Video processing method, system and device and storage medium

Abstract

The invention relates to the technical field of computers, and discloses a video processing method, system and device and a storage medium, and the video processing method comprises the steps: obtaining a target video file to be published to a video platform; obtaining feature data, wherein the feature data is one or more of video feature data of the target video file, equipment feature data of the video processing equipment and historically published feature data; based on the feature data, predicting a publishing probability that the target video file is successfully published to the video platform; and if the publishing probability is greater than a probability threshold value, pre-publishing the target video file in the pre-processing process before the target video file is published.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a video processing method, system, device and storage medium. Background Art

[0002] Pre-release means that during the process of video creation, some tasks of video release are performed in advance, such as video transcoding and synthesis tasks and uploading tasks of some videos. In this way, when users complete video creation and release videos, the waiting time of users can be greatly reduced, improving user experience.

[0003] However, the problem with pre-publishing is that some videos may not be actually published due to various reasons (such as user canceling the publication, device power failure, etc.), and thus the pre-publishing of these videos becomes a waste of bandwidth, affecting the bandwidth usage of other normal services.

[0004] In view of this, it is necessary to provide a video processing method that reduces bandwidth waste. Summary of the invention

[0005] In view of this, the embodiments of the present disclosure provide a video processing method, a video processing system, an electronic device, and a computer-readable storage medium, which can reduce bandwidth waste in pre-release scenarios.

[0006] On the one hand, the present disclosure provides a video processing method, the method comprising:

[0007] Get the target video file to be published to the video platform;

[0008] Acquire feature data, where the feature data is one or more of video feature data of the target video file, device feature data of a video processing device, and historically released feature data;

[0009] Based on the feature data, predicting the probability of the target video file being successfully published to the video platform;

[0010] If the release probability is greater than the probability threshold, the target video file is pre-released during the preprocessing process before the target video file is released.

[0011] Another aspect of the present disclosure further provides a video processing system, the system comprising:

[0012] The video acquisition module is used to acquire the target video file to be published to the video platform;

[0013] A feature acquisition module, used to acquire feature data, wherein the feature data is one or more of video feature data of the target video file, device feature data of a video processing device, and historically released feature data;

[0014] A prediction module, used for predicting the probability of the target video file being successfully published to the video platform based on the feature data;

[0015] The pre-publishing module is used to pre-publish the target video file during the pre-processing process before publishing the target video file if the publishing probability is greater than the probability threshold.

[0016] Another aspect of the present disclosure further provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the method described above is implemented.

[0017] Another aspect of the present disclosure provides an electronic device, which includes a processor and a memory, wherein the memory is used to store a computer program, and when the computer program is executed by the processor, the method described above is implemented.

[0018] In some embodiments of the present application, after acquiring the target video file, the release probability of the target video file can be predicted based on the feature data, and the target video file can be pre-released when the release probability is greater than the probability threshold. In this way, the target video file can be pre-released only when the target video file has a high probability of being successfully released to the video platform, avoiding the situation where the target video file is pre-released when the release probability of the target video file is low and may not be successfully released to the video platform, thereby reducing the problem of bandwidth waste in the pre-release scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The features and advantages of the present disclosure will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the present disclosure in any way. In the accompanying drawings:

[0020] Figure 1 A schematic diagram of a video publishing and playback scenario provided by an embodiment of the present application is shown;

[0021] Figure 2 A schematic diagram showing a flow chart of a video processing method provided by an embodiment of the present application is shown;

[0022] Figure 3 A schematic diagram of a process flow when a video processing device provided by an embodiment of the present application publishes a video file is shown;

[0023] Figure 4A module schematic diagram of a video processing system provided by an embodiment of the present application is shown;

[0024] Figure 5 A schematic diagram of an electronic device provided by an embodiment of the present application is shown. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solutions and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present disclosure.

[0026] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0027] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0028] As an optional but non-limiting implementation, in response to receiving an active request from the user, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0029] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet the relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0030] See also Figure 1 , which is a schematic diagram of a video publishing and playback scenario provided by an embodiment of the present application. Figure 1In the embodiment, the video processing device 11 may include a mobile phone, a computer and other devices. The video platform 12 may include a server and other devices. The video processing device 11 is connected to the video platform 12 in communication. On the one hand, the video processing device 11 may upload the video files stored locally or the video files collected by the video acquisition device to the video platform 12, and the video platform 12 stores and manages the video files. On the other hand, the video processing device 11 may also receive and play the video files sent by the video platform 12.

[0031] based on Figure 1 , this application provides a video processing method applied to a video processing device, which can reduce bandwidth waste in pre-release scenarios. Figure 2 , which is a flow chart of a video processing method provided in one embodiment of the present application. Figure 2 In the video processing method, the video processing method comprises the following steps:

[0032] Step S21, obtaining the target video file to be published to the video platform.

[0033] Specifically, when executing the method of the present application, the video processing device can display a file acquisition interface. The file acquisition interface may include one or more file acquisition entries. Different file acquisition entries correspond to different ways of acquiring the target video file. When a file acquisition entry is triggered, the target video file can be acquired according to the acquisition path corresponding to the file acquisition entry. For example, when file acquisition entry A is triggered, the video file locally stored in the video processing device can be displayed. Among the displayed video files, in response to a file selection operation, the selected video file can be used as the target video file. When file acquisition entry B is triggered, a video acquisition device (such as a camera) integrated in the video processing device can be started to acquire video, and the acquired video file can be used as the target video file.

[0034] In this embodiment, the file acquisition interface is specifically a video acquisition interface, that is, the target video file is acquired by video acquisition by default. The video acquisition interface may also include a file acquisition entry, and when the file acquisition entry is triggered, the target video file can be selected from the target video files locally stored in the video processing device.

[0035] Step S22, acquiring feature data, where the feature data is one or more of the video feature data of the target video file, the device feature data of the video processing device, and the historically released feature data.

[0036] Specifically, the video feature data refers to the content attributes and / or file attributes of the target video file, including but not limited to the content type, content source, resolution, whether special effects are used, file format, etc. of the target video file. Among them, whether special effects are used includes but is not limited to whether stickers are used, whether image quality is enhanced, etc.

[0037] Device characteristic data refers to device attributes of the video processing device, including but not limited to the remaining power, model, operating system, etc. of the video processing device.

[0038] Historical publishing feature data refers to the results obtained by analyzing and counting the publishing operations that have been performed by the video processing device, including but not limited to the number of times the video processing device has published within a specified historical time before the target video file, the number of drafts stored, the number of retries after failed publishing, the number of canceled publishing, the average publishing time, etc. Video Processing Device

[0039] In this embodiment, the video platform can analyze and count the publishing operations that have been performed by the video processing device. After the video processing device obtains the target video file, it can send a data acquisition request to the video platform to obtain historical publishing feature data from the video platform. At the same time, the video processing device can analyze the obtained target video file to obtain video feature data, and the video processing device can monitor its own device properties to obtain device feature data. In this way, feature data is obtained.

[0040] Step S23, based on the feature data, predict the probability of the target video file being successfully published to the video platform.

[0041] It is understandable that after obtaining the target video file, whether the target video file can be successfully published to the video platform may be affected by many factors. For example, before the target video file is published, the video processing device is shut down due to insufficient power, which may cause the target video file to fail to be published; for example, before the target video file is published, the button representing the cancellation of publication displayed by the video processing device is triggered, which may also cause the target video file to fail to be published. That is, whether the target video file can be successfully published to the video platform is probabilistic.

[0042] In this embodiment, the feature data can be input into a trained prediction model, and the prediction model predicts the probability of the target video file being successfully published to the video platform based on the feature data. The publishing probability represents the probability of the target video file being successfully published to the video platform. The greater the publishing probability, the higher the probability of the target video file being successfully published to the video platform.

[0043] Specifically, the prediction model can be obtained by training the video platform based on the historical feature data of one or more video processing devices. Before the video processing device executes the method of the present application, or during the execution of the method of the present application, the video platform can send the trained prediction model to the video processing device, so that the video processing device can predict the release probability of the target video file based on the prediction model. Among them, the historical feature data refers to the historical video files released by one or more video processing devices before the target video file. The historical video files may include a first historical video file that is successfully released and a second historical video file that fails to be released. The historical feature data may include the video feature data of the historical video file, the device feature data after the video processing device obtains the historical video file, and the historical release feature data of the video processing device before the historical video file. For ease of understanding, the following is explained by example. Assuming that before the target video file, video processing device 1 has released video file A, video file B and video file C, and video processing device 2 has released video file D and video file E, then video files A, B, C, D, and E can be called historical video files. The historical feature data may include:

[0044] Video feature data of video files A, B, C, D, and E;

[0045] The device characteristic data after the video processing device 1 obtains the video files A, B, and C, and the device characteristic data after the video processing device 2 obtains the video files D and E;

[0046] The video processing device 1 releases the feature data of the history of video files A, B, and C, and the video processing device 2 releases the feature data of the history of video files D and E.

[0047] The relevant model training principle is a technology that the technical personnel of this application should be familiar with, and this application will not be repeated here. After the prediction model training is completed, the feature data corresponding to the target video file is input into the prediction model, and the prediction model can judge the probability of the target video file being successfully published to the video platform based on the input feature data.

[0048] Step S24: If the release probability is greater than the probability threshold, the target video file is pre-released during the pre-processing process before the target video file is released.

[0049] In some embodiments, pre-processing before publishing includes editing the target video file, such as adding background music to the target video file, adjusting the color of the target video file, specifying the playback time of the target video file, adding stickers to the target video file, etc.

[0050] In some embodiments, pre-processing before release includes setting the permission attributes of the target video file on the video platform, such as setting whether to allow the entity of the target video file to be downloaded from the video platform to the local client, setting the client device that allows the target video file to be viewed on the video platform, setting the client device that allows the target video file to be forwarded on the video platform, etc.

[0051] In this embodiment, after acquiring the target video file, the video processing device may first display a video editing page. In the video editing page, the target video file may be edited. After the editing process is completed, you can jump from the video editing page to the video publishing page. The video publishing page may include a permission setting area and a publish button. In the permission setting area, the permission attributes of the target video file on the video platform may be set. After completing the setting of the permission attributes, in response to the publish button being triggered, the pre-processed target video file may be formally published to the video platform. Among them, in the process of pre-processing the target video file on the video editing page and the video publishing page, if the probability of publishing the target video file is greater than the probability threshold, the target video file may be pre-published at the same time; if the probability of publishing the target video file is less than or equal to the probability threshold, the target video file may not need to be pre-published.

[0052] In summary, in some embodiments of the present application, after acquiring the target video file, the release probability of the target video file can be predicted based on the feature data, and the target video file can be pre-released when the release probability is greater than the probability threshold. In this way, the target video file can be pre-released only when the target video file has a high probability of being successfully released to the video platform, avoiding the situation where the target video file is pre-released when the release probability of the target video file is low and may not be successfully released to the video platform, thereby reducing the problem of bandwidth waste in the pre-release scenario.

[0053] The solution of this application is further described below.

[0054] In some embodiments, among the video feature data, device feature data, and historically released feature data included in the feature data, some video feature data and some device feature data may change over time, that is, the feature data at different time points may not be exactly the same. For example, in the process of preprocessing the target video file, device feature data such as the remaining power of the video processing device may change over time. For another example, before editing the target video file, the video feature data that can be obtained may only include the content type, content source, and resolution of the target video file. However, after editing the target video file, in addition to obtaining the content type, content source, and resolution of the target video file, the special effects data of the target video file can also be obtained, that is, whether the target video file uses stickers and whether image quality enhancement is performed.

[0055] Based on the feature data at different time points, the release probability can be predicted separately. In theory, the richer the feature data, the more accurate the release probability can be. However, when the feature data is richer, the time point of predicting the release probability is usually later, which will affect the duration of pre-release. For example, based on the first feature data obtained before the editing process, the first release probability can be predicted; based on the second feature data obtained after the editing process, the second release probability can be predicted. Since the second feature data can include special effects data added to the target video file during the editing process, the second feature data can be richer than the first feature data, so the second release probability can be more accurate than the first release probability. However, the time point for determining the first release probability is before the editing process, and the time point for determining the second release probability is after the editing process. Based on the first release probability, pre-release can be performed before the editing process, so that more release tasks can be completed and the waiting time for the formal release can be reduced. Based on the second release probability, pre-release can only be performed after the editing process, and the pre-release time is relatively short, and relatively few release tasks can be completed, which may increase the waiting time for the formal release.

[0056] In view of the reasons mentioned above, in the preprocessing process before publishing the target video file, there may be multiple time points for obtaining feature data, and the feature data obtained at different time points are not exactly the same. The above-mentioned pre-publishing of the target video file may include:

[0057] According to the order of time points, if the release probability of the target video file is predicted to be greater than the probability threshold for the first time based on the feature data obtained at a time point, the target video file is pre-released starting from this time point.

[0058] For example, suppose there are three time points: 10:00, 10:01, and 10:02. According to the order of the time points, first based on the feature data obtained at 10:00, it is predicted that the release probability of the target video file is less than the probability threshold, so at 10:00, the target video file is not pre-released. Then based on the feature data obtained at 10:01, it is predicted that the release probability of the target video file is greater than the probability threshold, that is, the first prediction that the release probability of the target video file is greater than the probability threshold, then the target video file is pre-released starting from 10:01.

[0059] In this way, on the one hand, the prediction accuracy can be improved. For example, according to the chronological order of the time points, after predicting that the release probability of the target video file is less than the probability threshold at the first time point with less feature data, the release probability of the target video file can be continued to be predicted at the second time point with more feature data, so as to prevent the release probability at the first time point from being inaccurate, and the target video file is not pre-released when it should be pre-released. By predicting the release probability multiple times, the prediction accuracy can be improved. On the other hand, when the release probability of the target video file is predicted to be greater than the probability threshold for the first time, the target video file is pre-released, which can ensure sufficient pre-release time and reduce the release waiting time.

[0060] In this embodiment, the target video file is pre-published, specifically including:

[0061] At a first time point after entering the video editing page, first feature data is acquired, and a first release probability of the target video file is predicted based on the first feature data;

[0062] If the first release probability is greater than the probability threshold, pre-release the target video file after the first time point;

[0063] If the first release probability is less than or equal to the probability threshold, at a second time point after entering the video release page, obtaining second feature data, and predicting a second release probability of the target video file based on the second feature data;

[0064] If the second release probability is greater than the probability threshold, the target video file is pre-released after the second time point.

[0065] Among them, obtaining the first characteristic data after entering the video publishing page, and obtaining the second characteristic data after entering the video editing page, can ensure that the first characteristic data and the second characteristic data are not exactly the same. Specifically, the second characteristic data obtained after entering the video publishing page can be richer than the first characteristic data obtained after entering the video editing page. In the case where the first release probability predicted based on the first characteristic data is less than or equal to the probability threshold, re-predicting the second release probability of the target video file based on the second characteristic data can improve the accuracy of the prediction. In the case where the first release probability predicted based on the first characteristic data is greater than the probability threshold, pre-releasing the target video file from the first time point can ensure sufficient pre-releasing time and reduce the waiting time for release.

[0066] In some embodiments, when acquiring the first feature data, the feature data before entering the video editing page can be acquired as the first feature data, and when acquiring the second feature data, the feature data before entering the video publishing page can be acquired as the second feature data. In simple terms, on the video editing page, the feature data that can be acquired before entering the video editing page is used as the first feature data; and on the video publishing page, the feature data that can be acquired before entering the video publishing page is used as the second feature data. In this way, when predicting the release probability on the video editing page or the video publishing page, the feature data will not change with time, that is, the feature data is stable, which is convenient for predicting the release probability.

[0067] Further, in some embodiments, an attribute or a result in the feature data is referred to as a feature. For example, the content type of the target video file can be referred to as a feature, and the power level of the target video device can be referred to as another feature. The first feature data obtained on the video editing page and the second feature data obtained on the video publishing page may have different feature compositions. For example, the video feature data in the first feature data may only include the content type, content source, and resolution of the target video file. After the target video file is edited on the video editing page, the video feature data in the second feature data may include special effects data of the target video file in addition to the content type, content source, and resolution of the target video file, that is, whether the target video file uses stickers, whether the image quality is enhanced, etc. If the same prediction model is used when predicting the first release probability based on the first feature data and predicting the second release probability based on the second feature data, the accuracy of the prediction model will inevitably be affected, and the accuracy of the obtained release probability will be low.

[0068] In view of this, in some embodiments, predicting the first release probability of the target video file based on the first feature data may include:

[0069] Inputting the first feature data into a first prediction model, and having the first prediction model predict a first release probability of the target video file based on the first feature data;

[0070] The above-mentioned predicting the first release probability of the target video file based on the first feature data may include:

[0071] Inputting the second feature data into a second prediction model, and having the second prediction model predict a second release probability of the target video file based on the second feature data;

[0072] The first prediction model and the second prediction model are different prediction models. Specifically, the first prediction model and the second prediction model may have different model structures or different model parameters. In this way, the first prediction model may focus on the feature composition of the first feature data, and the second prediction model may focus on the feature composition of the second feature data, which may improve the model accuracy and make the predicted release probability more accurate.

[0073] Furthermore, in some embodiments, when predicting the release probability based on feature data, each feature in the feature data may have its own corresponding weight. The larger the weight of a feature, the more important the feature is in the release probability prediction. Conversely, the smaller the weight of a feature, the less important the feature is in the release probability prediction. For features with small weights, these features do not contribute much to the release probability prediction, and may even interfere with the release probability prediction. In view of this, the above-mentioned prediction of the release probability of the target video file being successfully released to the video platform based on feature data may include:

[0074] Obtain the importance score of each feature in the feature data. The importance score represents the weight of each feature when predicting the release probability.

[0075] From multiple features, select the target features whose importance scores are higher than the score threshold;

[0076] Based on the target features, predict the probability of the target video file being successfully published to the video platform.

[0077] For example, suppose the acquired feature data includes the content type, content source, resolution, whether special effects are used, file format, remaining power, model, and operating system of the target video file, and the importance score of the operating system is lower than the scoring threshold, and the importance scores of other features except the operating system are higher than the scoring threshold. Then, the features except the operating system can be used as target features, and the release probability of the target video file can be predicted based on the target features. In this way, the accuracy of the release probability can be improved.

[0078] Specifically, the prediction model may be an xgboost algorithm model. After all acquired feature data are input into the xgboost algorithm model, the xgboost algorithm model may screen out target features whose importance scores are higher than a score threshold, and predict the release probability of the target video file based on the target features.

[0079] Furthermore, in some embodiments, based on the feature data, predicting the probability of the target video file being successfully published to the video platform may include:

[0080] Perform a preprocessing operation on part or all of the feature data, and based on the feature data after the preprocessing operation, predict the probability of the target video file being successfully published to the video platform, wherein the preprocessing operation includes one or more of the following:

[0081] 1) If there is a first feature whose value range is a discrete value in the feature data, the value of the first feature is encoded.

[0082] For ease of understanding, let's take the feature "content type" as an example. Assume that the value range of "content type" is "tourism", "life", and "technology". Since "tourism", "life", and "technology" are discrete values, "content type" can be used as the first feature with a discrete value range. The value of "content type" can be one of "tourism", "life", and "technology".

[0083] For the first feature, the value of the first feature can be encoded. Different values ​​have different encodings. For example, take the content type of the target video file as an example. When the value of the content type is "tourism", "01" can be used instead of "tourism"; when the value of the content type is "life", "02" can be used instead of "tourism". Here, 01 and 02 are the encodings corresponding to the values ​​of the first feature. Using encoding to replace the value of the first feature can enable the prediction model to better identify the value of the first feature, thereby improving the prediction accuracy of the prediction model.

[0084] 2) If there is a second feature whose value range is a continuous value in the feature data, the value of the second feature is bucketed.

[0085] For ease of understanding, take the remaining power of a video processing device as an example. Assume that the value range of the remaining power is "0% to 100%". Since "0% to 100%" is a continuous value, the remaining power can be used as the second feature with a continuous value range. The value of the remaining power can be one of "0% to 100%".

[0086] The value range of the second feature is divided into multiple non-overlapping intervals, each of which is called a bucket. Each bucket has a number, and different buckets have different numbers. Bucketing the value of the second feature is to determine the target interval range where the value of the second feature is located, and use the number of the target interval range to replace the value of the second feature. In this way, the prediction model can better identify the value of the second feature, which is conducive to improving the prediction accuracy of the prediction model.

[0087] At this point, the relevant description of the video processing method of the present application is completed. The following is a description of the operation scenario of the video processing method.

[0088] In some embodiments, the video processing method of the present application is mainly operated during the network peak period. That is, during the network peak period, in order to avoid bandwidth waste, only the target video files whose release probability is greater than the probability threshold are pre-released. During non-network peak periods, there is no need to predict the release probability of the target video files, and all the acquired target video files can be pre-released directly. In this way, the user experience is improved.

[0089] In some embodiments, the video processing device has a pre-release function switch. Only when the pre-release function switch is turned on, the video processing method of the present application needs to be run to pre-release the target video file whose release probability is greater than the probability threshold.

[0090] In summary, please refer to Figure 3 , is a flow chart of a video processing device providing a video file publishing process according to an embodiment of the present application. Figure 3 The steps include:

[0091] Step S31, obtaining the target video file to be published to the video platform. For details, please refer to the relevant description of step S21, which will not be repeated here.

[0092] Step S32, determining whether the pre-release function of the video processing device is turned on. If so, executing step S33, if not, executing step S36.

[0093] Step S33, determine whether the time point of acquiring the target video file is during the network peak period. If yes, execute step S34, if not, execute step S36.

[0094] Step S34, determining whether the release probability of the target video file is greater than a probability threshold. If so, executing step S36, if not, executing step S35.

[0095] Step S35: Do not pre-publish the target video file.

[0096] Step S36, pre-publishing the target video file.

[0097] See also Figure 4 , is a module diagram of a video processing system provided by an embodiment of the present application. Figure 4 In the video processing system, the video processing system includes:

[0098] The video acquisition module is used to acquire the target video file to be published to the video platform;

[0099] A feature acquisition module, used to acquire feature data, wherein the feature data is one or more of video feature data of a target video file, device feature data of a video processing device, and historically released feature data;

[0100] A prediction module, used to predict the probability of a target video file being successfully published to a video platform based on feature data;

[0101] The pre-publishing module is used to pre-publish the target video file during the pre-processing process before publishing the target video file if the publishing probability is greater than the probability threshold.

[0102] See also Figure 5 , is a schematic diagram of an electronic device provided by an embodiment of the present application. The electronic device includes a processor and a memory, the memory is used to store a computer program, and when the computer program is executed by the processor, the above method is implemented.

[0103] The processor may be a central processing unit (CPU). The processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above chips.

[0104] As a non-transitory computer-readable storage medium, the memory can be used to store non-transitory software programs, non-transitory computer executable programs and modules, such as program instructions / modules corresponding to the method in the embodiment of the present invention. The processor executes various functional applications and data processing of the processor by running the non-transitory software programs, instructions and modules stored in the memory, that is, implementing the method in the above method embodiment.

[0105] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the processor, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0106] One embodiment of the present application further provides a computer-readable storage medium, which is used to store a computer program. When the computer program is executed by a processor, the above method is implemented.

[0107] Although the embodiments of the present disclosure have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present disclosure, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A video processing method, characterized in that: The method comprises: Get the target video file to be published to the video platform; Acquire feature data, where the feature data is one or more of video feature data of the target video file, device feature data of a video processing device, and historically released feature data; Based on the feature data, predicting the probability of the target video file being successfully published to the video platform; If the release probability is greater than the probability threshold, the target video file is pre-released during the preprocessing process before the target video file is released.

2. The method according to claim 1, characterized in that In the preprocessing process before publishing the target video file, there are multiple time points for obtaining feature data, and the feature data obtained at different time points are not completely the same; The pre-publishing of the target video file includes: According to the order of time points, if the release probability of the target video file is predicted to be greater than the probability threshold for the first time based on the feature data obtained at a time point, the target video file is pre-released starting from this time point.

3. The method according to claim 2, characterized in that The pre-publishing of the target video file includes: At a first time point after entering the video editing page, first feature data is acquired, and a first release probability of the target video file is predicted based on the first feature data; If the first release probability is greater than the probability threshold, pre-releasing the target video file after the first time point; If the first release probability is less than or equal to the probability threshold, at a second time point after entering the video release page, obtaining second feature data, and predicting a second release probability of the target video file based on the second feature data; If the second release probability is greater than the probability threshold, the target video file is pre-released after the second time point.

4. The method according to claim 3, characterized in that The obtaining of the first characteristic data comprises: Acquire feature data before entering the video editing page as the first feature data; The obtaining of the second characteristic data comprises: The feature data before entering the video publishing page is obtained as the second feature data.

5. The method according to claim 3, characterized in that The predicting the first release probability of the target video file based on the first feature data includes: Inputting the first feature data into a first prediction model, and having the first prediction model predict a first release probability of the target video file based on the first feature data; The predicting the first release probability of the target video file based on the first feature data includes: Inputting the second feature data into a second prediction model, and having the second prediction model predict a second release probability of the target video file based on the second feature data; The first prediction model and the second prediction model are different prediction models.

6. The method according to any one of claims 1 to 5, characterized in that: The predicting, based on the feature data, the probability of the target video file being successfully published to the video platform includes: Obtaining an importance score for each feature in the feature data, wherein the importance score represents a weight of each feature when predicting the release probability; Filtering out target features whose importance scores are higher than a score threshold from the multiple features; Based on the target features, the probability of the target video file being successfully published to the video platform is predicted.

7. The method according to any one of claims 1 to 5, characterized in that: The predicting, based on the feature data, the probability of the target video file being successfully published to the video platform includes: Performing a preprocessing operation on part or all of the feature data, and predicting the probability of the target video file being successfully published to the video platform based on the feature data after the preprocessing operation, wherein the preprocessing operation includes one or more of the following: If a first feature whose value range is a discrete value exists in the feature data, encoding the value of the first feature; If a second feature whose value range is a continuous value exists in the feature data, the value of the second feature is bucketed.

8. A video processing system, characterized in that: The system comprises: The video acquisition module is used to acquire the target video file to be published to the video platform; A feature acquisition module, used to acquire feature data, wherein the feature data is one or more of video feature data of the target video file, device feature data of a video processing device, and historically released feature data; A prediction module, used for predicting the probability of the target video file being successfully published to the video platform based on the feature data; The pre-publishing module is used to pre-publish the target video file during the pre-processing process before publishing the target video file if the publishing probability is greater than the probability threshold.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: The electronic device comprises a processor and a memory, wherein the memory is used to store a computer program, and when the computer program is executed by the processor, the method according to any one of claims 1 to 7 is implemented.

Images