usually measured with a power value of 2 or an f-stop. SDR pictures have a dynamic range of about 10f stops, also referred to below as dynamic, that is, a ratio of 1000 brightest to darkest pixels in the linear region, for example ITU-R BT.709 OETF (phototransfer function). ) (Rec.ITU-R BT.709-5, April 2002) or ITU-R BT.2020 OETF (Rec.ITU-R) Limited number of bits in the non-linear region (HDTV (High Definition Television System) and UHDTV (Ultra High Definition Television System)) by reducing dynamics using BT.2020-1, June 2014) ) Is most often 8 or 10). This limited non-linear representation does not allow correct rendering of small signal variations, especially in the dark and bright luminance ranges. In a high dynamic range picture (HDR (High-Dynamic-Range) picture), the signal is much more dynamic (up to 20f stop, the ratio of the brightest pixel to the darkest pixel is up to 1 million), and the signal covers the entire range. A new non-linear representation is needed to maintain the high accuracy of. In HDR pictures, the raw data is typically represented in a floating point format (32 or 16 bits per component, ie floating point or semi-floating point), with the most common format being the openEXR semi-floating point format (per RGB component). 16 bits, i.e. 48 bits per pixel), or an integer in long representation, usually at least 16 bits.

0006A color gamut is a particular complete set of colors. The most common usage refers to a set of colors that can be accurately represented in a given situation, for example, within a given color space or by a particular output device.

0007The color gamut may be defined by the RGB primary colors given in the CIE1931 color space chromaticity diagram shown in FIG. 1 and the white point.

0008It is common to define primary colors in the so-called CIE 1931 color space chromaticity diagram. This is a two-dimensional diagram (x, y) that defines color independently of the luminance component. And any color XYZ is projected on this figure using the following transformations: (Outside 1)<img id="000003" he="22" wi="144" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />The z = 1-xy component is also defined, but has no additional information.

0009In this figure, the color gamut is defined by a triangle whose vertices are a set of (x, y) coordinates of the three primary colors RGB. The white point W is another given (x, y) point belonging to the triangle, usually close to the center of the triangle.

0010A color volume is defined by a color space and a dynamic range of values represented in said color space.

0011For example, the color gamut is defined by the RGB ITU-R Recommendation BT.2020 color space for UHDTV. The older standard, ITU-R Recommendation BT.709, defines a smaller color gamut for HDTV. SDR officially defines a dynamic range of up to 100 knits (candela / square meter) for the color volume in which the data is encoded, but some display technologies can display brighter pixels.

0012As extensively described in A Review of RGB Color Spaces by Danny Pascale, changing the color gamut, that is, the transformation that maps the three primary colors and white points from one color gamut to another, is a linear RGB color space. Can be performed using a 3x3 matrix in. Also, the spatial change from XYZ to RGB is carried out by a 3x3 matrix. As a result, changing the color gamut can be performed by a 3x3 matrix, regardless of whether RGB or XYZ is the color space. For example, changing the color gamut from BT.2020 linear RGB to BT.709 XYZ can be done by a 3x3 matrix.

0013A high dynamic range picture (HDR picture) is a color picture represented by HDR dynamic whose brightness value is higher than that of the SDR picture dynamic.

0014HDR dynamic is not yet defined in the standard, but some expect a dynamic range of up to thousands of knits. For example, the HDR color volume is defined by the RGB BT.2020 color space, and the values represented in the RGB color space belong to the dynamic range of 0 to 4000 knits. Another example of an HDR color volume is defined by the RGB BT.2020 color space, where the values represented in the RGB color space belong to the dynamic range of 0 to 1000 knits.

0015Color grading of a picture (or video) is a process of changing / enhancing the color of a picture (or video). Color grading of a picture typically involves changing the color volume (color space and / or dynamic range) or color gamut of the picture. Therefore, two different color grading versions of the same picture are modified / enhanced according to the version of this picture whose values are represented in different color volumes (or color gamuts), or at least one of those colors according to different color grades. This is the version of the picture. This may include interaction with the user.

0016For example, in filmmaking, a three-color camera is used to capture a picture (video) into an RGB color value consisting of three components (red, green, and blue). The RGB color value depends on the characteristics (primary colors) of the three colors of the sensor.

0017An HDR colorgraded version of the captured picture (or video) is then obtained (using a particular theatrical grade) to obtain theatrical rendering. Typically, the value of the first color grading version of the captured picture (or video) is represented according to a standardized YUV format such as BT.2020, which defines the parameter values for UHDTV.

0018The YUV format typically applies a nonlinear function, the so-called photoelectric transfer function (OETF), to the linear RGB component to obtain the nonlinear component R'G'B', followed by a color transformation (usually a 3x3 matrix). ) Is applied to the obtained nonlinear R'G'B'component to obtain three component YUVs. The first component Y is the luminance component, and the two components U and V are the color difference (chrominance) components.

0019The Colorist then fine-tunes some color values, usually in collaboration with the cinematographer, to adjust the color values of the first color grading version of the captured picture (or video). It is carried out by fine-tuning and infusing artistic intent.

0020An SDR colorgraded version of the captured picture is also obtained to get a home release rendering (using a particular home Blu-ray Disc / DVD grade). Typically, the value of the second color grading version of the captured picture is the studio code of a standardized YUV format, eg, a standard digital television with a standard 4: 3 and widescreen 16: 9 aspect ratio. It is expressed according to ITU-R Recommendation BT.601 (Rec.601), which defines the conversion parameters, or ITU-R Recommendation BT.709, which defines the parameter values for high-definition television systems (HDTV).

0021Obtaining such an SDR color grading version of a captured picture is usually the color volume of the first color grading version of the captured picture (for example, RGB BT.2020 1000 knit modified by colorlist). It is provided to reduce the size so that the second colorgrading version of the captured picture belongs to the second color volume (eg RGB BT.709 1000 knit). It uses an automatic color mapping function (CMF) (for example, to map the RGB BT.2020 format to RGB BT.709), which is usually approximated by a 3D look-up table (also known as a 3D LUT). Step. Note that all YUV formats considered are characterized by primary color parameters that allow the definition of any RGB-to-YUV and YUV-to-RGB color mapping.

0022Colorists then work with the cinematographer to fine-tune / fine-tune some color values to the color values of the second color grading version of the captured picture for home release. It is carried out by infusing artistic intent into the color.

0023The problem to be resolved is to deliver both an HDR colorgraded version and an SDR colorgraded version of the captured picture (or video), i.e. compressed representing the colorgraded version of the captured picture (or video). A related SDR picture (or video) representing a colorgraded SDR version of the captured picture (or video) while delivering the HDR picture (or video) at the same time, for example for backward compatibility with a legacy SDR display. Video) is to be delivered. The associated SDR picture (or video) is sometimes referred to as an imposed SDR picture (video).

0024The direct solution is to simultaneously deliver both these HDR and SDR colorgrading pictures (or videos) on the delivery infrastructure. The disadvantage of this solution is that legacy infrastructure adapted to broadcast SDR pictures (or videos), such as HEVC Main 10 Profiles ("High Efficiency Video Coding", SERIES H: AUDIOVISUAL AND MULTIMEDIA SYSTEMS, Recommendation ITU- Compared to T H.265, Telecommunication Standardization Sector of ITU, October 2014), this is to substantially double the required bandwidth.

0025Using a legacy delivery infrastructure is a requirement for accelerating the emergence of HDR picture (or video) delivery. Also, the bit rate must be minimized while ensuring good quality for both HDR and SDR pictures (or video).

0026The following is a simplified overview of the present disclosure to provide a basic understanding of some aspects of the present disclosure. This overview is not an extensive overview of this disclosure. It does not identify the main or important elements of this disclosure. The following summary merely presents some aspects of the present disclosure in simplified form as a prelude to the more detailed description provided below.

0027This principle aims to remedy at least one of the drawbacks of the prior art, in a way that encodes both the HDR picture and the first SDR picture obtained from the HDR picture into at least one bitstream. There: --The step to get the second HDR picture from the HDR picture, --By applying a non-linear function that depends on the modulation value obtained from the brightness of the color picture to the brightness of the color picture so that the dynamic of the brightness component is reduced compared to the dynamic of the brightness of the color picture. Steps to get the brightness component and -It is a step to acquire two color difference components, --The step of getting at least one neutral color component by scaling each color component by a factor that depends on the luminance component, --With the step of obtaining the two color difference components from at least one intermediate color component With the steps to get by In the step to get by A second HDR picture is obtained by combining the luminance component and the two color difference components. Steps to get and --The steps to get the color mapping function that allows the color mapping of the second SDR picture to the color of the third SDR picture obtained from the first SDR picture, and --- Steps to encode the information representing the color mapping function into a bitstream, --With the steps to encode the 4th SDR picture from the 1st SDR picture into a bitstream It is done by the method.

0028According to an example of this principle, the third and fourth SDR pictures are the first SDR pictures.

0029According to an example of this principle, the fourth SDR picture is the first SDR picture and the third SDR picture is the encoded version of the first SDR picture.

0030According to an example of this principle, the third SDR picture is the first SDR picture and the fourth SDR picture is obtained by applying a color mapping function to the colors of the second SDR picture.

0031According to an example of this principle, the third SDR picture is a decoded version of the encoded first SDR picture, and the fourth SDR picture makes the color mapping function the color of the second SDR picture. Obtained by applying.

0032According to the other one of those aspects, the principle is a method of decoding an HDR picture from at least one bitstream: --The step of getting the first SDR picture decrypted from the fourth decrypted SDR picture obtained by decoding the bitstream, and --The step of getting the information representing the color mapping function by decoding the bitstream, and --The steps to obtain the decoded second SDR picture by applying the inverse of the color mapping function to the colors of the decoded third SDR picture obtained from the decoded first SDR picture, and --The step to get the decoded HDR picture from the decrypted second SDR picture, --The step of obtaining the first component by applying a non-linear function to the luminance component obtained from the bitstream so that the dynamic of the first component is increased compared to the dynamic of the luminance component. , --The step of obtaining at least one color component from the first component and the two color difference components obtained from the bitstream and the coefficient (L) depending on the luminance component. With the steps to get by With The decoded picture is obtained by combining the at least one color component. Regarding the method.

0033According to the other one of those aspects, the principle is a device that encodes both the HDR picture and the first SDR picture obtained from the HDR picture into at least one bitstream: --Getting a second HDR picture from an HDR picture, --By applying a non-linear function that depends on the modulation value obtained from the brightness of the color picture to the brightness of the color picture so that the dynamic of the brightness component is reduced compared to the dynamic of the brightness of the color picture. To get the brightness component and -To get two color difference components, --Getting at least one neutral color component by scaling each color component by a factor that depends on the brightness component, --Obtaining the two color difference components from at least one neutral color component To get by In getting by A second HDR picture is obtained by combining the luminance component and the two color difference components. To get and --Getting a color mapping function that allows you to map the color of the second SDR picture to the color of the third SDR picture obtained from the first SDR picture, --Encoding the information representing the color mapping function into a bitstream and --- Encoding the fourth SDR picture from the first SDR picture into a bitstream The present invention relates to a device comprising a processor configured to perform the above.

0034According to the other one of those aspects, the principle is a device that decodes an HDR picture from at least one bitstream. --Getting the decrypted first SDR picture from the decrypted fourth SDR picture obtained by decoding the bitstream, --Getting information that represents a color mapping function by decoding the bitstream, --Getting the decrypted second SDR picture by applying the inverse of the color mapping function to the color of the decoded third SDR picture obtained from the decoded first SDR picture, --Getting the decoded HDR picture from the decrypted second SDR picture, --To obtain the first component by applying a non-linear function to the luminance component obtained from the bitstream so that the dynamic of the first component is increased compared to the dynamic of the luminance component. , --To obtain at least one color component from the first component and the two color difference components obtained from the bitstream and the coefficient depending on the luminance component. To get by Equipped with a processor configured to do The decoded picture is obtained by combining the at least one color component. It is related to the device.

0035According to other aspects of those embodiments, the principle is a device comprising a processor configured to implement the above method and a program code instruction described above when the program is executed on a computer. Computer program products, including program code instructions for performing the steps of the above method, and at least a processor-readable medium containing the instructions for causing the processor to perform the steps of the above method, and program code instructions. It relates to a non-temporary storage medium that, when the program is executed on a computing device, holds instructions in the program code for performing the steps of the above method.

0036The particular nature of this disclosure, as well as other purposes, advantages, features and uses of this disclosure will become apparent from the description of the following examples in conjunction with the accompanying drawings.

0037The drawings show an embodiment of the present disclosure.

0038<figref num="1">It is a figure which shows the example of the CIE1931 color space chromaticity diagram.</figref><figref num="2">It is a block diagram of the step of the method for encoding both HDR picture and SDR picture by this principle.</figref><figref num="3">HDR picture I according to an example of this principle_HDRAnd SDR picture I_SDR1It is a figure of the step of the method for decoding.</figref><figref num="4">HDR picture I described in connection with Figure 2._HDRAnd the first SDR picture I_SDR1It is a figure of the step of an example of the method for encoding both of.</figref><figref num="5">It is a figure of the step of the method for encoding both the HDR picture and the SDR picture by one modification of FIG.</figref><figref num="6">HDR picture I described in connection with Figure 2._HDRAnd the first SDR picture I_SDR1It is a figure of the step of an example of the method for encoding both of.</figref><figref num="7">It is a figure of the step of the method for encoding both the HDR picture and the SDR picture by one modification of FIG.</figref><figref num="8a">It is a figure of the substep of step 210 according to the example of this principle.</figref><figref num="8b">It is a figure of the substep of step 210 according to the example of this principle.</figref><figref num="8c">It is a figure of the substep of step 210 according to the example of this principle.</figref><figref num="8d">It is a figure of the substep of step 210 according to the example of this principle.</figref><figref num="9">It is a figure of the step of the method for decoding both HDR picture and SDR picture by an example of this principle.</figref><figref num="10a">It is a figure of the substep of step 210 according to the example of this principle.</figref><figref num="10b">It is a figure of the substep of step 210 according to the example of this principle.</figref><figref num="10c">It is a figure of the substep of step 210 according to the example of this principle.</figref><figref num="11a">It is a figure of the step of the method of decoding an HDR picture and an SDR picture from at least one bit stream by an example of this principle.</figref><figref num="11b">It is a figure of the step of the method of decoding an HDR picture and an SDR picture from at least one bit stream by an example of this principle.</figref><figref num="11c">It is a figure of the step of the method of decoding an HDR picture and an SDR picture from at least one bit stream by an example of this principle.</figref><figref num="11d">It is a figure of the step of the method of decoding an HDR picture and an SDR picture from at least one bit stream by an example of this principle.</figref><figref num="12">It is a figure which shows an example of the architecture of the device by an example of this principle.</figref><figref num="13">It is a figure which shows two remote devices which communicate with each other through a communication network by an example of this principle.</figref>

0039Similar or identical elements are referred to by the same reference number.

0040The present disclosure is fully described below with reference to the accompanying drawings showing examples of the present disclosure. However, this disclosure may be embodied in a number of alternative forms and should not be construed as being limited to the examples described herein. Accordingly, various modifications and alternative forms of the present disclosure are possible, but specific embodiments thereof are shown as examples of drawings and are described in detail herein. However, this disclosure is not intended to be limited to the particular form disclosed, and conversely this disclosure is all modifications and equalities that fall within the spirit and scope of the disclosure as defined by the claims. Please understand that it covers things and alternatives.

0041The terms used herein are for illustration purposes only and are not intended to limit this disclosure. As used herein, the singular forms "a", "an" and "the" shall also include the plural, unless the context explicitly states otherwise. As used herein, the terms "provide," "provide," "include," and / or "include" are the features, integers, steps, actions, elements, and / or described. It will be further understood that it specifies the existence of a component, but does not preclude the existence or addition of one or more other features, integers, steps, behaviors, elements, components, and / or groups thereof. Furthermore, when one element is referred to as "responsive" or "connected" to another element, it may respond to or be connected directly to the other element, or there may be intervening elements. In contrast, when one element is referred to as "directly responding" or "directly connected" to another, there are no intervening elements. As used herein, the term "and / or" includes any and all combinations of one or more of the related listed items and may be abbreviated as "/".

0042It will be appreciated that terms such as first, second, etc. may be used herein to describe various elements, but these elements should not be limited by these terms. These terms are used only to distinguish one element from another. For example, without departing from the teachings of the present disclosure, the first element may be referred to as the second element, and similarly the second element may be referred to as the first element.

0043Some of the figures include arrows on the communication path to indicate the main direction of communication, but it should be understood that communication can occur in the opposite direction of the arrows shown.

0044Some embodiments relate to block diagrams and operating flow charts in which each block represents a circuit element, module, or code portion containing one or more executable instructions for implementing a specified logical function. Explained. It should also be noted that in other embodiments, the functions described within the block may occur out of the order described. For example, two blocks shown in succession can actually be executed at substantially the same time, or the blocks can be executed in reverse order depending on the functionality involved.

0045References herein to "one example," "one example," or "an example" include at least one implementation of the present disclosure of a particular feature, structure, or property described in connection with an example or example. Means that it can be included in. The appearance of the phrases "in one embodiment", "according to one embodiment", "in one example", or "according to one example" in various places herein is not necessarily all in the same embodiment or. It does not refer to an example, nor is it a separate or alternative example or example that is necessarily mutually exclusive with other examples or examples.

0046The reference numbers appearing in the claims are for illustration purposes only and do not limit the claims.

0047Although not explicitly stated, the examples and modifications can be used in any combination or subcombination.

0048Although this principle describes the coding / decoding of pictures, it extends to the coding / decoding of sequences (videos) of pictures, because each picture in the sequence is sequentially coded as described below. This is because it is converted / decrypted.

0049Figure 2 shows the HDR picture I based on this principle._HDRAnd SDR picture I_SDR1A diagram of the steps of the method for encoding both of the above is shown.

0050HDR picture I_HDRIs a colorgraded version of the captured picture (or video) according to the first grade, the first SDR picture I_SDR1Is a color grading version of the captured picture (or video) according to the second grade described above. This constraint on the encoding method is to preserve the artist's intent, SDR Picture I_SDR1Color grade must be rendered by the decoder, or at least the SDR picture is SDR picture I_SDR1To have a visual content that is very close to the visual content of.

0051In step 210, the module TM is the HDR picture I._HDRTone-map the second SDR picture I_SDR2To get.

0052The term "tone mapping" refers to HDR picture I_HDRMeans any method of reducing the dynamic range of. Examples of tone mapping methods are given in FIGS. 8a-d, 9 and 10a-d, but the present disclosure is not limited to a particular tone mapping method.

0053In step 220, module SDR1-to-SDR3 is the first SDR picture I._SDR1From the third SDR picture I_SDR3To get.

0054In step 230, the module CM is the second SDR picture I._SDR2And the third SDR picture I_SDR3Third SDR picture I that minimizes the difference with_SDR3Second SDR picture I to color_SDR2Gets the color mapping function CMF that enables the color mapping of.

0055For example, the color mapping function is SDR Picture I_SDR23rd SDR picture I from the pixel of_SDR3Obtained by minimizing the mean square error calculated by subtracting the pixel value of. An example of a color mapping function is given by the HEVC standard using the color remapping information SEI message (Appendix D.2.32). The present disclosure is not limited to a particular color mapping function, but extends to any type of mapping function.

0056In step 240, the encoder ENC1 encodes the information INF representing the color mapping function CMF into the bitstream F1.

0057According to one embodiment of the method, the information INF is an index that allows the color mapping function CMF to be retrieved from the list of color mapping functions.

0058According to one embodiment of the method, the information INF represents a parameter of the color mapping function CMF.

0059In step 250, module SDR1-to-SDR4 is the first SDR picture I._SDR1From 4th SDR Picture I_SDR4To get.

0060In step 260, the encoder ENC2 has a fourth SDR picture I._SDR4Is encoded in the bitstream F2.

0061Figure 3 shows an HDR picture I based on an example of this principle._HDRAnd SDR picture I_SDR1The figure of the step of the method for decoding is shown.

0062In step 310, the decoder DEC2 decodes the fourth SDR picture I, which is decoded by decoding the bitstream F2._SDR4Gets the decrypted SDR picture called.

0063In step 320, module SDR4-to-SDR1 is the decrypted fourth SDR picture I._SDR4First SDR picture I decoded from_SDR1To get.

0064In step 220, module SDR1-to-SDR3 is the decrypted first SDR picture I._SDR1Third SDR picture I decoded from_SDR3To get.

0065In step 330, the decoder DEC1 acquires the information INF representing the color mapping function CMF by decoding the bitstream F1 at least partially.

0066According to one variant, the information INF represents the inverse of the color mapping function CMF.

0067In step 340, the module AP^-1Is the decrypted third SDR picture I_SDR3Color mapping function to the color of CMF Inverse function of CMF CMF^-1Second SDR picture I decrypted by applying_SDR2To get.

0068In step 350, the module ITM is the decrypted second SDR picture I_SDR2HDR picture I decoded by applying reverse tone mapping to_HDRTo get.

0069Inverse tone mapping is the inverse function of tone mapping used in step 210 of FIG.

0070FIG. 4 is an HDR picture I described in connection with FIG._HDRAnd the first SDR picture I_SDR1A diagram of the steps of an example of a method for encoding both of the above is shown.

0071Modules SDR1-to-SDR3 and SDR1-to-SDR4 are SDR Picture I_SDR3And I_SDR4Is SDR picture I_SDR1Is configured to be equal to.

0072In other words, those modules do not implement any method.

0073In step 230, the color mapping function CMF is then acquired and the first SDR picture I_SDR1Second SDR picture I to color_SDR2Color mapping is possible, and in step 260, the first SDR picture I_SDR1Is directly encoded by the encoder ENC2.

0074Therefore, according to this example of this principle, the first SDR picture I colorgraded by the colorist._SDR1Is made available directly by decoding the bitstream F2. The first SDR picture I decrypted in this way_SDR1Is displayed, the artist's intent is saved.

0075FIG. 5 shows an HDR picture I according to a modification of FIG._HDRAnd SDR picture I_SDR1A diagram of the steps of the method for encoding both of the above is shown.

0076Module SDR1-to-SDR4 is the fourth SDR picture I_SDR4Is the first SDR picture I_SDR1It is configured to be. Therefore, the first SDR picture I colorgraded by the colorist_SDR1Is made available directly by decoding the bitstream F2. The first picture I decrypted in this way_SDR1Is displayed, the artist's intent is saved.

0077Module SDR1-to-SDR3 uses the encoder ENC2 to make the first SDR picture I_SDR1First SDR picture I encoded and encoded according to decoder DEC2_SDR1Third SDR picture I by decoding_SDR3Is configured to be obtained (step 310).

0078In step 230, the color mapping function CMF is then acquired and encoded in the first SDR picture I._SDR1SDR picture I to the color of the decrypted version of_SDR2Color mapping is possible.

00791st SDR picture I_SDR1First encoded SDR picture I, not from_SDR1Determining the color mapping function CMF from the decoded version of the second SDR picture I whose contents were used on the coding side_SDR2Decrypted second SDR picture I (obtained on the decoding side) closer to the content of_SDR2Bring. Then the decrypted second SDR picture I_SDR2And the decrypted second SDR picture I_SDR2The decoded HDR picture obtained from the color mapping function determined from has a visual content closer to the visual content of the original HDR picture on the coding side, and the HDR coding / decoding method of FIG. Performance is improved.

0080FIG. 6 is an HDR picture I described in connection with FIG._HDRAnd the first SDR picture I_SDR1A diagram of the steps of an example of a method for encoding both of the above is shown.

0081Module SDR1-to-SDR3 is SDR Picture I_SDR3Is SDR picture I_SDR1It is configured to be.

0082In step 230, the color mapping function CMF is then acquired and the first SDR picture I_SDR1Second SDR picture I to color_SDR2Color mapping is possible.

0083Module SDR1-to-SDR4 is the second SDR picture I_SDR2To the color of (SDR picture I_SDR1By applying the color mapping function CMF (obtained from), the fourth SDR picture I_SDR4It is provided with a module AP (step 610) for acquiring.

0084Therefore, the fourth SDR picture I_SDR4The content of the first SDR picture I_SDR1The reason is that the color mapping function CMF is determined to minimize the difference between these two pictures.

0085FIG. 7 shows an HDR picture I according to a modification of FIG._HDRAnd SDR picture I_SDR1A diagram of the steps of the method for encoding both of the above is shown.

0086Module SDR1-to-SDR3 uses the encoder ENC2 to make the first SDR picture I_SDR1(Step 260), the first SDR picture I encoded according to decoder DEC2_SDR1Third SDR picture I by decoding_SDR3Is configured to be obtained (step 310).

00871st SDR picture I_SDR1First SDR picture I encoded, not from_SDR1Determining the color mapping function CMF from the decoded version of the second SDR picture I whose contents were used on the coding side_SDR2Decrypted second SDR picture I (obtained on the decoding side) closer to the content of_SDR2Bring. Then the decrypted second SDR picture I_SDR2And the decrypted second SDR picture I_SDR2The decoded HDR picture obtained from the color mapping function determined from has a visual content closer to the visual content of the original HDR picture on the coding side, and the HDR coding / decoding method of FIG. Performance is improved.

0088According to an example of this principle, in step 210, the module TM is an HDR picture I._HDRHDR Picture I to reduce the dynamic range of the brightness to the target dynamic range_HDRApply the tone mapping operator to.

0089The present invention is not limited to any particular tone mapping operator. This single condition is that the tone mapping operator must be reversible.

0090For example, the tone mapping operators defined by Reinhard (Reinhard, E., Stark, M., Shirley, P., and Ferrerda, J., "Photographic tone reproduction for digital images", ACM Transactions on Graphics 21 (July 2002). Mon)), or Boatard, R., Bouatouch, K., Cozot, R., Thoreau, D., and Gruson, A. (2012), Temporal coherency for video tone mapping, AMJvan Eijk, CCDavis, SM Hammel, and Those defined by AKMajumdar (Eds.), Proc.SPIE8499, Applications of Digital Image Processing (p.84990D-84990D-10)) may be used.

00918a-d show a diagram of the sub-steps of step 210 according to an example of this principle.

0092As shown in Figure 8a, the module TM is an HDR picture I_HDRIt has a module BAM configured to get the backlight picture Ba from (step 2101).

0093According to one embodiment of step 2101 shown in Figure 8b, the module BAM is an HDR picture I._HDRIt is equipped with a module BI that acquires the backlight picture Ba from the brightness component L of.

0094HDR picture I_HDRIf is in the RGB color space, the luminance component L is obtained, for example, in the 709 color gamut by a linear combination given by L = 0.2127.R + 0.7152.G + 0.0722.B

0095According to one embodiment, the backlight picture Ba is a shape function given by the following equation ( shape functions) ψ_iDetermined as a weighted linear combination of: Ba = Σ_ia_iψ_i (1) Here a_iIs the weighting factor.

0096Therefore, determining the backlight picture Ba from the brightness component L is the optimum weighting factor (potentially, if it is unknown in advance, the optimum shape) so that the backlight picture Ba matches the brightness component L. Functions also include finding).

0097Weight coefficient a_iThere are many well-known methods for discovering. For example, the root mean square error method can be used to minimize the mean square error between the backlight picture Ba and the luminance component L.

0098The shape function can be the true physical response of the backlight of the display (for example, one made of LEDs, and each shape function corresponds to the response of one LED), or to the luminance component. Note that it can be a pure mathematical structure that best fits.

0099According to a variant of this example shown in Figure 8c, the module BAM is the HDR picture I acquired by the module HL._HDRAverage brightness value L_{mean mean}Further comprises a module BM that modulates the backlight picture Ba (given by Eq. (1)) using.

0100According to one example, the module HL has an average luminance value L over the total luminance component L._{mean mean}Is configured to calculate.

0101According to one example, the module HL has an average luminance value L according to the following equation._{mean mean}Is configured to calculate: (Outside 2)<img id="000004" he="14" wi="129" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Here, β is a coefficient less than 1, and E (X) is the mathematically expected value (average value) of the luminance component L.

0102This last example is HDR Picture I_HDRAverage brightness value L by some pixels with extremely high values, usually accompanied by very unpleasant time average brightness instability when_{mean mean}It is advantageous because it avoids being affected.

0103In the present invention, the average brightness value L_{mean mean}Is not limited to a specific embodiment for calculating.

0104According to the variant shown in FIG. 8d, module N normalizes the backlit image Ba (given by Eq. (1)) with its mean value E (Ba) for the HDR picture (or HDR picture). (For all HDR pictures if belongs to a sequence or group of pictures) Backlit picture Ba_grayGet (has a midgray equal to 1): (Outside 3)<img id="000005" he="20" wi="131" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0105Module BM then uses the following relationship to HDR picture I_HDRAverage brightness value L_{mean mean}Backlit picture Ba_grayIs configured to modulate: Ba_mod cst_mod.L_{mean mean}^α.Ba_gray (2) Here cst_modIs the modulation factor and α is any other modulation factor less than 1, typically 1/3. For example, cst about a backlit picture_mod1.7 is obtained by the minimum mean square.

0106In fact, due to linearity, all operations to modulate the backlight picture have a backlight factor a._iApplies to as a correction factor, which is the factor a_iThe new coefficient (Outside 4)<img id="000006" he="11" wi="124" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Is to convert to and get: (Outside 5)<img id="000007" he="17" wi="124" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0107The present disclosure is HDR Picture I_HDRIt is not limited to any method for obtaining the backlight picture Ba from.

0108In step 2102 of FIG. 8a, the second SDR picture I_SDR2Is HDR picture I for each pixel_HDRIs obtained by dividing by the backlight picture Ba.

0109In step 2103, the encoder ENC3 encodes the backlight picture Ba into the bitstream F3.

0110HDR Picture Dividing HDR by Backlit Picture Ba reduces the dynamic range of HDR pictures. Therefore, the method described in relation to Figures 8a-d is HDR Picture I._HDRCan be thought of as a tone mapping of.

0111FIG. 9 shows a diagram of the steps of a method for decoding both an HDR picture and an SDR picture according to an example of this principle.

0112This example makes it possible to retrieve both HDR and SDR pictures if they were pre-encoded by the methods described in connection with FIGS. 8a-d.

0113The module ITM includes a decoder DEC3 that obtains the decoded backlight picture Ba by decoding the bitstream F3 in step 350 (step 3501). In step 3502, the second SDR picture I_SDR2Decrypted HDR picture I by multiplying by the decoded backlight picture Ba_HDRIs obtained.

0114Second SDR picture I_SDR2Multiplying the decoded backlight picture Ba gives the resulting HDR picture the dynamic range of the second SDR picture I._SDR2Increased compared to, i.e., such multiplication can be considered an inverse tone mapping.

0115FIG. 10a-c shows a diagram of substeps of step 210 according to an example of this principle.

0116In this example, HDR Picture I_HDRIs HDR Picture I_HDRIt is considered to have three color components Ec (c = 1, 2 or 3) in which the pixel value of is expressed.

0117The present disclosure is not limited to any color space in which the three component Ec is represented, but extends to any color space, such as RGB, CIELUV, XYZ, CIELab, and the like.

0118Basically, the luminance component L and the two color difference components C1 and C2 are HDR picture I._HDRIt is determined from the three color components Ec. The luminance and color difference components form an SDR color picture in which the pixel values are represented within the color space (L, C1, C2). The SDR color picture is visible on a legacy SDR display, i.e., has sufficient visual quality to be seen on a legacy SDR display.

0119In step 100a, the module IC is HDR picture I_HDRThe component Y, which represents the brightness of, is obtained by linearly combining the three components Ec: (Outside 6)<img id="000008" he="18" wi="144" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Where A1 is the first row of the 3 × 3 matrix A that defines the color space transformation from the (E1, E2, E3) color space to the color space (Y, C1, C2).

0120In step 130a, the module BMM gets the module value Bm from component Y.

0121According to an example of step 130a, the modulation value Bm is the average, median, minimum or maximum value of the pixel values of component Y. These operations are for linear HDR luminance domain Y_linIn, or in the non-linear region, eg, ln (Y) or Y of γ &lt;1^γCan be carried out in.

0122In step 110a, the module FM obtains the luminance component L by applying the nonlinear function f to the component Y: L = f (Bm, Y) (3)

0123Applying the nonlinear function f to component Y reduces its dynamic range. In other words, the dynamics of the luminance component L are reduced as compared to the dynamics of the component Y.

0124Basically, the dynamic range of component Y is reduced so that the luminance value of component L is represented using 10 bits.

0125According to one embodiment, the component Y is divided by the modulation value Bm before applying the nonlinear function f: L = f (Y / Bm) (4)

0126According to one embodiment, the nonlinear function f is a gamma function: (Outside 7)<img id="000009" he="16" wi="127" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Here Y₁Is equal to Y or Y / Ba, depending on the embodiment of equation (3) or (4), B is a constant value, and γ is a parameter (strictly a real value less than 1).

0127According to one example, the nonlinear function f is the S-Log function: L = a.ln (Y₁+ b) + c Here, a, b, and c are determined to be continuous at 1 when f (0) and f (1) are invariant and the derivative of the SLog curve is extended by a gamma curve less than 1. It is a parameter (real value) of the SLog curve. Therefore, a, b and c are functions of the parameter γ. Typical values are shown in Table 1.

0128<tables num="1"><img id="000010" he="43" wi="127" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" /></tables>

0129In one advantageous embodiment, a value of γ close to 1 / 2.5 is efficient in terms of HDR compression performance as well as good visibility of the resulting SDR Luma. Therefore, the three parameters can advantageously take the following values: a = 0.444955114, b = 0.12123691, c = 0.994855684.

0130According to one example, the nonlinear function f is either gamma correction or SLog correction, depending on the pixel value of component Y.

0131Applying gamma correction to component Y raises dark areas, but does not reduce enough highlights to avoid saturation of bright pixels.

0132Then, according to one embodiment, the module FM applies either gamma correction or SLog correction depending on the pixel value of component Y. The information data Inf can indicate whether gamma correction or SLog correction is applied.

0133For example, if the pixel value of component Y is less than the threshold (equal to 1), gamma correction is applied, otherwise SLog correction is applied.

0134According to one example, when the method is used to encode several HDR pictures that belong to a sequence of pictures, the modulation value Bm is for each HDR picture, group of pictures (GOP), or HDR picture. Part of, for example, but not limited to, is determined for slices or transfer units defined in HEVC.

0135According to one embodiment, the value Bm and / or non- parameters of the linear function f (for example, a, b, c or gamma), and / or information data Inf is stored in local or remote memory, and / or Added to Bitstream F3.

0136In step 120a, at least one color component EC (c = 1, 2, 3) is HDR picture I_HDRObtained from. Color component Ec HDR picture I for color conversion, either directly from local or remote memory_HDRCan be obtained by applying to.

0137In step 140a, the neutral color component E'c (c = 1, 2 or 3) is obtained by scaling each color component Ec by a coefficient r (L) that depends on the luminance component L: (Outside 8)<img id="000011" he="24" wi="135" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Here, r (L (i)) is a coefficient (real value) determined by the module RM (step 150a), which depends on the value of the pixel i of the component L, and is E._c'(i) is the value of pixel i of the neutral color component E'c, and E_c(i) is the value of pixel i of the color component Ec.

0138Scaling by a coefficient means multiplying the coefficient or dividing by the reciprocal of the coefficient.

0139Scaling each color component Ec by a coefficient r (L) that depends on the luminance component L is an HDR picture I._HDRSave the hue of the color of.

0140According to an example of step 150a, the coefficient r (L) is the ratio of the luminance component L to the component Y: (Outside 9)<img id="000012" he="23" wi="133" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Here, Y (i) is the value of the pixel i of the component Y. In practice, the pixel value Y (i) of the component Y clearly depends on the pixel value L (i) of the luminance component L, so the ratio can be written as a function of L (i) only.

0141In this example, each color component Ec can be scaled by a coefficient r (L) that further depends on the component Y, HDR picture I._HDRIt is advantageous because it preserves the hue of the colors in the image and improves the visual quality of the decoded color picture.

0142More precisely, in color science and color theory, colorfulness, chroma, and saturation refer to the perceived intensity of a particular color. Color saturation is the degree of difference between color and gray. Saturation is the chromaticity relative to the brightness of other colors that appear white under similar observation conditions. Saturation is the color saturation of a color for its brightness.

0143Very colorful stimuli are vivid and intense, less colorful stimuli appear more subdued and closer to gray. If there is no color saturation, the color is "neutral" gray (a picture that has no color saturation in any of the colors is called grayscale). Any color can be described by its color saturation (or chromaticity or saturation), lightness (or brightness), and hue.

0144Definitions of hue and saturation of a color depend on the color space used to represent the color.

0145For example, if the CIELUV color space is used, the saturation s_uvIs the brightness L^*Saturation C for_uv^*Defined as a ratio of: (Outside 10)<img id="000013" he="22" wi="133" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0146And the hue is given by: (Outside 11)<img id="000014" he="19" wi="133" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0147According to another example, when the CIELAB color space is used, saturation is defined as the ratio of chromaticity to luminance: (Outside 12)<img id="000015" he="24" wi="133" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0148And the hue is given by: (Outside 13)<img id="000016" he="17" wi="128" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0149These equations are rational predictors of saturation and hue that are consistent with human perception of saturation, and determine the brightness in the CIELAB (or CIELUV) color space at angles a * / b * (or u * /). It is shown that adjusting while fixing v *) affects the perception of hue and thus the same color. In step 140a, scaling the color component Ec by the same coefficient preserves this angle and thus the hue.

0150Here, HDR picture I_HDRIs represented in the CIELUV color space, HDR Picture I_HDRBy combining the luminance component L of the CIELUV color space, whose dynamic range is reduced compared to the luminance dynamic range of (step 110a), and the two color difference components U (= C1) and V (= C2), 2 SDR picture I_SDR2Consider that is formed. Therefore, the second SDR picture I_SDR2Colors are perceived differently by humans because of the varying saturation and hue of the colors. The method described in connection with FIG. 10a is the second SDR picture I._SDR2Hue of color is HDR picture I_HDRSecond SDR Picture I to best match the hue of the color of_SDR2Determine the color difference components C1 and C2 of.

0151According to an example of step 150a, the coefficient r (L) is given by: (Outside 14)<img id="000017" he="23" wi="128" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0152This last embodiment is advantageous because it prevents the coefficients from going to zero for very dark pixels, i.e., the ratio can be inverted regardless of the pixel value.

0153In step 160a, the two color difference components C1 and C2 are obtained from the at least one neutral color component E'c.

0154According to one embodiment of step 160a shown in FIG. 10b, at least one intermediate component Dc (c = 1, 2 or 3) applies OETF to each intermediate color component (E'c) (step). Obtained by 161b): (Outside 15)<img id="000018" he="23" wi="128" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />For example, OETF is defined by ITU-R Recommendation BT.709 or BT.2020 and is described as: (Outside 16)<img id="000019" he="20" wi="130" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0155This embodiment allows for reduction of dynamic range in response to a particular OETF, but involves complex decoding processes as detailed below.

0156According to a variant of this example shown in Figure 10c, the OETF is approximated by a square root, i.e. at least one intermediate component Dc (c = 1, 2 or 3) is each intermediate color component (E'c). ) To take the square root (step 161c): (Outside 17)<img id="000020" he="31" wi="131" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0157This variant provides a good approximation of the OETF as defined by ITU-R Recommendation BT.709 or BT.2020 and is advantageous as it leads to low complexity decoders.

0158According to another variant, OETF is approximated by a cube root, i.e., obtained by having at least one intermediate component Dc (c = 1, 2 or 3) take the cube root of each intermediate color component (E'c). Be: (Outside 18)<img id="000021" he="28" wi="131" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0159This variant is advantageous because it provides a good approximation of the OETF as defined by ITU-R Recommendation BT.709 or BT.2020, but it is somewhat more than the decoder obtained if the OETF was approximated by the square root. It leads to more complicated decoders.

0160In step 162b, module LC1 obtains two color difference components C1 and C2 by linearly combining the three intermediate components Dc: (Outside 19)<img id="000022" he="23" wi="131" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Here, A2 and A3 are the second and third rows of the 3 × 3 matrix A.

0161In step 170a shown in FIG. 10a, the module COM combines the luminance component L with the color difference components C1 and C2 to create a second SDR picture I._SDR2To get.

016211a-d show a diagram of the steps of a method of decoding an HDR picture and an SDR picture from at least one bitstream according to an example of this principle.

0163In step 111a, the module DECOMB has a second SDR picture I._SDR2The luminance component L and the two color difference components C1 and C2 are obtained from.

0164In step 113a, the module IFM is a nonlinear function f.^-1Is applied to the luminance component L so that the dynamic of the first component Y is increased compared to the dynamic of the luminance component L to obtain the first component Y. Y = f^-1(Ba, L) (5)

0165Non-linear function f^-1Is the inverse of the nonlinear function f (step 110a).

0166Therefore, the function f^-1The example of is defined according to the example of the function f.

0167According to one example, the value Bm and / or the nonlinear function f^-1Parameters (eg, a, b, c or γ) and / or informational data Inf are obtained from local or remote memory (eg lookup table) and / or from the bitstream F3 shown in Figure 11a. ..

0168According to one embodiment, the luminance component L is a nonlinear function f.^-1After applying, the modulation value Bm is multiplied: Y = Bm * f^-1(L) (6)

0169According to one example, the nonlinear function f^-1Is the inverse of the gamma function.

0170And component Y is given by (Outside 20)<img id="000023" he="19" wi="128" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Here Y₁Is equal to Y or Y / Bm, depending on the embodiment of equation (5) or (6), B is a constant value, and γ is a parameter (strictly a real value less than 1).

0171According to one embodiment, the nonlinear function f^-1Is the inverse of the S-Log function. Component Y₁Is given by: (Outside 21)<img id="000024" he="20" wi="128" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0172According to one embodiment, the nonlinear function f is the inverse function of either gamma correction or SLog correction depending on the pixel value of component Y. This is indicated by the information data Inf.

0173In step 112a, the module ILC obtains at least one color component Ec from the first component Y, the two color difference components C1 and C2, and the luminance component L-dependent coefficient r (L). Then the decoded HDR picture I_HDRIs obtained by combining the at least one color component Ec.

0174The coefficient r (L) can be obtained from local or remote memory (eg, a look-up table), or a bitstream.

0175If a general OETF is applied to each neutral component E'c (step 161b in Figure 10b), the intermediate component Dc is associated with component Y, the two color difference components C1, C2, and the coefficient r (L). :: (Outside 22)<img id="000025" he="30" wi="146" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />as well as, (Outside 23)<img id="000026" he="18" wi="133" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Here, EOTF (light transfer function) is the inverse function of OETF applied in step 161b.

0176Equation (7b) gives: (Outside 24)<img id="000027" he="25" wi="124" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Here, OETF (E_c) = D_cAnd (Outside 25)<img id="000028" he="13" wi="124" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Is a constant according to the matrix A, L_iIs also a linear function according to the matrix A. And equation (7a) is: r (L) * Y = A₁₁EOTF (D₁) + A₁₂EOTF (D₂) + A₁₃EOTF (D₃) (9) And (Outside 26)<img id="000029" he="20" wi="146" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0177Equation (10) is D₁It is an implicit expression about only. Depending on the EOTF equation, equation (10) can be solved approximately simply. When solved, D₁Is obtained, D₂, D₃Is D by equation (8)₁Derived from. Then, by applying EOTF to the three obtained intermediate components Dc, the neutral color component E'c is obtained, that is, E'c = EOTF (Dc).

0178In this general case, that is, when the general OETF (which does not have any particular property) is applied to each neutral color component E'c, there is no analytical solution in equation (10). For example, if the OETF is ITU-R BT.709 / 2020 OETF, equation (10) is numerical using the so-called Newton's method, or any other numerical method for finding the roots of holomorphic functions. Can be solved by. However, this leads to very complex decoders.

0179In this general case, according to the first example of step 112a shown in FIG. 11b, in step 1121a, the module ILEC is the first component Y, the two color difference components, as described above. Obtain three neutral color components E'c from C1, C2, and the coefficient r (L). In step 1122a, the three color components Ec are obtained by scaling each neutral color component E'c by a factor r (L): Ec (i) = E'c (i) / r (L (i)) Here, r (L (i)) is a coefficient given by step 150a, which depends on the value of pixel i of the component L (output of step 111a), and is E._c'(i) is the value of pixel i of the neutral color component E'c, and E_c(i) is the value of pixel i of the color component Ec.

0180In practice, this order in which step 1121a precedes step 1122a is the reverse of the coding method (FIG. 10b), in which step 161b is followed by step 162b.

0181According to a variant of this first example, OEFT is a square root function and EOTF is a square function.

0182According to another variant of this first example, OEFT is a cube root function, and EOTF is a cubic function.

0183If the OETF used in step 161b meets the commutation criteria, i.e.: OETF (x * y) = OETF (x) * OETF (y) Component Y and color component Ec are associated by: (Outside 27)<img id="000030" he="23" wi="127" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Where Fc is a component equal to OETF (Ec), and (Outside 28)<img id="000031" he="20" wi="158" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />As a result, the commutation condition gives: (Outside 29)<img id="000032" he="20" wi="135" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0184Equation (11) gives: (Outside 30)<img id="000033" he="39" wi="168" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />here, (Outside 31)<img id="000034" he="18" wi="116" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Is a constant that depends on the matrix A, L_iIs a linear function that also depends on the matrix A.

0185And equation (11) is: Y = A₁₁EOTF (F₁) + A₁₂EOTF (F₂) + A₁₃EOTF (F₃) (13) And (Outside 32)<img id="000035" he="25" wi="149" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0186If the OETF satisfies the commutation condition, according to the second example of step 112a shown in FIG. 11c, in step 1121c, the two intermediate components C'1 and C'2 have the coefficients OEFT (r (L (L (L)). Obtained by scaling the two color difference components C1 and C2 by i))), where OETF is the function used in step 161b of Figure 10b: (Outside 33)<img id="000036" he="32" wi="125" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Here, r (L (i)) is a coefficient given by step 150a, which depends on the value of pixel i of the component L (output of step 111a), and is C.₁'(I c₂'(i) is the value of pixel i of components C'1 and C'2, respectively, and C₁(I c₂(i) is the value of pixel i of the components C1 and C2, respectively.

0187In step 1122c, the module ILEC obtains three color components Ec from the first component Y and the two intermediate color difference components C'1 and C'2 as described above.

0188According to a variant of this second example, OEFT is a square root function and EOTF is a square function. Then, in step 1122c, the two intermediate components C'1 and C'2 are coefficients. (Outside 34)<img id="000037" he="19" wi="123" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Obtained by scaling the two color difference components C1 and C2: (Outside 35)<img id="000038" he="34" wi="129" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0189Equation (11) is: (Outside 36)<img id="000039" he="24" wi="139" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />as well as, (Outside 37)<img id="000040" he="23" wi="132" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />As a result, the communication gives: (Outside 38)<img id="000041" he="24" wi="132" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0190Equation (14) is: (Outside 39)<img id="000042" he="23" wi="118" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />as well as, (Outside 40)<img id="000043" he="24" wi="139" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0191Equation (17) is a quadratic equation that can be solved analytically. This analytical solution leads to a specific embodiment of step 1122c shown in Figure 11d. This example is advantageous because it allows an analytical expression of the decoded component of the EOTF (the inverse function of the OETF), and thus the HDR picture. In addition, EOTF is then a square function, which is a low-complexity process on the decoding side.

0192In step 11221c, the module SM obtains the second component S by combining the two intermediate color difference components C'1, C'2, and the first component Y: (Outside 41)<img id="000044" he="20" wi="118" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Here k₀, K₁And k₂Is a parameter value (Outside 42)<img id="000045" he="20" wi="124" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Means the square of the component C'c (c = 1 or 2).

0193In step 11222c, module LC2 obtains three solver component Fc by linearly combining the intermediate color difference components C'1, C'2, and the second component S: (Outside 43)<img id="000046" he="22" wi="128" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Where C is a 3 × 3 matrix defined as the inverse of the matrix A.

0194In step 11223c, the three color components Ec are obtained by taking the square of each neutral color component (Dc): (Outside 44)<img id="000047" he="28" wi="128" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0195The matrix A is the HDR picture I to be coded from the color space (E1, E2, E3) to which the pixel value of the HDR picture to be coded is represented to the color space (Y, C1, C2)._HDRDetermine the conversion of.

0196Such a matrix is the HDR picture I to be encoded._HDRDepends on the color gamut of.

0197For example, if the HDR picture to be encoded is represented by the BT709 color gamut defined by ITU-R Rec.709, the matrix A is given by: (Outside 45)<img id="000048" he="24" wi="128" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Matrix C is given by: (Outside 46)<img id="000049" he="25" wi="128" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0198According to a variant of this second embodiment, OETF is a cube root function and EOTF is a cubic function. Then, in step 1121c of FIG. 11c, the two intermediate components C'1 and C'2 are then coefficiented by the two color difference components C1 and C2. (Outside 47)<img id="000050" he="20" wi="128" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />Can be obtained by scaling by: (Outside 48)<img id="000051" he="33" wi="142" file="JP2018524924A_D0001.tif" img-format="tif" img-content="drawing" />

0199And EOTF becomes a cubic function, thereby F₁Equation (17) becomes a more complicated cubic equation, which can be solved analytically by the so-called Cardano method.

0200Very complex analytical solutions also exist for quartic equations (Ferrari method), but as stated by Abel-Ruffini's theory, they no longer exist for any order greater than or equal to 5.

0201The decoder DEC1 (DEC2, DEC3, respectively) is configured to decode the data encoded by the encoder ENC1 (ENC2, ENC3, respectively). The encoders ENC1 and / or ENC2 and / or ENC3 (and the decoders DEC1 and / or DEC2 and / or DEC3) can be block-based processing.

0202The encoders ENC1 and / or ENC2 and / or ENC3 (and the decoders DEC1 and / or DEC2 and / or DEC3) are not limited to a particular encoder (decoder).

0203According to one embodiment, the encoder ENC1 is configured to encode the information INF into an SEI message such as the color remapping information SEI message defined in the HEVC standard (Annex D.2.32).

0204According to one embodiment, the encoder ENC3 encodes the backlight picture Ba as an auxiliary picture or using frame packing (Appendix D.2.16) as described in the HEVC standard. Alternatively, the weighting factor and, in some cases, the shape function are configured to be encoded in an SEI message (HEVC standard, Annex D1).

0205According to one embodiment, the decoder DEC3 is obtained from an auxiliary picture or packed frame encoded in bitstream F1 as described in the HEVC standard, or from an SEI message in bitstream F1. It is configured to ... the decoded backlight picture Ba obtained from the weighting factor obtained and possibly the shape function.

0206Encoders ENC1 and / or ENC2 (and decoders DEC1 and / or DEC2) are not limited to a particular encoder, for example, lossy image / video encoders such as JPEG, JPEG2000, MPEG2, HEVC recommendation or H264 / AVC recommendation ("Advanced video coding for generic audiovisual Services", SERIES H: AUDIOVISUAL AND MULTIMEDIA SYSTEMS, Recommendation ITU-T H.264, Telecommunication Standardization Sector of ITU, February 2014)) be able to.

0207Bitstreams F1, F2, F3 can be multiplexed together to form a single bitstream.

0208In Figure 1-11d, the module is a functional unit, which may or may not relate to a distinguishable physical unit. For example, these modules or parts thereof may be grouped into unique parts or circuits, or may contribute to the functionality of the software. On the contrary, some modules can potentially consist of separate physical entities. Devices compatible with this disclosure use pure hardware, eg, ASIC or FPGA or VLSI, such as "application specific integrated circuits", "field programmable gate arrays", "ultra-large scale integration", etc., respectively. Implemented using dedicated hardware from, or from several integrated electronic components embedded in the device, or from a blend of hardware and software components.

0209FIG. 12 represents an exemplary architecture of a device 1200 that may be configured to implement the methods described in relation to FIGS. 1-11d.

0210The device 1200 is connected by the data and address bus 1201 together with the following elements: --Microprocessor 1202 (or CPU), for example DSP (ie digital signal processor), --ROM (ie, read-only memory) 1203 and --RAM (ie, random access memory) 1204 and --I / O interface 1205 for receiving data for transmission from the application, --Battery 1206 and To be equipped.

0211As an example, the battery 1206 is outside the device. In each of the above memories, the word "register" as used herein is a small area (several bits) or a very large area (eg, the entire program, or a large amount of reception or decoding. Data) can be handled. ROM1203 contains at least programs and parameters. ROM1203 can store algorithms and instructions for performing techniques according to this principle. When switched on, CPU1202 uploads the program to RAM and executes the corresponding instruction.

0212RAM1204 is executed by CPU1202 and is used to register the program uploaded after the device 1200 is switched on, to register the input data, to register the intermediate data in different states of this method, and to execute this method. Include other variables in registers.

0213The implementations described herein can be implemented, for example, in methods or processes, devices, software programs, data streams, or signals. Even if it is discussed only in the context of a single form of implementation (eg, only as a method or device), the implementation of the feature discussed is also in other forms (eg, a program). Can be implemented. The device can be implemented, for example, with the appropriate hardware, software, and firmware. The method may be implemented, for example, in devices including computers, microprocessors, integrated circuits, or programmable logic devices, which generally refer to processing devices, such as processors. Processors further include communication devices such as computers, mobile phones, portable / personal digital assistants ("PDAs"), and other devices that facilitate the communication of information between end users.

0214According to an example of a coding or encoder, an HDR or SDR picture is taken from the source. For example, the source is --Local memory (1203 or 1204), such as video memory or RAM (ie, random access memory), flash memory, ROM (ie, read-only memory), hard disk, and --A storage interface (1205), eg, an interface with a large storage, RAM, flash memory, ROM, optical disk or magnetic support, --Communication interfaces (1205), such as wired interfaces (eg, bus interfaces, wide area network interfaces, local area network interfaces), or wireless interfaces (eg, IEEE802.11 interfaces or Bluetooth® interfaces), and --With a picture capture circuit (eg, a sensor, eg, a CCD (ie, a charge-coupled device), or a CMOS (ie, a complementary metal oxide semiconductor)) Belongs to a set containing.

0215According to an example of a decoder or decoder, the decoded SDR or HDR picture is sent to the destination, specifically the destination. --Local memory (1203 or 1204), such as video memory or RAM, flash memory, hard disk, --A storage interface (1205), eg, an interface with a large storage, RAM, flash memory, ROM, optical disk or magnetic support, --Communication interface (1205), eg wired interface (eg bus interface (eg USB (ie universal serial bus)), wide area network interface, local area network interface, HDMI® (high quality multimedia interface) ) Interface), or a wireless interface (eg, IEEE802.11 interface, WiFi® or Bluetooth® interface), --With display Belongs to a set containing.

0216According to the coding or encoder example, the bitstreams F1, F2 and / or F3 are transmitted to the destination. As an example, one or both bitstreams of bitstreams F1, F2 and F3 are stored in local or remote memory, such as video memory (1204) or RAM (1204), hard disk (1203). In one variant, one or both bitstreams are sent to a storage interface (1205), eg, an interface with a mass storage, flash memory, ROM, optical disk or magnetic support, and / or a communication interface (1205). , For example, transmitted over an interface with a point-to-point link, a communication bus, a point-to-multipoint link, or a broadcast network.

0217According to the decoder or decoder example, the bitstreams F1, F2 and / or F3 are taken from the source. Illustratively, the bitstream is read from local memory, such as video memory (1204), RAM (1204), ROM (1203), flash memory (1203) or hard disk (1203). In one variant, the bit stream is received from an interface with a storage interface (1205), such as a mass storage, RAM, ROM, flash memory, optical disk or magnetic support, and / or a communication interface (1205), eg. , Point-to-point link, bus, point-to-multipoint link, or received from an interface with a broadcast network.

0218According to an example, a device 1200 configured to implement the coding method described in connection with one of FIGS. 2, 4-8d, 10a-c --Mobile devices and --Communication device and --With game devices --Tablet (or tablet computer) and --Laptop and --Still image camera and --With a video camera --Coded chip and --Still image server and --With a video server (eg broadcast server, video on demand server or web server) Belongs to a set containing.

0219According to an example, a device 1200 configured to implement the decryption method described in connection with one of FIGS. 3, 9, and 11a-d. --Mobile devices and --Communication device and --With game devices --Set-top box and --TV set and --Tablet (or tablet computer) and --Laptop and --Display and --With a decryption chip Belongs to a set containing.

0220According to one embodiment shown in FIG. 13, in the context of transmission between two remote devices A, B over the communication network NET, device A is shown in FIGS. 2, 4-8d, 10a-c. Device B comprises memory RAM and ROM and associated processor configured to implement the method for encoding the picture described in connection with one, and device B is shown in FIGS. 3, 9, 11a-d. It comprises a memory RAM and a processor associated with ROM configured to implement the method for decryption described in connection with one of the above.

0221According to one example, the network is a broadcast network adapted to broadcast a still image or video picture from device A to a decoding device including device B.

0222Implementations of the various processes and features described herein can be embodied in a variety of different devices or applications. Examples of such devices are encoders, decoders, post-processors that process the output from the decoders, preprocessors that provide inputs to the encoders, video encoders, video decoders, video codecs, web servers. , Set-top boxes, laptops, personal computers, mobile phones, PDAs, and any other device for processing pictures or videos, or other communication devices. As should be clear, the device can be mobile and even installed in a mobile vehicle.

0223The method may also be implemented by instructions executed by a processor, such instructions (and / or data values generated by one implementation) may be stored on a computer-readable storage medium. A computer-readable storage medium is embodied in one or more computer-readable media, in the form of a computer-readable program product on which computer-readable program code executable by a computer is embodied. Can be taken. The computer-readable storage medium used herein is a non-temporary storage medium that is endowed with the unique ability to store information internally and to provide the retrieval of information from it. Conceivable. The computer-readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination described above. The following are portable computer diskettes, hard disks, read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, Alternatively, any suitable combination described above provides a more specific example of a computer-readable storage medium to which this principle can be applied, but is exemplary, as will be readily appreciated by those skilled in the art. Please understand that it is just a non-exhaustive list.

0224Instructions can form a tangibly embodied application program on a processor-readable medium.

0225The instructions can be, for example, hardware, firmware, software, or a combination. Instructions can be contained, for example, in an operating system, a separate application, or a combination of the two. Thus, a processor can be characterized, for example, as both a device that is configured to perform a process and a device that contains a processor-readable medium (such as a storage device) that has instructions to perform the process. .. In addition, processor-readable media can store data values produced by an implementation in addition to or instead of instructions.

0226As will be apparent to those of skill in the art, implementations can, for example, generate various signals formatted to hold information that can be stored or transmitted. The information can include, for example, instructions to implement the method, or data generated by one of the described implementations. For example, the signal retains the rules for writing or reading the syntax of the described embodiment as data, or the actual syntax value written by the described embodiment as data. Can be formatted. Such signals can be formatted, for example, as electromagnetic waves (eg, using the radio frequency portion of the spectrum) or as baseband signals. Formatting can include, for example, encoding the data stream and modulating the carrier wave with the encoded data stream. The information held by the signal can be, for example, analog or digital information. The signal can be transmitted over a variety of different wired or wireless links, as is known. The signal can be stored on a processor readable medium.

0227Several implementations are listed. Nevertheless, it will be understood that various modifications can be made. For example, elements of different implementations may be combined, supplemented, modified, or removed to generate other implementations. In addition, one of ordinary skill in the art can use other structures and processes in place of those disclosed, and the resulting implementation will at least substantially in order to achieve at least substantially the same results as the disclosed implementation. You will understand that you will perform at least substantially the same function in the same way. Therefore, these and other implementations are contemplated by this application.