HDR cloud video processing method, system and device

By matching the graphics card format in cloud games and crawling and converting HDR video data, and adapting it to the client, the color overexposure problem of HDR video data when displayed in cloud games is solved, and the correct dynamic range of HDR video is achieved show.

CN114422829BActive Publication Date: 2024-11-22HANGZHOU WULIAN TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210113659.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-01-30
Publication Date
2024-11-22
Estimated Expiration
2042-01-30

AI Technical Summary

Technical Problem

In cloud games or cloud computers, HDR video data has a problem of overexposure when displaying the client due to mismatched formats. The existing technology has failed to effectively adapt to HDR video, resulting in incorrect dynamic range.

Method used

By determining the cache format of the graphics card on the host side, crawling and converting the video data into a format that matches the graphics card format, using an encoder to encode and send the data to the client, the client decodes and converts it back to the original format for display, and according to the metadata Set up the graphics card exchange chain to ensure the correct display of HDR dynamic range.

Benefits of technology

Avoid the color overexposure of HDR video data when displayed in SDR format, ensure that the dynamic range of HDR video is displayed correctly on the client side, and avoid the problem of color overexposure and dimness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114422829B_ABST
    Figure CN114422829B_ABST
Patent Text Reader

Abstract

The present invention discloses an HDR cloud video processing method, system and device. By adopting a first format that is the same as the cache format of a graphics card on a host side to capture video data, when HDR is displayed on the host side, the video data of the current display screen is captured in a format corresponding to HDR, thereby avoiding the phenomenon in the prior art that medium-brightness pixels in the video data in the HDR format become overexposed when displayed on the client after being captured in the SDR format due to the adoption of the SDR video format to capture HDR video. The dynamic range of the HDR video is correctly displayed on the client side without the phenomenon of color overexposure.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of cloud gaming technology, and in particular to an HDR cloud video processing method, system and device. Background Art

[0002] With the construction and development of 5G networks, cloud gaming, cloud computing and other services are developing rapidly. Cloud gaming or cloud computing is achieved by encoding the current display screen, audio data and cursor data of the host and sending them to the client through the network. The client decodes the current display screen, audio data and cursor data and displays the current display screen of the host on the client screen and plays the audio data. At the same time, the client sends the mouse displacement and other data to the host, and performs operations corresponding to the mouse displacement and other data on the host, so as to achieve the effect of controlling the host in real time through the client.

[0003] The dynamic range of HDR (High Dynamic Range) video can reach more than 1000 nits. Compared with SDR (Standard Dynamic Range) video, HDR has richer colors and stronger contrast, giving people an immersive feeling. However, in current cloud games or cloud computer services, there is no corresponding adaptation for HDR, and the format used when capturing video data is different from the cache format of the graphics card on the host side. Since the video data in HDR format is not mapped one by one to the SDR video format when capturing video data, the medium-brightness pixels in the HDR video data will become very bright when captured in SDR format, resulting in the inability to correctly display the dynamic range of the video data when the client displays the video data after HDR is turned on on the host side, resulting in overexposure of colors. Summary of the invention

[0004] The purpose of the present invention is to provide an HDR cloud video processing method, system and device, so that the dynamic range of the HDR video can be correctly displayed on the client without color overexposure.

[0005] In order to solve the above technical problems, the present invention provides an HDR cloud video processing method, comprising:

[0006] Determine that the cache format of the graphics card on the host side is the first format;

[0007] Capturing video data of a current display screen of the host terminal according to the first format;

[0008] Converting the video data in the first format into a second format acceptable to the encoder;

[0009] The encoder encodes the video data in the second format and sends it to the client, so that the client decodes the video data in the second format, converts the video data in the second format into video data in the first format, and then displays it on the client.

[0010] Preferably, before the encoder encodes the video data in the second format and sends it to the client, the method further includes:

[0011] Determine the EOTF and color space of the currently displayed picture;

[0012] Determine whether the current display image is HDR;

[0013] If it is HDR, obtain metadata of the currently displayed image;

[0014] Setting the encoder according to the EOTF and color space of the current display picture;

[0015] The method includes encoding the video data in the second format by the encoder and sending the encoded video data to the client, so that the client decodes the video data in the second format, converts the video data in the second format into the video data in the first format, and then displays the encoded video data on the client, including:

[0016] After encoding the video data in the second format through the encoder, the metadata and the encoded video data are sent to the client, so that the client decodes the video data in the second format and converts the video data in the second format into video data in the first format, and sets the exchange chain of the graphics card of the client according to the metadata to display the video data in the first format on the client.

[0017] Preferably, determining whether the currently displayed picture is HDR includes:

[0018] Determine whether the current display image is HDR according to the color space.

[0019] Preferably, judging whether the currently displayed picture is HDR according to the color space includes:

[0020] When the color space is the BT.2020 color space, it is determined that the current display image is HDR.

[0021] Preferably, the metadata includes:

[0022] The coordinate values ​​of the color space of the currently displayed picture and the maximum brightness, minimum brightness and average brightness of the currently displayed picture.

[0023] Preferably, the client sets a swap chain of a graphics card of the client according to the metadata, including:

[0024] Normalizing the coordinate values ​​of the color space of the currently displayed picture and the maximum brightness, minimum brightness and average brightness of the currently displayed picture;

[0025] Set the SetHDRMETADATA function interface according to the normalized coordinate values ​​of the color space of the current display image and the maximum brightness, minimum brightness and average brightness of the current display image;

[0026] Use the SetHDRMETADATA function interface to set the swap chain.

[0027] Preferably, after setting the encoder according to the EOTF and color space of the current display picture, the method further includes:

[0028] A QP offset parameter for luma and a QP offset parameter for chroma are set in the encoder.

[0029] Preferably, before converting the video data in the first format into a second format acceptable to the encoder, the method further includes:

[0030] If the current display screen of the host is HDR and the client does not support displaying HDR, the video data in the first format is converted into SDR video data through tone mapping.

[0031] The present invention also provides an HDR cloud video processing system, comprising:

[0032] A first format determining unit, used to determine that the cache format of the graphics card on the host side is a first format;

[0033] A video data capture unit, used for capturing the video data of the current display screen of the host terminal according to the first format;

[0034] A format conversion unit, used for converting the video data in the first format into a second format that can be received by the encoder;

[0035] The video data sending unit is used to encode the video data in the second format through the encoder and send it to the client, so that the client can decode the video data in the second format, convert the video data in the second format into video data in the first format, and then display it on the client.

[0036] The present invention also provides an HDR cloud video processing device, comprising:

[0037] Memory for storing computer programs;

[0038] A processor is used to implement the steps of the above-mentioned HDR cloud video processing method when executing the computer program.

[0039] The present invention provides an HDR cloud video processing method, system and device, which adopt a first format that is the same as the cache format of a graphics card on the host side to capture video data, so that when the host side displays in HDR, because the video data of the current display screen is captured in a format corresponding to HDR, the phenomenon in the prior art that medium-brightness pixels in the video data in the HDR format become very bright when displayed on the client after being captured in the SDR format due to the use of the SDR video format to capture HDR video is avoided, so that the dynamic range of the HDR video is correctly displayed on the client without the phenomenon of color overexposure. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the prior art and the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0041] Figure 1 A flowchart of an HDR cloud video processing method provided by the present invention;

[0042] Figure 2 A schematic diagram of the structure of an HDR cloud video processing system provided by the present invention;

[0043] Figure 3 A schematic diagram of the structure of an HDR cloud video processing device provided by the present invention. DETAILED DESCRIPTION

[0044] The core of the present invention is to provide an HDR cloud video processing method, system and device, so that the dynamic range of the HDR video can be correctly displayed on the client without color overexposure.

[0045] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] Please refer to Figure 1 , Figure 1 A flowchart of an HDR cloud video processing method provided by the present invention.

[0047] An HDR cloud video processing method, comprising:

[0048] S1: Determine that the cache format of the graphics card on the host side is the first format;

[0049] Considering that when displaying HDR content and SDR content, the cache format of the graphics card on the host side is different, and there are multiple options. For example, when displaying HDR, the cache format of the graphics card is usually DXGI_FORMAT_R16G16B16A16_FLOAT or DXGI_FORMAT_R10G10B10A10_UNORM format, and when displaying SDR on the host side, the cache format of the graphics card is DXGI_FORMAT_R10G10B10A10_UNORM or DXGI_FORMAT_R8G8B8A8_UNORM. Therefore, in order to adapt to the HDR display so that when HDR is turned on on the host side, the client can correctly display the dynamic range of HDR, in this embodiment, it is necessary to first determine that the cache format of the graphics card on the host side is the first format.

[0050] The first format here may be, but is not limited to, DXGI_FORMAT_R16G16B16A16_FLOAT / DXGI_FORMAT_R10G10B10A10_UNORM / DXGI_FORMAT_R8G8B8A8_UNORM, and this application does not make any special limitation on this.

[0051] S2: Capture video data of the current display screen of the host terminal according to the first format;

[0052] When the current display screen of the host side is HDR, assuming that the graphics card cache format is a 16-bit integer, the range of the video data is 0-65535, and if the video data in 16-bit format is captured in 10-bit format, that is, in the SDR video format HDR video, the range of the captured video data is only 0-1024. Since the original video data, that is, the video data in the range of 0-65535, is not mapped one by one to the range of 0-1024 when capturing the video data of the current display screen of the host side, the value greater than 1024 will be truncated to 1024 for storage, causing the pixels with medium brightness in HDR to become very bright, resulting in overexposure. In order to solve the above problem, in this embodiment, the video data of the current display screen of the host side is captured in the first format that is the same as the cache format of the graphics card of the host side, that is, the same first format as the current display screen of the host side, so as to avoid the overexposure of the display when the client displays HDR due to the difference between the cache format of the graphics card of the host side and the format of capturing the current display screen when the host side displays HDR.

[0053] S3: converting the video data in the first format into a second format that can be received by the encoder;

[0054] Considering that the format adopted by the encoder during encoding and decoding is different from the format used by the host and client to display video data. Therefore, in order for the encoder to be able to normally encode and decode the video data of the current display screen, it is necessary to convert the video data in the first format into a second format that the encoder can receive. For example, RGB related formats are usually used in display systems, while Y'Cb'Cr' related formats are usually used in encoders. When the first format is DXGI_FORMAT_R16G16B16A16_FLOAT, the second format is P016 or YUV420P10LE, when the first format is DXGI_FORMAT_R10G10B10A10_UNORM, the second format is YUV420P10LE, and when the first format is DXGI_FORMAT_R8G8B8A8_UNORM, the second format is YUV420P.

[0055] It should also be noted that in order to convert the video data in the first format into the video data in the second format, the video data needs to be downsampled, which is not particularly limited in the present application.

[0056] S4: The video data in the second format is encoded by an encoder and sent to the client, so that the client decodes the video data in the second format, converts the video data in the second format into video data in the first format, and then displays the video data on the client.

[0057] In order to display the video data in the second format in the encoder on the client, after receiving the encoded video data in the second format sent by the host, the client first decodes the video data in the second format, and then converts the video data in the second format into video data in the first format and displays it on the client screen. For example, FFmpeg conversion based on the CPU or CUDA (Compute Unified Device Architecture) conversion based on the GPU (Graphics Processing Unit) can be used.

[0058] In addition, the encoder here can be but is not limited to a HEVC (High Efficiency Video Coding) encoder and an AVC (Advanced Video Coding). Users can choose according to the current display screen and actual needs of the host side, and this application does not make any special limitations here.

[0059] It should also be noted that when restoring the video data in the second format to the video data in the first format, it is also necessary to upsample the video data through filtering operations to avoid color loss. This application does not specifically limit the specific method used for upsampling.

[0060] To summarize, in this embodiment, the first format that is the same as the cache format of the graphics card on the host side is adopted to capture video data, so that when the host side displays in HDR, since the video data of the currently displayed picture is captured in a format corresponding to HDR, the phenomenon in the prior art that medium-brightness pixels in the video data in the HDR format become very bright when displayed on the client after being captured in the SDR format due to the adoption of the SDR video format to capture the HDR video is avoided, so that the dynamic range of the HDR video is correctly displayed on the client without the phenomenon of overexposure of colors.

[0061] Based on the above embodiments

[0062] As a preferred embodiment, before encoding the video data in the second format by the encoder and sending it to the client, the method further includes:

[0063] Determine the EOTF (Electrical-Optical Transfer Function) and color space of the current display image;

[0064] Determine whether the current display image is HDR;

[0065] If it is HDR, get the metadata of the currently displayed image;

[0066] Set the encoder according to the EOTF and color space of the current display;

[0067] The video data in the second format is encoded by an encoder and sent to a client, so that the client decodes the video data in the second format, converts the video data in the second format into video data in the first format, and then displays the video data on the client, including:

[0068] After encoding the video data in the second format through the encoder, the metadata and the encoded video data are sent to the client, so that the client decodes the video data in the second format and converts the video data in the second format into video data in the first format, and displays the video data in the first format on the client after setting the exchange chain of the graphics card of the client according to the metadata.

[0069] Considering that the color space and EOTF of HDR and SDR are different, HDR usually uses BT.2020 color space and PQ (Perceptual Quantizer) conversion curve, while SDR usually uses BT.709 color space and gamma 2.2 curve. If the host is currently displaying HDR images, and the BT.709 color space and gamma 2.2 curve are used to set the encoding parameters of the encoder, the colors will be dim when the HDR images are displayed on the client. In order to solve the above problem, in this embodiment, the EOTF and color space of the current display screen of the host side will be determined first. For example, the color space of the current display screen of the host side can be queried through the DXGIOutput6::GetDesc interface. If it is DXGI_COLOR_SPACE_RGB_FULL_G2084_NONE_P2020, it means that the current display screen of the host side is the BT.2020 color space. If it is DXGI_COLOR_SPACE_RGB_FULL_G22_NONE_P709, it means that the current display screen of the host side is the BT.709 color space. Set the encoding parameters of the encoder according to the EOTF and color space of the current display screen on the host side. For example, set the encoder profile to main10 profile or main profile according to the color space and conversion curve of the current display screen on the host side, and set the syntax elements such as video_signal_type_present_flag, colour_description_present_flag, colour_primaries, transfer_characteristics, matrix_coeffs, etc. in the SPS (Sequence parameter set). Since HDR video has the unique attribute of metadata, while SDR does not, the metadata contains the coordinate values ​​and brightness of the color space of the HDR video. Therefore, when the current display screen on the host side is HDR, the swap chain of the client graphics card needs to be set according to the HDR metadata, otherwise the color will be dim. In this embodiment, it is determined whether the current display screen on the host side is HDR. When the current display screen on the host side is HDR, the metadata of the current display screen on the host side is obtained and sent to the client. Before the client displays the video data in the first format, the exchange chain of the client's graphics card is set according to the metadata, so that the dynamic range of HDR is correctly displayed and the phenomenon of dim colors is avoided.

[0070] As a preferred embodiment, determining whether the current display image is HDR includes:

[0071] Determine whether the current display image is HDR based on the color space.

[0072] Considering that when displaying HDR and SDR images on the host side, the color space of the graphics card on the host side is different, for example, HDR is usually BT.2020 color space, and SDR is usually BT.709 color space. Therefore, in this embodiment, it is determined whether the current display image on the host side is HDR based on the color space, which realizes the detection of HDR and the implementation method is simple.

[0073] As a preferred embodiment, judging whether the current display image is HDR according to the color space includes:

[0074] When the color space is the BT.2020 color space, it is determined that the current display image is HDR.

[0075] Since HDR is usually a BT.2020 color space in a cloud computer or cloud gaming system, in this embodiment, when the color space is the BT.2020 color space, the current display image is determined to be HDR.

[0076] As a preferred embodiment, the metadata includes:

[0077] The coordinate values ​​of the color space of the currently displayed image and the maximum brightness, minimum brightness and average brightness of the currently displayed image.

[0078] In this embodiment, the metadata includes the coordinate values ​​of the color space of the current display image, such as the coordinate value of the red primary color, the coordinate value of the blue primary color, the coordinate value of the green primary color, and the coordinate value of the white point. The metadata also includes the maximum brightness, the minimum brightness, and the average brightness of the current display image.

[0079] It should also be noted that, since the coordinate value of the color space of the current display screen is a fixed value, it can be obtained only once, or it can be obtained periodically or in real time. The maximum brightness, minimum brightness and average brightness of the current display screen will change in real time. Considering that the human eye cannot perceive the change of brightness within a certain range, the maximum brightness, minimum brightness and average brightness of the current display screen can be obtained in real time or periodically, and this application does not make any special restrictions here.

[0080] As a preferred embodiment, the client sets the swap chain of the graphics card of the client according to the metadata, including:

[0081] Normalize the coordinate values ​​of the color space of the current display image and the maximum brightness, minimum brightness and average brightness of the current display image;

[0082] Set the SetHDRMETADATA function interface according to the normalized coordinate values ​​of the color space of the current display image and the maximum brightness, minimum brightness and average brightness of the current display image;

[0083] Use the SetHDRMETADATA function interface to set the swap chain.

[0084] In this embodiment, when setting the swap chain of the graphics card, the coordinate values ​​of the color space of the current display screen and the maximum brightness, minimum brightness and average brightness of the current display screen are first normalized, and then the relevant setting items in the SetHDRMETADATA function interface are set according to the normalized coordinate values ​​of the color space of the current display screen and the maximum brightness, minimum brightness and average brightness of the current display screen, such as DXGI_HDR_METADATA in the SetHDRMETADATA function interface. Then, the SetHDRMETADATA function interface is used to set the swap chain of the graphics card of the client, so that the dynamic range of the HDR video in the first format can be correctly displayed on the client.

[0085] As a preferred embodiment, after setting the encoder according to the EOTF and color space of the current display picture, the method further includes:

[0086] The QP (Quantizer Parameter) offset parameter of the luminance and the QP offset parameter of the chrominance are set in the encoder.

[0087] Taking into account that chroma leakage may occur during downsampling, in order to compensate for the loss caused by chroma leakage, in this embodiment, the QP bias parameter of brightness and the QP bias parameter of chroma are set in the encoder, which further ensures that the dynamic range of HDR can be correctly displayed on the client.

[0088] As a preferred embodiment, before converting the video data in the first format into the second format acceptable to the encoder, the method further includes:

[0089] If the current display screen of the host is HDR and the client does not support displaying HDR, the video data in the first format is converted to SDR video data through tone mapping.

[0090] Considering that if the host supports HDR, but the client only supports SDR, that is, the client cannot convert the video data in the second format back to the video data in the first format, so that the client cannot display the received video data normally. In order to solve the above problem, in this embodiment, when the current display screen of the host is HDR and the client does not support the display of HDR, before converting the video data in the first format to the second format, the video data in the first format, that is, the HDR video data, will be first converted to the video data in the SDR format through tone mapping, so that the client can display the screen of the host after receiving the video data. After the HDR video data is converted to the video data in the SDR format through tone mapping, the encoding parameter settings of the encoder and the related settings when the client is displayed on the screen are set according to the relevant parameters of SDR, and this application does not make any special restrictions on this.

[0091] It should also be noted that when multiple clients are connected to a host and the current display screen of the host is HDR, if some of the multiple clients support HDR, for the clients that support HDR, the host converts the video data in the first format into video data in the second format and sends it to the clients that support HDR, so that the clients that support HDR convert the video in the second format into HDR video data in the first format and display HDR; for the clients that do not support HDR, the video data in the first format is converted into SDR video data through tone mapping and sent to the clients that do not support HDR, so that the clients that do not support HDR display SDR video data.

[0092] Please refer to Figure 2 , Figure 2 A schematic diagram of the structure of an HDR cloud video processing system provided by the present invention.

[0093] The present invention also provides an HDR cloud video processing system, comprising:

[0094] A first format determining unit 21, used to determine that the cache format of the graphics card of the host end is a first format;

[0095] The video data capture unit 22 is used to capture the video data of the current display screen of the host terminal according to the first format;

[0096] A format conversion unit 23, configured to convert the video data in a first format into a second format that can be received by the encoder;

[0097] The video data sending unit 24 is used to encode the video data in the second format through an encoder and send it to the client, so that the client decodes the video data in the second format, converts the video data in the second format into video data in the first format and displays it on the client.

[0098] Please refer to Figure 3 , Figure 3 A schematic diagram of the structure of an HDR cloud video processing device provided by the present invention.

[0099] The present invention also provides an HDR cloud video processing device, comprising:

[0100] A memory 31, used for storing computer programs;

[0101] The processor 32 is used to implement the steps of the above-mentioned HDR cloud video processing method when executing the computer program.

[0102] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0103] It should also be noted that, in this specification, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0104] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A HDR cloud video processing method, characterized in that: include: Determine that the cache format of the graphics card on the host side is the first format; Capturing video data of a current display screen of the host terminal according to the first format; Convert the video data in the first format into a second format that can be received by the encoder; wherein the correspondence between the video data in the first format and the second format includes at least one of a first format of DXGI_FORMAT_R16G16B16A16_FLOAT corresponding to a second format of P016 or YUV420P10LE, a first format of DXGI_FORMAT_R10G10B10A10_UNORM corresponding to a second format of YUV420P10LE, and a first format of DXGI_FORMAT_R8G8B8A8_UNORM corresponding to a second format of YUV420P; Determine the EOTF and color space of the currently displayed picture; Determine whether the current display image is HDR; If it is HDR, obtain metadata of the currently displayed image; Setting the encoder according to the EOTF and color space of the currently displayed picture; If the current display screen of the host end is HDR and the client does not support displaying HDR, the video data in the first format is converted into SDR video data through tone mapping; After encoding the video data in the second format through the encoder, the metadata and the encoded video data are sent to the client, so that the client decodes the video data in the second format and converts the video data in the second format into video data in the first format, and after setting the exchange chain of the graphics card of the client according to the metadata, the converted data is up-sampled through a filtering operation to display the video data in the first format on the client.

2. The HDR cloud video processing method according to claim 1, characterized in that: Determining whether the current display image is HDR includes: Determine whether the current display image is HDR according to the color space.

3. The HDR cloud video processing method according to claim 2, characterized in that: Determining whether the current display image is HDR according to the color space includes: When the color space is the BT.2020 color space, it is determined that the current display image is HDR.

4. The HDR cloud video processing method according to claim 1, characterized in that: The metadata includes: The coordinate values ​​of the color space of the currently displayed picture and the maximum brightness, minimum brightness and average brightness of the currently displayed picture.

5. The HDR cloud video processing method according to claim 4, characterized in that: The client sets a swap chain of a graphics card of the client according to the metadata, including: Normalizing the coordinate values ​​of the color space of the currently displayed picture and the maximum brightness, minimum brightness and average brightness of the currently displayed picture; Set the SetHDRMETADATA function interface according to the normalized coordinate values ​​of the color space of the current display image and the maximum brightness, minimum brightness and average brightness of the current display image; Use the SetHDRMETADATA function interface to set the swap chain.

6. The HDR cloud video processing method according to claim 1, characterized in that: After the encoder is set according to the EOTF and the color space of the current display picture, the method further includes: A QP offset parameter for luma and a QP offset parameter for chroma are set in the encoder.

7. An HDR cloud video processing system, characterized in that: include: A first format determining unit, used to determine that the cache format of the graphics card on the host side is a first format; A video data capture unit, used for capturing the video data of the current display screen of the host terminal according to the first format; a format conversion unit, configured to convert the video data in the first format into a second format that can be received by the encoder; wherein the correspondence between the video data in the first format and the second format includes at least one of a first format of DXGI_FORMAT_R16G16B16A16_FLOAT corresponding to a second format of P016 or YUV420P10LE, a first format of DXGI_FORMAT_R10G10B10A10_UNORM corresponding to a second format of YUV420P10LE, and a first format of DXGI_FORMAT_R8G8B8A8_UNORM corresponding to a second format of YUV420P; A video data sending unit is used to determine the EOTF and color space of the current display picture; determine whether the current display picture is HDR; if it is HDR, obtain the metadata of the current display picture; set the encoder according to the EOTF and color space of the current display picture; if the current display picture of the host end is HDR and the client does not support displaying HDR, convert the video data in the first format into SDR video data through tone mapping; after encoding the video data in the second format through the encoder, send the metadata and the encoded video data to the client, so that the client decodes the video data in the second format and converts the video data in the second format into video data in the first format, and after setting the exchange chain of the graphics card of the client according to the metadata, upsample the converted data through filtering operation to display the video data in the first format on the client.

8. An HDR cloud video processing device, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the HDR cloud video processing method according to any one of claims 1 to 6 when executing the computer program.