Method and device for determining video transcoding scheme and video transcoding method and device

Abstract

The present disclosure relates to a method and device for determining a video transcoding scheme, and a video transcoding method and device. The method for determining a video transcoding scheme includes: obtaining at least one video and at least one candidate video transcoding scheme; transcoding each video in the at least one video based on the at least one candidate video transcoding scheme, and determining a video quality evaluation index set and a bit rate set for each video; based on the video quality evaluation index set, bit rate set, and target bit rate of each video, determining the video transcoding scheme corresponding to each video from at least one candidate video transcoding scheme, wherein the target bit rate is the expected average bit rate after transcoding the at least one video. According to the method and device for determining a video transcoding scheme, and the method and device for video transcoding of the present disclosure, a more suitable video transcoding scheme can be selected according to different video contents, thereby improving the average video quality of the transcoded video.

Description

Technical Field

[0001] The present disclosure relates to the field of video technology, and more specifically, to a method and device for determining a video transcoding scheme, a method and device for training a video transcoding decision model, and a video transcoding method and device. Background Art

[0002] In recent years, with the development of communications and internet technologies, the amount of video content on the internet has continued to grow. To reduce bandwidth costs, videos delivered to users are often transcoded and compressed. To improve transcoded video quality and compression efficiency, video pre-processing, such as denoising, sharpening, and blurring, is required. The selection of transcoding parameters also plays a significant role in compression efficiency.

[0003] At present, fixed pre-processing parameters and fixed CRF (Constant Rate Factor) values ​​are usually used to pre-process and transcode a batch of videos. For example, when performing video pre-processing, a representative video sequence is usually selected as a test sequence, and the test sequence is transcoded using different pre-processing algorithms and pre-processing intensities. Through various subjective and objective evaluation methods, the pre-processing algorithm and intensity combination with the best average performance on this test sequence is selected to process a batch of videos. However, the most suitable processing solutions for videos with different contents are also different. If fixed pre-processing parameters and fixed CRF values ​​are used for transcoding, the average quality of the transcoded videos may not be high. Summary of the Invention

[0004] The present disclosure provides a method and device for determining a video transcoding scheme, a method and device for training a video transcoding decision model, and a video transcoding method and device, to at least solve the problems in the above-mentioned related technologies.

[0005] According to a first aspect of an embodiment of the present disclosure, a method for determining a video transcoding scheme is provided, comprising: obtaining at least one video and at least one candidate video transcoding scheme; transcoding each video in the at least one video based on the at least one candidate video transcoding scheme, and determining a video quality evaluation indicator set and a bit rate set for each video in the at least one video; based on the video quality evaluation indicator set, the bit rate set and the target bit rate of each video in the at least one video, determining a video transcoding scheme corresponding to each video in the at least one video from the at least one candidate video transcoding scheme, wherein the target bit rate is an expected average bit rate after transcoding the at least one video.

[0006] Optionally, the at least one candidate video transcoding scheme is obtained by the following operations: obtaining a first processing scheme, the first processing scheme including at least one transcoding parameter setting strategy for transcoding the at least one video; and determining the at least one transcoding parameter setting strategy as the at least one candidate video transcoding scheme.

[0007] Optionally, the at least one candidate video transcoding scheme is also obtained by the following operations: obtaining at least one type of second processing scheme, each type of second processing scheme in the at least one type of second processing scheme includes at least one video pre-processing strategy for performing video pre-processing on the at least one video; selecting a video pre-processing strategy from the at least one video pre-processing strategy of each type of second processing scheme; selecting a transcoding parameter setting strategy from the at least one transcoding parameter setting strategy; combining the video pre-processing strategy selected from each type of second processing scheme with the transcoding parameter setting strategy selected from the at least one transcoding parameter setting strategy in a preset order to obtain the at least one candidate video transcoding scheme.

[0008] Optionally, the preset order is the execution order of each type of second processing scheme in the at least one type of second processing scheme when being executed.

[0009] Optionally, the method of determining a video processing scheme corresponding to each video in the at least one video from the at least one candidate video transcoding scheme based on the video quality evaluation indicator set, bit rate set and target bit rate of each video in the at least one video includes: based on the video quality evaluation indicator set and bit rate set of each video in the at least one video, under the condition that the average bit rate of the at least one video does not exceed the target bit rate, determining a video processing scheme corresponding to each video in the at least one video from the at least one candidate video transcoding scheme so that the total video quality evaluation indicator of the at least one video reaches a preset standard.

[0010] Optionally, determining from the at least one candidate video transcoding scheme a video processing scheme corresponding to each video in the at least one video when the total video quality evaluation index of the at least one video reaches a preset standard includes: using a dynamic programming method to determine from the at least one candidate video transcoding scheme a video processing scheme corresponding to each video in the at least one video when the total video quality evaluation index of the at least one video reaches a preset standard.

[0011] Optionally, the video quality evaluation index set is a set of video quality evaluation indicators after each video in the at least one video is transcoded according to each candidate video transcoding scheme in the at least one candidate video transcoding scheme, and the video quality evaluation indicators include objective quality evaluation indicators and / or subjective quality evaluation indicators.

[0012] According to a second aspect of an embodiment of the present disclosure, a method for training a video transcoding decision model is provided, comprising: predicting a video transcoding strategy of a training video through a video transcoding decision model to obtain a predicted video transcoding strategy of the training video; calculating a predicted loss using a loss function based on the predicted video transcoding strategy and a real video transcoding strategy, wherein the real video transcoding strategy is obtained through a target video transcoding scheme, and the target video transcoding scheme is determined in advance through the method for determining a video transcoding scheme described in any one of the first aspects; and training the video transcoding decision model by adjusting model parameters of the video transcoding decision model according to the predicted loss.

[0013] According to a third aspect of an embodiment of the present disclosure, a video transcoding method is provided, comprising: obtaining at least one video feature of a current video; inputting the at least one video feature into a corresponding video transcoding decision model to obtain a video transcoding strategy for the current video, wherein the video transcoding decision model is trained using the training method for the video transcoding decision model described in the second aspect; and transcoding the current video using the video processing strategy.

[0014] Optionally, the video transcoding decision model includes a first model, which is used to obtain a transcoding parameter setting strategy. The step of inputting the at least one video feature into the corresponding video transcoding decision model to obtain a video transcoding strategy for the current video includes: inputting a video feature corresponding to the first model in the at least one video feature into the first model to obtain a transcoding parameter setting strategy; and using the obtained transcoding parameter setting strategy as the video transcoding strategy.

[0015] Optionally, the video transcoding decision model also includes at least one second model, which is used to obtain a video pre-processing strategy. The inputting of the at least one video feature into the corresponding video transcoding decision model to obtain a video transcoding strategy for the current video includes: inputting the video feature corresponding to the at least one second model in the at least one video feature into the at least one second model to obtain at least one video pre-processing strategy; and combining the at least one video pre-processing strategy with the transcoding parameter setting strategy in a preset order as the video transcoding strategy.

[0016] Optionally, the preset order is the execution order of the at least one video pre-processing strategy when being executed.

[0017] According to a fourth aspect of an embodiment of the present disclosure, a device for determining a video transcoding scheme is provided, comprising: a data acquisition unit, configured to: acquire at least one video and at least one candidate video transcoding scheme; a parameter determination unit, configured to: transcode each video in the at least one video respectively based on the at least one candidate video transcoding scheme, and determine a video quality evaluation indicator set and a bit rate set for each video in the at least one video; a scheme determination unit, configured to: determine, from the at least one candidate video transcoding scheme, a video transcoding scheme corresponding to each video in the at least one video based on the video quality evaluation indicator set, the bit rate set and the target bit rate of each video in the at least one video, wherein the target bit rate is an expected average bit rate after transcoding the at least one video.

[0018] Optionally, the data acquisition unit is configured to: acquire a first processing scheme, the first processing scheme including at least one transcoding parameter setting strategy for transcoding the at least one video; and determine the at least one transcoding parameter setting strategy as the at least one candidate video transcoding scheme.

[0019] Optionally, the data acquisition unit is further configured to: obtain at least one type of second processing scheme, each type of second processing scheme in the at least one type of second processing scheme includes at least one video pre-processing strategy for performing video pre-processing on the at least one video; select a video pre-processing strategy from the at least one video pre-processing strategy of each type of second processing scheme; select a transcoding parameter setting strategy from the at least one transcoding parameter setting strategy; combine the video pre-processing strategy selected from each type of second processing scheme with the transcoding parameter setting strategy selected from the at least one transcoding parameter setting strategy in a preset order to obtain the at least one candidate video transcoding scheme.

[0020] Optionally, the preset order is the execution order of each type of second processing scheme in the at least one type of second processing scheme when being executed.

[0021] Optionally, the scheme determination unit is configured to: based on a set of video quality evaluation indicators and a bit rate set for each video in the at least one video, and under the condition that the average bit rate of the at least one video does not exceed the target bit rate, determine from the at least one candidate video transcoding scheme a video processing scheme corresponding to each video in the at least one video when the total video quality evaluation indicator of the at least one video reaches a preset standard.

[0022] Optionally, the scheme determination unit is configured to: use dynamic programming to determine, from the at least one candidate video transcoding scheme, a video processing scheme corresponding to each video in the at least one video when the total video quality evaluation index of the at least one video reaches a preset standard.

[0023] Optionally, the video quality evaluation index set is a set of video quality evaluation indicators after each video in the at least one video is transcoded according to each candidate video transcoding scheme in the at least one candidate video transcoding scheme, and the video quality evaluation indicators include objective quality evaluation indicators and / or subjective quality evaluation indicators.

[0024] According to a fifth aspect of an embodiment of the present disclosure, a training device for a video transcoding decision model is provided, comprising: a transcoding strategy prediction unit, configured to: predict the video transcoding strategy of a training video through a video transcoding decision model to obtain the predicted video transcoding strategy of the training video; a prediction loss calculation unit, configured to: calculate the prediction loss using a loss function based on the predicted video transcoding strategy and the real video transcoding strategy, wherein the real video transcoding strategy is obtained through a target video transcoding scheme, and the target video transcoding scheme is pre-determined by the method for determining the video transcoding scheme described in any one of the first aspects; a model parameter adjustment unit, configured to: train the video transcoding decision model by adjusting the model parameters of the video transcoding decision model according to the predicted loss.

[0025] According to a sixth aspect of an embodiment of the present disclosure, a video transcoding device is provided, comprising: a feature acquisition unit, configured to acquire at least one video feature of a current video; a strategy acquisition unit, configured to input the at least one video feature into a corresponding video transcoding decision model, respectively, to obtain a video transcoding strategy for the current video, wherein the video transcoding decision model is trained using the training method of the video transcoding decision model described in the second aspect; and a video transcoding unit, configured to transcode the current video using the video processing strategy.

[0026] Optionally, the video processing decision model includes a first model, which is used to obtain a transcoding parameter setting strategy. The strategy acquisition unit is configured to: input a video feature corresponding to the first model in the at least one video feature into the first model to obtain a transcoding parameter setting strategy; and use the obtained transcoding parameter setting strategy as the video transcoding strategy.

[0027] Optionally, the video transcoding decision model also includes at least one second model, which is used to obtain a video pre-processing strategy. The strategy acquisition unit is further configured to: input the video feature corresponding to the at least one second model in the at least one video feature into the at least one second model to obtain at least one video pre-processing strategy; combine the at least one video pre-processing strategy with the transcoding parameter setting strategy in a preset order as the video transcoding strategy.

[0028] Optionally, the video transcoding unit is configured such that: the preset order is the execution order of the at least one video pre-processing strategy when being executed.

[0029] According to a seventh aspect of an embodiment of the present disclosure, an electronic device is provided, comprising: at least one processor; and at least one memory storing computer-executable instructions, wherein the computer-executable instructions, when executed by the at least one processor, cause the at least one processor to execute the method for determining a video transcoding scheme according to the first aspect of the present disclosure, the method for training a video transcoding decision model according to the second aspect, or the video transcoding method according to the third aspect.

[0030] According to an eighth aspect of an embodiment of the present disclosure, a computer-readable storage medium storing instructions is provided. When the instructions are executed by at least one processor, the at least one processor is prompted to execute the method for determining the video transcoding scheme according to the first aspect of the present disclosure, the method for training the video transcoding decision model according to the second aspect, or the video transcoding method according to the third aspect.

[0031] According to a ninth aspect of an embodiment of the present disclosure, a computer program product is provided, in which instructions can be executed by a processor of a computer device to complete the method for determining a video transcoding scheme according to the first aspect of the present disclosure, the method for training a video transcoding decision model according to the second aspect, or the video transcoding method according to the third aspect.

[0032] The technical solutions provided by the embodiments of the present disclosure bring at least the following beneficial effects:

[0033] According to the method and device for determining a video transcoding scheme, the method and device for training a video transcoding decision model, and the video transcoding method and device disclosed in the present invention, a more suitable video transcoding scheme is provided for each video sequence based on the characteristics of the video content, which can improve the average quality of the transcoded video, and thus significantly reduce the average bit rate of the video under the same average video quality conditions. In addition, based on the dynamic programming method, a video transcoding scheme with the highest average objective quality under the condition of a given average bit rate can be obtained.

[0034] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute an improper limitation of the present disclosure.

[0036] Figure 1 is an exemplary system architecture illustrating an exemplary system in which exemplary embodiments of the present disclosure may be applied.

[0037] Figure 2 is a flowchart illustrating a method for determining a video transcoding scheme according to an exemplary embodiment of the present disclosure.

[0038] Figure 3 Detailed description is a flowchart illustrating a method for training a video transcoding decision model according to an exemplary embodiment of the present disclosure.

[0039] Figure 4 is a flowchart illustrating a video transcoding method according to an exemplary embodiment of the present disclosure.

[0040] Figure 5 is a block diagram illustrating an apparatus for determining a video transcoding scheme according to an exemplary embodiment of the present disclosure.

[0041] Figure 6 is a block diagram illustrating a training apparatus for a video transcoding decision model according to an exemplary embodiment of the present disclosure.

[0042] Figure 7 is a block diagram illustrating a video transcoding apparatus according to an exemplary embodiment of the present disclosure.

[0043] Figure 8 is a block diagram illustrating an electronic device 800 according to an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0044] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0045] It should be noted that the terms "first," "second," and the like in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the numbers used in this manner are interchangeable where appropriate so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The implementation methods described in the following examples do not represent all implementation methods consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with certain aspects of the present disclosure as detailed in the appended claims.

[0046] It should be noted that the phrase "at least one of the items" in this disclosure includes three types of parallel situations: "any one of the items", "a combination of any multiple items of the items", and "all of the items". For example, "including at least one of A and B" includes the following three parallel situations: (1) including A; (2) including B; (3) including A and B. For another example, "performing at least one of step 1 and step 2" includes the following three parallel situations: (1) performing step 1; (2) performing step 2; and (3) performing steps 1 and 2.

[0047] Before transcoding, the video source needs to be filtered (i.e., video pre-processing, for example, denoise filtering can be performed on the video source to suppress noise in the video source; and unsharp filtering can enhance the texture of the video source, so that the texture detail quality of the transcoded video is better). Afterwards, the video source is transcoded using a target bitrate control mode (for example, CQP (Constant Quantization Parameter), ABR (Average Bitrate), CRF (Constant Rate Factor), etc.). To ensure similar quality after transcoding different videos, the CRF parameters can be used to control the quality and bitrate of the transcoded video. Specifically, the CRF transcoding mode takes into account the subjective characteristics of the video: in areas with rich textures and intense motion, the human eye is not sensitive and distortion is not easy to detect, so quantization distortion is increased to save bitrate; in areas with simple textures and gentle motion, distortion is easily detected, so quantization distortion is reduced to improve quality. In actual transcoding, the CRF transcoding mode is often used, that is, a CRF value is given and each video sequence is encoded according to this CRF value.

[0048] In related technologies, before video transcoding, a representative video sequence is first selected as a test sequence, and the test sequence is transcoded with different pre-processing algorithms and intensities. The pre-processing algorithm and intensity combination with the best average performance on the test sequence is selected through various subjective and objective evaluation methods, and the pre-processing algorithm and intensity combination are used as fixed pre-processing parameters to pre-process different video sequences; alternatively, the video characteristics can be analyzed, and different pre-processing schemes can be formulated for videos with different content characteristics (for example, the noise level of the video source is analyzed, and the video source judged to be a high-noise video is denoise filtered). After that, the CRF transcoding mode is used for video transcoding. However, using fixed pre-processing parameters and fixed CRF values ​​cannot process different video sequences according to content characteristics. For individual sequences whose content characteristics deviate far from the average characteristics of the test sequence, the video quality may be significantly reduced. In addition, the optimal CRF values ​​for videos with different content characteristics may also be different, which leads to low average video quality after transcoding. Manual feature classification and pre-processing scheme formulation require complex feature analysis, which is very computationally intensive and makes it difficult to provide more appropriate features for decision-making for various pre-processing schemes.

[0049] In order to improve the average video quality of transcoded videos, the present disclosure proposes a method and device for determining a video transcoding scheme and a video transcoding method and device, which provide a more suitable video transcoding scheme for each video sequence according to the characteristics of the video content, thereby improving the average quality of the transcoded video, and further significantly reducing the average bit rate of the video under the same average video quality conditions. In addition, based on the dynamic programming method, the video transcoding scheme with the highest average objective quality under the condition of a given average bit rate can be obtained. Figures 1 to 8 The present invention specifically describes a method and apparatus for determining a video transcoding scheme, a method and apparatus for training a video transcoding decision model, and a method and apparatus for video transcoding according to exemplary embodiments of the present invention.

[0050] Figure 1 An exemplary system architecture 100 is shown in which exemplary embodiments of the present disclosure may be applied.

[0051] like Figure 1As shown, system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. Network 104 is a medium used to provide a communication link between terminal devices 101, 102, 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables. Users can use terminal devices 101, 102, 103 to interact with server 105 via network 104 to receive or send messages (e.g., requests to determine a video transcoding solution, video transcoding requests, requests to train a video transcoding decision model), etc. Terminal devices 101, 102, 103 may be installed with various video applications. Terminal devices 101, 102, 103 may be hardware or software. When terminal devices 101, 102, 103 are hardware, they may be various electronic devices with display screens and capable of audio and video playback, recording, editing, etc., including but not limited to smartphones, tablet computers, laptop computers, and desktop computers. When the terminal devices 101, 102, and 103 are software, they can be installed in the electronic devices listed above, and can be implemented as multiple software or software modules (for example, to provide distributed services) or as a single software or software module. No specific limitation is given here.

[0052] The terminal devices 101, 102, and 103 may be equipped with an image acquisition device (e.g., a camera) to collect video data. In practice, the smallest visual unit that constitutes a video is a frame. Each frame is a static image. Combining a temporally continuous sequence of frames together forms a dynamic video. In addition, the terminal devices 101, 102, and 103 may also be equipped with a component for converting electrical signals into sound (e.g., a speaker) to play sound, and may also be equipped with a device for converting analog audio signals into digital audio signals (e.g., a microphone) to collect sound.

[0053] The server 105 may be a server that provides various services, such as a backend server that supports multimedia applications installed on the terminal devices 101, 102, and 103. The backend server may parse, store, and process received data such as requests for determining a video transcoding solution, video transcoding requests, and requests for training a video transcoding decision model.

[0054] It should be noted that the server can be either hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When the server is software, it can be implemented as multiple software programs or software modules (for example, to provide distributed services), or as a single software program or software module. No specific limitations are given here.

[0055] It should be noted that the methods for determining a video transcoding scheme, the video transcoding methods, and the methods for training a video transcoding decision model provided in the embodiments of the present disclosure are typically executed by a terminal device, but may also be executed by a server, or may be executed collaboratively by a terminal device and a server. Accordingly, the apparatus for determining a video transcoding scheme, the apparatus for transcoding, and the apparatus for training a video transcoding decision model may be provided in a terminal device, a server, or both.

[0056] It should be understood that Figure 1 The number of terminal devices, networks, and servers in the embodiment is merely illustrative. Any number of terminal devices, networks, and servers may be provided as required, and the present disclosure does not limit this.

[0057] Figure 2 is a flowchart illustrating a method for determining a video transcoding scheme according to an exemplary embodiment of the present disclosure.

[0058] Reference Figure 2 In step 201, at least one video and at least one candidate video transcoding scheme may be obtained. Here, at least one video to be transcoded may be obtained, and the at least one video to be transcoded may constitute a video set. In some embodiments, multiple popular videos may be selected, for example, videos with a high number of views within a current period of time, etc., without limitation.

[0059] A candidate video transcoding scheme is a video transcoding scheme that can be adopted for each video in a video set when transcoding. According to an exemplary embodiment of the present disclosure, at least one candidate video transcoding scheme can be obtained by the following operations: First, a first processing scheme is obtained. Here, the first processing scheme includes at least one transcoding parameter setting strategy for transcoding each video in the video set. The at least one transcoding parameter setting strategy can be determined as at least one candidate video transcoding scheme. Specifically, the first processing scheme can be a video transcoding scheme based on a CRF transcoding mode, a CQP transcoding mode, or an ABR transcoding mode, and the transcoding parameter setting strategy can be the transcoding parameters for each transcoding mode. For example, in a video transcoding scheme based on a CRF transcoding mode, the transcoding parameter setting strategy is a CRF parameter. In both the x264 encoder and the x265 encoder, the CRF parameter range is 0 to 51. A smaller CRF value indicates higher video quality after transcoding, and accordingly, a higher bitrate after transcoding. Here, multiple CRF parameters can be provided, for example, the CRF value can be 26, 27, 28, and 29, without limitation. In some embodiments, in order to further improve the video quality of transcoded videos, the at least one candidate video transcoding scheme can also be obtained by the following operations: obtaining at least one type of second processing scheme, each type of second processing scheme in the at least one type of second processing scheme includes at least one video pre-processing strategy for performing video pre-processing on each video in the video set, and a video pre-processing strategy can be selected from at least one video pre-processing strategy of each type of second processing scheme, and then a transcoding parameter setting strategy can be selected from at least one transcoding parameter setting strategy, and the video pre-processing strategy selected from each type of second processing scheme is combined with the transcoding parameter setting strategy selected from at least one transcoding parameter setting strategy in a preset order to obtain the at least one candidate video transcoding scheme. Specifically, the second processing scheme may include at least one of an image quality enhancement scheme (for example, an image quality enhancement algorithm based on a neural network, etc.), a noise reduction scheme, a sharpening scheme, and a blurring scheme; the preset order is the execution order of each second processing scheme in at least one type of the second processing scheme when it is executed. For example, the preset order may be the execution order of image quality enhancement processing, noise reduction processing, sharpening processing, and blurring processing, or the execution order of noise reduction processing, blurring processing, sharpening processing, and image quality enhancement, and there is no limitation on this. When combining schemes, for example, but not limited to, the noise reduction scheme, the blurring scheme, and the video transcoding scheme based on the CRF transcoding mode may be combined as a candidate video transcoding scheme, or the image quality enhancement scheme, the noise reduction scheme, and the video transcoding scheme based on the CRF transcoding mode may be combined as another candidate video transcoding scheme, etc.

[0060] According to an exemplary embodiment of the present disclosure, each type of second processing scheme includes at least one video pre-processing strategy. For example, the image quality enhancement scheme may include three strategies: De-art (distortion removal) algorithm, De-art-GAN (distortion removal based on generative adversarial network) algorithm, and no image quality enhancement processing of the video. The noise reduction scheme may include two strategies: a noise reduction algorithm based on video noise characteristics and no noise reduction processing of the video. The sharpening scheme may include two strategies: using the USM (Unsharp Mask) sharpening algorithm and not performing sharpening processing on the video. The blurring scheme may include two strategies: using the Gblur (GaussianBlur, Gaussian blur) algorithm and not blurring the video. Of course, in the specific implementation process, the video pre-processing execution strategy can be added or reduced in each type of second processing scheme according to actual conditions, and there is no limitation on this.

[0061] According to an exemplary embodiment of the present disclosure, when combining at least one type of second processing scheme with the first processing scheme in a preset order, a video pre-processing strategy can be selected from at least one video pre-processing strategy of each type of second processing scheme in the at least one type of second processing scheme, and a transcoding parameter setting strategy can be selected from at least one video transcoding strategy, and then the video pre-processing strategy selected from each type of second processing scheme can be combined with the transcoding parameter setting strategy selected from at least one transcoding parameter setting strategy in a preset order. For example, the three video pre-processing strategies of the De-art algorithm, the USM sharpening algorithm, and the non-blurring of the video can be combined with the video transcoding strategy of taking a CRF value of 29 in the execution order of image quality enhancement processing, sharpening processing, and blurring processing to obtain a candidate video processing strategy. Here, according to the execution order of image quality enhancement processing, noise reduction processing, sharpening processing and blurring processing, the aforementioned image quality enhancement scheme includes three strategies, the noise reduction scheme, sharpening scheme and blurring scheme each include two strategies, and the video transcoding scheme includes four strategies as an example. For each video in the video set, there are 96 (M=3*2*2*2*4=96) candidate video transcoding schemes.

[0062] Return to reference Figure 2 In step 202, each video in the at least one video is transcoded based on at least one candidate video transcoding scheme, and a set of video quality evaluation indicators and a set of bitrates are determined for each video in the at least one video. Here, each video in the video set can be transcoded according to each candidate transcoding scheme, and the video quality evaluation indicator and transcoding bitrate of the transcoded video are calculated to form a set of video quality evaluation indicators and a set of bitrates.

[0063] According to an exemplary embodiment of the present disclosure, a video quality evaluation indicator set is a set of video quality evaluation indicators after each video in at least one video is transcoded according to each candidate video transcoding scheme in at least one candidate video transcoding scheme, and the video evaluation indicators include objective quality evaluation indicators and / or subjective quality evaluation indicators. Here, the objective quality evaluation indicators, for example, but not limited to, are at least one of VMAF (Video Multimethod Assessment Fusion), PSNR (Peak Signal to Noise Ratio), SSIM (Structural Similarity) and MS-SSIM (Multiscale Structural Similarity). In the present disclosure, the video quality evaluation indicator set is the VMAF set of the video.

[0064] In step 203, based on the video quality evaluation indicator set, bit rate set and target bit rate of each video in the at least one video, a video transcoding scheme corresponding to each video in the at least one video is determined from at least one candidate video transcoding scheme, where the target bit rate is the expected average bit rate after transcoding the at least one video.

[0065] According to an exemplary embodiment of the present disclosure, based on a video quality evaluation indicator set and a bit rate set for each video in at least one video, and under the condition that the average bit rate of the at least one video does not exceed the target bit rate, a video processing scheme corresponding to each video in the at least one video when the total video quality evaluation indicator of the at least one video reaches a preset standard (for example, when the video quality evaluation indicator is VMAF, the preset standard is the maximum value of the total VMAF) can be determined from the at least one candidate video transcoding scheme. In some embodiments, a dynamic programming method can be used to determine from the at least one candidate video transcoding scheme the video processing scheme corresponding to each video in the at least one video when the total video quality evaluation indicator of the at least one video reaches the preset standard. Specifically, the video quality evaluation indicator set, the bit rate set and the target bit rate can be first input into the dynamic programming algorithm, and then the output of the dynamic programming algorithm can be used as the video transcoding scheme corresponding to each video in the at least one video determined from the at least one candidate video transcoding scheme.

[0066] Taking into account the computational complexity of determining the optimal video transcoding scheme by traversing candidate video transcoding schemes during the actual transcoding process, the present disclosure transforms the problem of determining the optimal video transcoding scheme into predicting the optimal video transcoding strategy by establishing a video transcoding decision model, and combining the predicted optimal video transcoding strategies in a preset order to obtain the optimal video transcoding scheme.

[0067] Figure 3 Detailed description is a flowchart illustrating a method for training a video transcoding decision model according to an exemplary embodiment of the present disclosure.

[0068] Reference Figure 3 In step 301, the video transcoding strategy of the training video can be predicted by the video transcoding decision model to obtain the predicted video transcoding strategy of the training video. Here, the video transcoding decision model can be, for example, but not limited to, a fully connected neural network (for example, a 2-layer fully connected neural network), a CNN (Convolutional Neural Network) or an SVM (Support Vector Machines), etc., and there is no limitation on this. The input of the video transcoding decision model can be the training video or the features of the training video. As an example, the features of the training video can be complexity, blockiness, blur, noise and other features, wherein complexity can be calculated by spatial gradient and other methods, and features such as blockiness, blur, and noise can be obtained by using a feature extraction algorithm.

[0069] In step 302, the predicted loss can be calculated using a loss function based on the predicted video transcoding strategy and the actual video transcoding strategy. Here, the actual video transcoding strategy can be obtained through the target video transcoding scheme, and the target video transcoding scheme is determined in advance by the video transcoding scheme determination method of the present disclosure (for example, Figure 2 Specifically, the target video transcoding scheme may be a transcoding parameter setting strategy, or a combination of at least one video pre-processing strategy and the transcoding parameter setting strategy in a preset order. The video transcoding decision model disclosed herein can predict one transcoding parameter setting strategy or one video pre-processing strategy each time. Therefore, when calculating the predicted loss, when the target video transcoding scheme is a transcoding parameter setting strategy, the predicted video transcoding strategy and the transcoding parameter setting strategy can be used to calculate the predicted loss; and when the target video transcoding scheme is a combination of at least one video pre-processing strategy and the transcoding parameter setting strategy in a preset order, the predicted loss can be calculated using the predicted video transcoding strategy and a corresponding video pre-processing strategy or transcoding parameter setting strategy in the target video transcoding scheme.

[0070] In step 303 , the video transcoding decision model may be trained by adjusting model parameters of the video transcoding decision model according to the prediction loss.

[0071] Figure 4 is a flowchart illustrating a video transcoding method according to an exemplary embodiment of the present disclosure.

[0072] Reference Figure 4 In step 401, at least one video feature of the current video is obtained. Here, the video features include complexity, blockiness, blur, noise, etc. Complexity can be calculated by spatial gradient, etc., and blockiness, blur, noise, etc. can be obtained by feature extraction algorithm.

[0073] In step 402, at least one video feature is input into a corresponding video transcoding decision model to obtain a video transcoding strategy for the current video. Here, the video transcoding decision model is trained using the video transcoding decision model training method disclosed herein.

[0074] According to an exemplary embodiment of the present disclosure, a video transcoding decision model includes a first model, which is used to obtain a transcoding parameter setting strategy. A video feature corresponding to the first model in at least one video feature can be input into the first model to obtain a transcoding parameter setting strategy, and the transcoding parameter setting strategy is used as the video transcoding strategy. In some embodiments, to improve the average quality of the transcoded video, the video transcoding decision model also includes at least one second model, which is used to obtain a video pre-processing strategy. A video feature corresponding to at least one second model in at least one video feature can be input into the at least one second model to obtain at least one video pre-processing strategy. The at least one obtained video pre-processing strategy is then combined with the transcoding parameter setting strategy obtained by the first model in a preset order as the video transcoding strategy. Here, the preset order is the execution order of each type of second processing scheme in at least one type of second processing scheme when it is executed. For example, the preset order can be the execution order of image quality enhancement processing, noise reduction processing, sharpening processing and blurring processing, or it can be the execution order of noise reduction processing, blurring processing, sharpening processing and image quality enhancement, and there is no limitation to this. Specifically, to make the video transcoding strategy obtained by the video transcoding decision model more accurate, a video transcoding decision model can be trained for the first processing scheme and each second processing scheme. The model for the first processing scheme can be used as the first model, and the model for the second processing scheme can be used as the second model. For example, the present disclosure exemplarily trains five video transcoding decision models, which are used to obtain image quality enhancement strategy, noise reduction strategy, sharpening strategy, blurring strategy, and transcoding parameter setting strategy, respectively.

[0075] In step 403, the video processing strategy is used to transcode the current video. Here, for each video in the video set, a video transcoding strategy adapted to the characteristics of each video is derived based on its video features using the video transcoding decision model. Therefore, by performing video transcoding according to the video transcoding method disclosed herein, the average quality of the transcoded video can be improved, thereby significantly reducing the average bitrate of the video while maintaining the same average video quality.

[0076] Figure 5 is a block diagram illustrating an apparatus for determining a video transcoding scheme according to an exemplary embodiment of the present disclosure.

[0077] Reference Figure 5 According to an exemplary embodiment of the present disclosure, a video transcoding scheme determination apparatus 500 may include a data acquisition unit 501 , a parameter determination unit 502 , and a scheme determination unit 503 .

[0078] The data acquisition unit 501 can acquire at least one video and at least one candidate video transcoding scheme; the parameter determination unit 502 can transcode each video in the at least one video based on the at least one candidate video transcoding scheme, and determine a video quality evaluation index set and a bit rate set for each video in the at least one video; the scheme determination unit 503 can determine a video transcoding scheme corresponding to each video in the at least one video from the at least one candidate video transcoding scheme based on the video quality evaluation index set, the bit rate set and the target bit rate of each video in the at least one video, where the target bit rate is the expected average bit rate after transcoding the at least one video.

[0079] because Figure 2 The method for determining the video transcoding scheme shown can be determined by Figure 5 The video transcoding scheme determination device 500 shown in FIG. 1 is used to perform the above operation, and the data acquisition unit 501, the parameter determination unit 502 and the scheme determination unit 503 can respectively perform the same operation as described above. Figure 2 The operations corresponding to steps 201, 202 and 203 in FIG. 2 are as follows: Figure 5 Any details about the operations performed by the units in Figure 2 The corresponding descriptions are not repeated here.

[0080] Figure 6 is a block diagram illustrating a training apparatus for a video transcoding decision model according to an exemplary embodiment of the present disclosure.

[0081] Reference Figure 6 According to an exemplary embodiment of the present disclosure, a training device 600 for a video transcoding decision model may include a transcoding strategy prediction unit 601 , a prediction loss calculation unit 602 , and a model parameter adjustment unit 603 .

[0082] The transcoding strategy prediction unit 601 can predict the video transcoding strategy of the training video through the video transcoding decision model to obtain the predicted video transcoding strategy of the training video; the prediction loss calculation unit 602 can calculate the prediction loss using the loss function based on the predicted video transcoding strategy and the real video transcoding strategy. Here, the real video transcoding strategy can be obtained through the target video transcoding scheme, and the target video transcoding scheme is pre-determined by the video transcoding scheme determination method of the present disclosure (for example, Figure 2 The model parameter adjustment unit 603 trains the video transcoding decision model by adjusting the model parameters of the video transcoding decision model according to the prediction loss.

[0083] because Figure 3 The training method of the video transcoding decision model shown can be obtained by Figure 6 The video transcoding decision model training device 600 is executed, and the transcoding strategy prediction unit 601, the prediction loss calculation unit 602 and the model parameter adjustment unit 603 can respectively execute the same Figure 3 The operations corresponding to steps 301, 302 and 303 in FIG. 301 are as follows: Figure 6 Any details about the operations performed by the units in Figure 3 The corresponding descriptions are not repeated here.

[0084] Figure 7 is a block diagram illustrating a video transcoding apparatus according to an exemplary embodiment of the present disclosure.

[0085] Reference Figure 7 According to an exemplary embodiment of the present disclosure, a video transcoding apparatus 700 may include a feature acquisition unit 701 , a strategy acquisition unit 702 , and a video transcoding unit 703 .

[0086] The feature acquisition unit 701 can obtain at least one video feature of the current video; the strategy acquisition unit 702 can input at least one video feature into the corresponding video transcoding decision model to obtain a video transcoding strategy for the current video. Here, the video transcoding decision model is trained by the training method of the video transcoding decision model disclosed in the present invention; the video transcoding unit 703 can use the video processing strategy to transcode the current video.

[0087] because Figure 4 The video transcoding method shown can be Figure 7 The video transcoding device 700 shown in FIG. 1 is used to perform the above operations, and the feature acquisition unit 701, the strategy acquisition unit 702 and the video transcoding unit 703 can respectively perform the same operations as those in FIG. Figure 4The operations corresponding to steps 401, 402, and 403 in FIG. 4 are as follows: Figure 7 Any details about the operations performed by the units in Figure 4 The corresponding descriptions are not repeated here.

[0088] Figure 8 is a block diagram of an electronic device 800 according to an exemplary embodiment of the present disclosure.

[0089] Reference Figure 8 The electronic device 800 includes at least one memory 801 and at least one processor 802, wherein the at least one memory 801 stores a set of computer-executable instructions. When the computer-executable instruction set is executed by the at least one processor 802, the method for determining the video transcoding scheme or the method for training the video transcoding decision model or the video transcoding method according to the present disclosure is executed.

[0090] As an example, the electronic device 800 can be a PC, a tablet device, a personal digital assistant, a smart phone, or other device capable of executing the above-mentioned instruction set. Here, the electronic device 800 is not necessarily a single electronic device, but can also be any device or circuit that can execute the above-mentioned instructions (or instruction sets) individually or in combination. The electronic device 800 can also be part of an integrated control system or system manager, or can be configured as a portable electronic device that is interconnected with a local or remote (e.g., via wireless transmission) interface.

[0091] In electronic device 800, processor 802 may include a central processing unit (CPU), a graphics processing unit (GPU), a programmable logic device, a dedicated processor system, a microcontroller, or a microprocessor. By way of example and not limitation, the processor may also include an analog processor, a digital processor, a microprocessor, a multi-core processor, a processor array, a network processor, etc.

[0092] The processor 802 can execute instructions or codes stored in the memory 801, wherein the memory 801 can also store data. Instructions and data can also be sent and received over the network via the network interface device, wherein the network interface device can use any known transmission protocol.

[0093] The memory 801 can be integrated with the processor 802, for example, by placing RAM or flash memory within an integrated circuit microprocessor or the like. Furthermore, the memory 801 can include a separate device, such as an external disk drive, a storage array, or any other storage device usable by a database system. The memory 801 and the processor 802 can be operatively coupled or can communicate with each other, for example, via an I / O port, a network connection, or the like, such that the processor 802 can access files stored in the memory.

[0094] In addition, the electronic device 800 may further include a video display (such as a liquid crystal display) and a user interaction interface (such as a keyboard, a mouse, a touch input device, etc.) All components of the electronic device 800 may be connected to each other via a bus and / or a network.

[0095] According to an exemplary embodiment of the present disclosure, a computer-readable storage medium storing instructions may also be provided, wherein when the instructions are executed by at least one processor, the at least one processor is prompted to execute the method for determining a video transcoding scheme or the method for training a video transcoding decision model or the video transcoding method according to the present disclosure. Examples of computer-readable storage media here include: read-only memory (ROM), random access programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), flash memory, non-volatile memory, CD-ROM, CD-R, CD+R, CD-RW, CD+RW, DVD-ROM, DVD-R, DVD+R, DVD-RW, DVD+RW, DVD-RAM, BD-ROM, BD-R, BD-R LTH, BD-RE, Blu-ray or optical disk storage, hard disk drive (HDD), solid state drive (SSD), card storage (such as, multimedia card, secure digital (SD) card or ultra-fast digital (XD) card), magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk and any other device, any other device configured to store the computer program and any associated data, data files and data structures in a non-transitory manner and provide the computer program and any associated data, data files and data structures to a processor or computer so that the processor or computer can execute the computer program. The computer program in the above-mentioned computer-readable storage medium can be run in an environment deployed in a computer device such as a client, a host, an agent device, a server, etc. In addition, in one example, the computer program and any associated data, data files and data structures are distributed on a networked computer system so that the computer program and any associated data, data files and data structures are stored, accessed and executed in a distributed manner by one or more processors or computers.

[0096] According to an exemplary embodiment of the present disclosure, a computer program product may also be provided, the instructions in which may be executed by a processor of a computer device to complete a method for determining a video transcoding scheme or a method for training a video transcoding decision model or a video transcoding method according to an exemplary embodiment of the present disclosure.

[0097] According to the method and device for determining a video transcoding scheme, the method and device for training a video transcoding decision model, and the video transcoding method and device disclosed in the present invention, a more suitable video transcoding scheme is provided for each video sequence based on the characteristics of the video content, which can improve the average quality of the transcoded video, and thus significantly reduce the average bit rate of the video under the same average video quality conditions. In addition, based on the dynamic programming method, a video transcoding scheme with the highest average objective quality under the condition of a given average bit rate can be obtained.

[0098] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the following claims.

[0099] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for determining a video transcoding solution, characterized in that: include: Obtain multiple videos and at least one candidate video transcoding scheme; Transcoding each of the multiple videos according to the at least one candidate video transcoding scheme, and determining a video quality evaluation indicator set and a bit rate set for each of the multiple videos; Based on a video quality evaluation index set, a bit rate set, and a target bit rate for each video in the multiple videos, and under the condition that the average bit rate of the multiple videos does not exceed the target bit rate, a video transcoding scheme corresponding to each video in the multiple videos when the total video quality evaluation index of the multiple videos meets a preset standard is determined by dynamic programming from the at least one candidate video transcoding scheme, wherein the target bit rate is an expected average bit rate after transcoding the multiple videos. The video quality evaluation index set is a set of objective quality evaluation indicators of each video in the multiple videos after being transcoded according to each candidate video transcoding scheme in the at least one candidate video transcoding scheme, and the preset standard is the maximum value of the total objective quality evaluation indicators of the multiple videos. The at least one candidate video transcoding solution is obtained by the following operations: Obtaining a first processing solution, the first processing solution comprising at least one transcoding parameter setting strategy for transcoding the plurality of videos; Acquire multiple types of second processing solutions, each type of the multiple types of second processing solutions including multiple video pre-processing strategies for performing video pre-processing on the multiple videos; Selecting a video pre-processing strategy from the multiple video pre-processing strategies of each type of the second processing scheme; Selecting a transcoding parameter setting strategy from the at least one transcoding parameter setting strategy; Combining the video pre-processing strategy selected from each type of the second processing scheme with the transcoding parameter setting strategy selected from the at least one transcoding parameter setting strategy in a preset order to obtain the at least one candidate video transcoding scheme, The preset order is the execution order of each type of the second processing scheme in the plurality of types of the second processing schemes when being executed. The multiple types of second processing solutions include a first type of second processing solution and a second type of second processing solution, wherein the first type of second processing solution and the second type of second processing solution include different numbers of video pre-processing strategies.

2. A training method for a video transcoding decision model, characterized in that: include: Predicting a video transcoding strategy for a training video using a video transcoding decision model to obtain a predicted video transcoding strategy for the training video; Calculating a predicted loss using a loss function based on the predicted video transcoding strategy and the actual video transcoding strategy, wherein the actual video transcoding strategy is obtained by a target video transcoding scheme, and the target video transcoding scheme is pre-determined by the method for determining a video transcoding scheme according to claim 1; The video transcoding decision model is trained by adjusting model parameters of the video transcoding decision model according to the prediction loss.

3. A video transcoding method, characterized in that: include: Obtain at least one video feature of the current video; Inputting the at least one video feature into a corresponding video transcoding decision model to obtain a video transcoding strategy for the current video, wherein the video transcoding decision model is trained using the video transcoding decision model training method according to claim 2; The video transcoding strategy is used to transcode the current video, wherein the video transcoding decision model includes a first model and multiple second models, the first model is used to obtain a transcoding parameter setting strategy, and the second model is used to obtain a video pre-processing strategy. The at least one video feature is input into the corresponding video transcoding decision model to obtain a video transcoding strategy for the current video, including: Inputting a video feature corresponding to the first model among the at least one video feature into the first model to obtain a transcoding parameter setting strategy; Inputting video features corresponding to the multiple second models in the at least one video feature into the multiple second models to obtain multiple video pre-processing strategies; Combining the multiple video pre-processing strategies with the transcoding parameter setting strategy in a preset order as the video transcoding strategy, The preset order is the execution order of the multiple video pre-processing strategies when they are executed.

4. A device for determining a video transcoding scheme, characterized in that: include: A data acquisition unit is configured to: acquire a plurality of videos and at least one candidate video transcoding scheme; A parameter determination unit is configured to: transcode each of the multiple videos according to the at least one candidate video transcoding scheme, and determine a video quality evaluation indicator set and a bit rate set for each of the multiple videos; The scheme determination unit is configured to: based on a video quality evaluation index set, a bit rate set, and a target bit rate for each video in the multiple videos, determine, using a dynamic programming method, from the at least one candidate video transcoding scheme so that the total video quality evaluation index of the multiple videos meets a preset standard, a video transcoding scheme corresponding to each video in the multiple videos, under the condition that the average bit rate of the multiple videos does not exceed the target bit rate, wherein the target bit rate is an expected average bit rate after the multiple videos are transcoded. The video quality evaluation index set is a set of objective quality evaluation indicators of each video in the multiple videos after being transcoded according to each candidate video transcoding scheme in the at least one candidate video transcoding scheme, and the preset standard is the maximum value of the total objective quality evaluation indicators of the multiple videos. Wherein, the data acquisition unit is configured to: Obtaining a first processing solution, the first processing solution comprising at least one transcoding parameter setting strategy for transcoding the plurality of videos; Acquire multiple types of second processing solutions, each type of the multiple types of second processing solutions including multiple video pre-processing strategies for performing video pre-processing on the multiple videos; Selecting a video pre-processing strategy from the multiple video pre-processing strategies of each type of the second processing scheme; Selecting a transcoding parameter setting strategy from the at least one video transcoding parameter setting strategy; Combining the video pre-processing strategy selected from each type of the second processing scheme with the transcoding parameter setting strategy selected from the at least one transcoding parameter setting strategy in a preset order to obtain the at least one candidate video transcoding scheme, The preset order is the execution order of each type of the second processing scheme in the plurality of types of the second processing schemes when being executed. The multiple types of second processing solutions include a first type of second processing solution and a second type of second processing solution, wherein the first type of second processing solution and the second type of second processing solution include different numbers of video pre-processing strategies.

5. A training device for a video transcoding decision model, characterized in that: include: The transcoding strategy prediction unit is configured to: predict the video transcoding strategy of the training video through the video transcoding decision model to obtain the predicted video transcoding strategy of the training video; a prediction loss calculation unit, configured to: calculate the prediction loss using a loss function based on the predicted video transcoding strategy and the actual video transcoding strategy, wherein the actual video transcoding strategy is obtained by a target video transcoding scheme, and the target video transcoding scheme is pre-determined by the method for determining the video transcoding scheme according to claim 1; The model parameter adjustment unit is configured to train the video transcoding decision model by adjusting the model parameters of the video transcoding decision model according to the prediction loss.

6. A video transcoding device, characterized in that: include: The feature acquisition unit is configured to: acquire at least one video feature of the current video; a strategy acquisition unit configured to: input the at least one video feature into a corresponding video transcoding decision model to obtain a video transcoding strategy for the current video, wherein the video transcoding decision model is trained using the video transcoding decision model training method according to claim 2; The video transcoding unit is configured to: transcode the current video using the video transcoding strategy, The video transcoding decision model includes a first model and multiple second models, the first model is used to obtain a transcoding parameter setting strategy, the second model is used to obtain a video pre-processing strategy, and the strategy acquisition unit is configured to: Inputting a video feature corresponding to the first model among the at least one video feature into the first model to obtain a transcoding parameter setting strategy; Inputting video features corresponding to the multiple second models in the at least one video feature into the multiple second models to obtain multiple video pre-processing strategies; Combining the multiple video pre-processing strategies with the transcoding parameter setting strategy in a preset order as the video transcoding strategy, The preset order is the execution order of the multiple video pre-processing strategies when they are executed.

7. An electronic device, characterized in that: include: at least one processor; at least one memory storing computer-executable instructions, Wherein, when the computer executable instructions are executed by the at least one processor, they prompt the at least one processor to execute the method for determining the video transcoding scheme as described in claim 1 or the method for training the video transcoding decision model as described in claim 2 or the video transcoding method as described in claim 3.

8. A computer-readable storage medium storing instructions, characterized in that: When the instruction is executed by at least one processor, it prompts the at least one processor to execute the method for determining the video transcoding scheme as described in claim 1, the method for training the video transcoding decision model as described in claim 2, or the video transcoding method as described in claim 3.

9. A computer program product comprising computer instructions, characterized in that When the computer instructions are executed by at least one processor, the method for determining a video transcoding scheme according to claim 1 or the method for training a video transcoding decision model according to claim 2 or the video transcoding method according to claim 3 is implemented.

Images