Video transcoding decision-making method and device, equipment, storage medium and product
By combining historical popularity and video feature information to determine the transcoding value of broadcasters, and then filtering and sorting them, the problem of resource waste in existing technologies is solved, and more efficient utilization and allocation of video transcoding resources are achieved.
Patent Information
- Application Number
- CN202510924255.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-04
- Publication Date
- 2025-11-07
AI Technical Summary
Existing video transcoding decision-making methods rely on the popularity of a single live stream, resulting in a waste of video transcoding resources and an inability to effectively utilize the valuable resources of live streamers who are highly valuable for video transcoding.
By acquiring the historical popularity information and video history data of the anchors for initial screening, and combining video feature information to determine the transcoding value, the videos are sorted and filtered according to the transcoding value to select target users for video transcoding.
It improves the utilization rate of video transcoding resources, reduces resource waste, enables a more comprehensive analysis of the transcoding value of broadcasters, and improves the efficiency of transcoding decisions and the balanced allocation of resources.
Smart Images
Figure CN120915970A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] Embodiments of the present application relate to the technical field of computer, and in particular, to a video transcoding decision method and device, equipment, storage medium and product. BACKGROUND
[0002] With the development of computer technology and Internet technology, the application of video live streaming platforms is also becoming more and more widespread, and the number of video live streaming users is also increasing. In order to improve video quality, the video provided by the host is often transcoded and then different level videos are issued to the audience.
[0003] In order to improve the utilization efficiency of video transcoding resources, video transcoding decisions need to be made for multiple hosts to select hosts for video transcoding processing from multiple hosts. Currently, video transcoding decisions are generally based on live streaming popularity, for example, hosts with high live streaming popularity are selected as hosts for video transcoding processing. However, this video transcoding decision relies on a single live streaming popularity, which can easily lead to hosts with low video transcoding value being selected as hosts for video transcoding processing, resulting in waste of video transcoding resources. SUMMARY
[0004] Embodiments of the present application provide a video transcoding decision method, device, equipment, storage medium and product, which obtains multiple second users by performing first screening processing on multiple first users according to historical popularity information and video historical data, determines the transcoding value of the multiple second users according to video feature information and historical popularity information, and performs second screening processing on the multiple second users by sorting the multiple second users according to the transcoding value to obtain multiple target users for video transcoding processing, effectively solving the technical problem that video transcoding decisions in related technologies rely on a single live streaming popularity, which can easily lead to waste of video transcoding resources, and more comprehensively analyzing the video transcoding value of different hosts to reduce waste of video transcoding resources and improve the utilization rate of video transcoding resources.
[0005] In a first aspect, embodiments of the present application provide a video transcoding decision method, comprising: obtaining historical popularity information and video historical data of multiple first users, and performing first screening processing on the multiple first users according to the historical popularity information and the video historical data to obtain multiple second users; obtaining video feature information of the multiple second users, and determining the transcoding value of the multiple second users according to the historical popularity information and the video feature information; performing sorting processing on the multiple second users according to the transcoding value to obtain transcoding ranking information of the multiple second users; According to the transcoding ranking information, the second screening module performs a second screening process on the second users to obtain target users, the target users being users determined to perform video transcoding processing.
[0006] In a second aspect, an embodiment of the present application provides a video transcoding decision device, comprising a first screening module, a value analysis module, a ranking processing module and a second screening module, wherein: The first screening module is configured to obtain historical heat information and video historical data of a plurality of first users, perform a first screening process on the first users according to the historical heat information and the video historical data, and obtain second users. The value analysis module is configured to obtain video feature information of the second users, determine transcoding values of the second users according to the historical heat information and the video feature information. The ranking processing module is configured to perform an ordering process on the second users according to the transcoding values to obtain transcoding ranking information of the second users. The second screening module is configured to perform a second screening process on the second users according to the transcoding ranking information to obtain target users, the target users being users determined to perform video transcoding processing.
[0007] In a third aspect, an embodiment of the present application provides a video transcoding decision device, comprising a memory and one or more processors. The memory is configured to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the video transcoding decision method of the first aspect.
[0008] In a fourth aspect, an embodiment of the present application provides a non-volatile storage medium storing computer executable instructions, the computer executable instructions being used to execute the video transcoding decision method of the first aspect when executed by a computer processor.
[0009] In a fifth aspect, an embodiment of the present application provides a computer program product, the computer program product comprising a computer program stored in a computer readable storage medium, at least one processor of a device reading and executing the computer program from the computer readable storage medium, so that the device executes the video transcoding decision method of the first aspect.
[0010] The embodiment of the application obtains a plurality of second users by performing first screening processing on a plurality of first users according to historical hotness information and video historical data of the plurality of first users, obtains video feature information of the plurality of second users, determines transcoding values of the plurality of second users according to the historical hotness information and the video feature information, and performs sorting processing on the plurality of second users according to the transcoding values to obtain transcoding ranking information of the plurality of second users. The plurality of second users can be subjected to second screening processing according to the transcoding ranking information to obtain a plurality of target users subjected to video transcoding processing. The first screening processing can effectively reduce the waste of video transcoding decision resources caused by users with low transcoding values, improve the video transcoding decision efficiency, and consider the transcoding values of users from multiple dimensions of historical hotness information and video historical data. The video transcoding decision no longer depends on a single hotness, effectively reduces the case that users with low video transcoding values are decided to be users needing video transcoding, can more comprehensively analyze the video transcoding values of different anchors, reduces the waste of video transcoding resources, and improves the utilization rate of video transcoding resources. BRIEF DESCRIPTION OF DRAWINGS
[0011] Figure 1 is a flowchart of a video transcoding decision method provided by the embodiment of the application; Figure 2 is a flowchart of another video transcoding decision method provided by the embodiment of the application; Figure 3 is a structural schematic diagram of a video transcoding decision device provided by the embodiment of the application; Figure 4 is a structural schematic diagram of a video transcoding decision device provided by the embodiment of the application. DETAILED DESCRIPTION
[0012] In order to make the objectives, technical solutions and advantages of the present application clearer, the following further describes the specific embodiments of the present application with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application. In addition, it should be noted that, for the convenience of description, only parts related to the present application are shown in the drawings, but not all contents. Before discussing the example embodiments in more detail, it should be mentioned that some example embodiments are described as processes or methods depicted by flowcharts. Although the flowcharts describe the operations (or steps) as sequential processes, many of the operations can be implemented in parallel, concurrently or simultaneously. In addition, the order of the operations can be rearranged. The above processes can be terminated when the operations are completed, but can also have additional steps not included in the drawings. The above processes can correspond to methods, functions, procedures, subroutines, subprograms, etc.
[0013] The video transcoding decision method provided in the application can be applied to the video transcoding decision scene of a video live broadcast platform, and aims to more comprehensively analyze the video transcoding value of different anchors, reduce the waste of video transcoding resources, and improve the utilization rate of video transcoding resources.
[0014] In video transcoding, the number of servers of a live broadcast platform is usually limited, and the cost of a GPU card is usually higher than that of a CPU server, the cost of a CPU server is higher than that of hardware transcoding, and the cost of hardware transcoding is higher than that of no transcoding. There are often contradictions between the number of anchors and the number of servers, and contradictions between the transcoding value of anchors and the viewing popularity of anchors. The transcoding cost (total server cost) and the number of servers invested by the live broadcast platform are fixed, and the number of anchors is usually higher than the number of servers. In this case, it is difficult to perform video transcoding on all anchors, resulting in an inevitable contradiction between the number of anchors and the limited number of servers in the video transcoding link. In addition, video transcoding for anchors can improve the picture quality and viewing fluency, but the improvement gradient changes with the video content, and the gradient of improvement can reflect the value of transcoding for anchors, that is, under the same server resource investment, the income brought by different videos is different. From this point of view, it is more appropriate for the server to be allocated to anchors with high transcoding value. However, from the perspective of the audience, as long as the anchor has positive transcoding value, the higher the viewing popularity of the anchor, the higher the viewing income brought by the anchor. Since the transcoding value of the anchor and the video viewing popularity are independent of each other, the allocation of the server needs to balance between the transcoding value and the viewing popularity. In the existing video transcoding decision scheme, it is generally based on live broadcast popularity, for example, anchors with high live broadcast popularity are selected as anchors for video transcoding. This video transcoding decision scheme depending on single live broadcast popularity is easy to decide that anchors with low video transcoding value but high popularity need to be video transcoded, resulting in waste of video transcoding resources, or anchors with high video transcoding value but low popularity do not need to be video transcoded, and the video transcoding resources are not fully utilized. Based on this, the video transcoding decision method provided in the embodiments of the application is provided to solve the technical problem that the existing video transcoding decision scheme is easy to cause waste of video transcoding resources.
[0015] Figure 1 A flowchart of the video transcoding decision method provided in the embodiments of the application is given. The video transcoding decision method provided in the embodiments of the application can be executed by a video transcoding decision device, which can be realized by hardware and / or software and integrated in a video transcoding decision equipment (such as a transcoding system).
[0016] The video transcoding decision method executed by the video transcoding decision device is described below. Referring to Figure 1 , the video transcoding decision method comprises the following steps. S110: Obtain historical heat information and video historical data of a plurality of first users, and perform first screening processing on the plurality of first users according to the historical heat information and the video historical data to obtain a plurality of second users.
[0017] Exemplarily, the historical heat information and the video historical data of the plurality of first users are obtained. The first user can be a host of a live broadcast platform, and the historical heat information can be used to reflect the heat of the first user in a recent time period. The live room heat information can be determined by a combination of one or more of the total number of online people, the peak number of people, the average number of online people, and the integral score (such as active points, gift points, etc.) in the live room. The video historical data can be used to reflect the video features of the first user in a recent time period. The video features can be a combination of one or more of the video frame rate, the video code rate, and the video resolution.
[0018] In an embodiment, the plurality of first users are subjected to the first screening processing according to the historical heat information and the video historical data of the first users to obtain a plurality of second users, so as to realize the preliminary screening of the users, reduce the number of users participating in the subsequent transcoding value calculation, and improve the video transcoding decision efficiency. For example, the first users with low live video frame rate have high video delay and high stalling rate after video transcoding, and the negative impact of video transcoding is greater than the positive impact. Therefore, these first users are screened out and do not need to enter the subsequent transcoding value calculation. For the first users with low live video resolution and low video code rate, the smoothness improvement after video transcoding is low, and the cost performance of video transcoding is low. Therefore, these first users are also screened out and do not need to enter the subsequent transcoding value calculation, so as to reduce the waste of video transcoding decision resources of the users with low transcoding value.
[0019] Optionally, the first users with historical heat information lower than a first preset heat threshold and video historical data outside a preset video data range are screened out from the plurality of first users, and the remaining plurality of first users are determined as second users. The first users with video historical data outside the preset video data range can be understood as users with poor video quality or high video quality in a recent time period. The video transcoding of the users with poor video quality will result in high video delay and high stalling rate, and the negative impact of video transcoding is greater than the positive impact. Therefore, these users need to be screened out. The video transcoding of the users with high video quality will result in low smoothness improvement, and the cost performance of video transcoding is low. Therefore, these users also need to be screened out. Alternatively, the historical heat information and the video historical data are input into a trained preliminary screening model (such as a support vector machine (SVM) model), and the historical heat information and the video historical data are analyzed and processed by the preliminary screening model to determine whether to retain or screen out the corresponding first user.
[0020] S120: Obtain video feature information of the plurality of second users, and determine transcoding values of the plurality of second users according to the historical hotness information and the video feature information.
[0021] Exemplarily, the video feature information of the plurality of second users is obtained, and the video feature information can be obtained by feature extraction according to live video frames collected by the second users during live streaming. After obtaining the video feature information of the second users, the transcoding values of the plurality of second users are determined according to the historical hotness information and the video feature information. The higher the transcoding value is, the higher the income obtained by video transcoding of the second user is.
[0022] Optionally, the historical hotness information and the video feature information can be normalized respectively, and weighted sum processing is performed on the normalized historical hotness information and the video feature information, and the weighted sum processing result is taken as the transcoding value of the second user. The historical hotness information and the video feature information can also be input into a trained neural network model, and the historical hotness information and the video feature information are analyzed and processed by the neural network model to obtain the transcoding value of the second user.
[0023] S130: Perform sorting processing on the plurality of second users according to the transcoding values to obtain transcoding ranking information of the plurality of second users.
[0024] Exemplarily, after the transcoding values of the second users are determined, the plurality of second users are sorted in descending order of the transcoding values to obtain the transcoding ranking information of the plurality of second users. Optionally, the transcoding ranking information of each second user can be recorded by a transcoding ranking table.
[0025] The higher the live streaming hotness of the second user is and the longer the audience viewing time is, the greater the income brought by the same transcoding resource is. However, the transcoding income is also affected by the video features, that is, the better the video quality of the anchor is, the lower the value of the video transcoding enhancement link is, and the lower the video code rate of the anchor is, the lower the transcoding compression rate income is. The application comprehensively considers the historical hotness information and the video feature information of the second user, sorts the second users according to the transcoding values of the second users, and the second users in the front can have priority transcoding permission. The transcoding server resource can be preferentially allocated to the second users in the front, and the utilization efficiency of the video transcoding resource is improved.
[0026] S140: Perform second screening processing on the plurality of second users according to the transcoding ranking information to obtain a plurality of target users, and the target user is a user determined to be processed by video transcoding.
[0027] Exemplarily, the second screening processing is performed on the plurality of second users according to the transcoding ranking information determined above, to obtain a plurality of target users that need to be subjected to video transcoding processing. For example, a first quantity of second users can be determined as target users according to the transcoding ranking information. The first quantity can be a preset quantity, or can be determined according to the resource quantity of the transcoding server.
[0028] According to the above, the plurality of second users are obtained by performing the first screening processing on the plurality of first users according to the historical heat information and the video historical data of the plurality of first users, the video feature information of the plurality of second users is obtained, the transcoding values of the plurality of second users are determined according to the historical heat information and the video feature information, the plurality of second users are sorted to obtain the transcoding ranking information of the plurality of second users, the second screening processing is performed on the plurality of second users according to the transcoding ranking information, to obtain a plurality of target users that need to be subjected to video transcoding processing, the first screening processing can effectively reduce the waste of video transcoding decision resources by users with low transcoding values, improve the video transcoding decision efficiency, and consider the transcoding values of users from multiple dimensions of historical heat information and video historical data, so that the video transcoding decision is no longer dependent on a single heat, effectively reducing the situation that users with low video transcoding values are decided to be users that need to be subjected to video transcoding, more comprehensively analyzing the video transcoding values of different anchors, reducing the waste of video transcoding resources, and improving the utilization rate of video transcoding resources.
[0029] On the basis of the above-described embodiments, Figure 2 A flowchart of another video transcoding decision method provided by the embodiments of the present application is given, which is a specific embodiment of the above-described video transcoding decision method. Referring to Figure 2 The video transcoding decision method comprises: S210: Obtain historical heat information and video historical data of a plurality of first users, and perform first screening processing on the plurality of first users according to the historical heat information and the video historical data, to obtain a plurality of second users.
[0030] In one possible embodiment, the video transcoding decision method provided by the present application obtains historical heat information and video historical data of a plurality of first users, and performs first screening processing on the plurality of first users according to the historical heat information and the video historical data, to obtain a plurality of second users, which can comprise: S211: Obtain historical heat information of a plurality of first users, and a combination of one or more of video historical frame rate, video historical code rate and video historical resolution.
[0031] S212: Perform first screening processing on the plurality of first users according to the historical heat information, the combination of one or more of the video historical frame rate, the video historical code rate and the video historical resolution, to obtain a plurality of second users.
[0032] In an embodiment, the video history data provided by the present application can include a combination of one or more of the video history frame rate, the video history code rate, and the video history resolution. Accordingly, a combination of one or more of the historical heat information, the video history frame rate, the video history code rate, and the video history resolution of the plurality of first users is obtained, and the plurality of first users is first screened according to the combination of one or more of the historical heat information, the video history frame rate, the video history code rate, and the video history resolution to obtain a plurality of second users. The video history frame rate, the video history code rate, and the video history resolution can be monitored and collected for the live video of each first user, and the video history frame rate, the video history code rate, and the video history resolution can be the average video frame rate, the average video code rate, and the average video resolution in the recent preset time period, or the video history frame rate, the video history code rate, and the video history resolution in the last live broadcast.
[0033] In an embodiment, the first user in the plurality of first users can be screened out if the historical heat information is lower than the second preset heat threshold, and the video history frame rate is outside the preset frame rate range, the video history code rate is outside the preset code rate range, and the video history resolution is outside the preset resolution range. Alternatively, the combination of one or more of the historical heat information, the video history frame rate, the video history code rate, and the video history resolution can be input into the trained preliminary screening model, and the combination of one or more of the historical heat information, the video history frame rate, the video history code rate, and the video history resolution is analyzed and processed by the preliminary screening model to determine whether to retain or screen out the corresponding first user. The present application considers the combination of one or more of the historical heat information, the video history frame rate, the video history code rate, and the video history resolution to first screen the plurality of first users, reduces the calculation of the subsequent transcoding value of the first user with low video transcoding revenue, and improves the video transcoding decision efficiency.
[0034] In a possible embodiment, the historical heat information and the video history data of the plurality of first users obtained in the video transcoding decision method provided by the present application can be: obtaining a plurality of heat information records and video history records of the plurality of first users in a preset time length; and determining the historical heat information and the video history data of the plurality of first users according to the plurality of heat information records and the plurality of video history records.
[0035] Illustratively, the plurality of heat information records and the video history records of the plurality of first users in the preset time length are obtained, and the historical heat information and the video history data of the first user are calculated according to the plurality of heat information records and the plurality of video history records of each first user. The preset time length can be 1 day, 3 days, 7 days, 30 days, etc.
[0036] For example, the plurality of heat information records can be recorded as historical heat information with an average value of live broadcast time length, and the plurality of video history records can be recorded as video history data with an average value of live broadcast time length. The live broadcast time length can be a total time length of live broadcast of the first user within a preset time length. The plurality of heat information records can be weighted and summed to obtain the historical heat information, and the plurality of video history records can be weighted and summed to obtain the video history data according to a time distance from a current time. The smaller the time distance from the current time, the greater the corresponding weight.
[0037] The application can effectively reduce the situation that the short-term change of the heat information record or the video history record leads to an error estimation of the transcoding value of the first user, and improve the accuracy and stability of the video transcoding decision, by determining the historical heat information and the video history data of the plurality of first users according to the plurality of heat information records and the video history records of the plurality of first users within a preset time length.
[0038] S220: obtaining video feature information of the plurality of second users, and determining a transcoding value of the plurality of second users according to the historical heat information and the video feature information.
[0039] Optionally, the video feature information provided by the application can be obtained by performing feature extraction on historical video frames of the second user stored in a preset database. Alternatively, the video feature information can be obtained by performing feature extraction on historical video frames of the second user in advance and storing the video feature information in the preset database, and the video feature information of the second user can be directly obtained from the preset database.
[0040] In one possible embodiment, the video feature information provided by the application can be obtained by performing feature extraction on a plurality of historical video frames. Based on this, in the video transcoding decision method provided by the application, the video feature information of the plurality of second users can be obtained by obtaining a plurality of historical video frames of the plurality of second users and performing feature extraction on the historical video frames to obtain the video feature information.
[0041] For example, during the live broadcast of each user, live broadcast video frames of the user are collected and stored as historical video frames in a preset database, for example, live broadcast video frames of the user are collected at a set time interval and stored in the preset database. When making a video transcoding decision, a plurality of historical video frames of the plurality of second users can be obtained from the preset database, and feature extraction can be performed on the plurality of historical video frames of each second user based on a preset video feature extraction algorithm or feature extraction to obtain video feature information of each second user.
[0042] Correspondingly, when determining the transcoding values of the plurality of second users according to the historical hotness information and the video feature information, the transcoding values of the plurality of second users can be determined according to the historical hotness information and the plurality of video feature information. The plurality of video feature information can be fused by splicing, and then the transcoding values of the plurality of second users are determined according to the historical hotness information and the fused video feature information. The video feature information is obtained by feature extraction on the plurality of historical video frames of the plurality of second users, accurately reflects the video features of each second user in the live streaming process, ensures accurate determination of the transcoding values of each second user, and improves the accuracy of video transcoding decision.
[0043] In one possible embodiment, the method for video transcoding decision provided by the present application can determine the transcoding values of the plurality of second users according to the historical hotness information and the video feature information, which can be: inputting the historical hotness information and the video feature information into a trained value determination model, analyzing and processing the historical hotness information and the video feature information by the value determination model, and obtaining the transcoding values of the plurality of second users.
[0044] In one embodiment, the coupling relationship between the historical hotness information and the video feature information of each second user can be processed by the trained value determination model, and the transcoding value of the second user is output by the value determination model. The value determination model can be built based on a multilayer perceptron (MLP).
[0045] Exemplarily, the historical hotness information and the video feature information obtained above are input into the value determination model, the historical hotness information and the video feature information are analyzed and processed by the value determination model, and the transcoding values of the plurality of second users are obtained. The value determination model can analyze and process the historical hotness information and the video feature information, which can be: extracting the embedding vector information of the historical hotness information, submitting the embedding vector information of the historical hotness information and the video feature information (for example, splicing the embedding vector information and the video feature information) to the classification layer structure (MLP Head) in the value determination model, and analyzing and processing the embedding vector information and the video feature information by the classification layer structure and outputting the corresponding transcoding value. The historical hotness information and the video feature information are analyzed and processed by the value determination model, the transcoding values of each second user are accurately determined, and the accuracy of video transcoding decision is effectively improved.
[0046] S230: The plurality of second users are sorted according to the transcoding values to obtain transcoding ranking information of the plurality of second users.
[0047] S240: The plurality of second users are secondarily selected according to the transcoding ranking information to obtain a plurality of target users, and the target users are users determined to be processed by video transcoding.
[0048] In a possible embodiment, the video transcoding decision method provided in the application further obtains transcoding resource information before performing second screening on the plurality of second users according to the transcoding ranking information to obtain the plurality of target users.
[0049] Correspondingly, the video transcoding decision method provided in the application performs second screening on the plurality of second users according to the transcoding ranking information to obtain the plurality of target users, which can be: performing second screening on the plurality of second users according to the transcoding resource information and the transcoding ranking information to obtain the plurality of target users.
[0050] Exemplarily, when it is necessary to determine the target users for video transcoding processing, the transcoding resource information corresponding to each transcoding server can be determined, and the number of transcoding users that can perform video transcoding processing can be determined according to the transcoding resource information. Alternatively, the transcoding resource information corresponding to each transcoding server can be added to obtain the total transcoding resource amount, and the ratio of the total transcoding resource amount to the preset single-user transcoding resource consumption amount is determined as the number of transcoding users. The preset single-user transcoding resource consumption amount can be understood as the resource amount consumed by the transcoding server when performing video transcoding on a single user.
[0051] In an embodiment, after the number of transcoding users is determined, the second users with the number of transcoding users can be determined as target users from the plurality of second users in descending order of the transcoding ranking information. The application determines the target users by performing second screening on the plurality of second users according to the transcoding resource information and the transcoding ranking information of each transcoding server, which can perform video transcoding processing on the target users with higher transcoding value while ensuring that the transcoding resources are fully utilized, thereby improving the utilization efficiency of transcoding resources.
[0052] S250: distributing the plurality of target users to the plurality of transcoding servers, and using the transcoding servers to perform transcoding processing on the target users distributed thereto.
[0053] Exemplarily, after the plurality of target users are obtained by performing second screening on the plurality of second users according to the transcoding ranking information, the plurality of target users can be distributed to the plurality of transcoding servers (randomly distributing the target users to the transcoding servers, or distributing the target users to the transcoding servers based on a preset load balancing strategy), and the transcoding servers can obtain the live video stream provided by the target users distributed thereto, perform transcoding processing on the live video stream to obtain transcoded video streams, and send the transcoded video streams to the audience or the live server. The application distributes the plurality of target users to the plurality of transcoding servers for transcoding processing, thereby achieving balanced distribution of video transcoding resources and improving the utilization efficiency of transcoding resources.
[0054] In an embodiment, when the transcoding server transcodes the live video stream, the live video stream can be pre-processed (for example, one or more of a combination of quality enhancement, noise reduction, frame interpolation, contrast enhancement, super-resolution processing), and then the pre-processed live video stream is transcoded, which can effectively improve the smoothness and picture quality of the transcoded video stream.
[0055] The above, by the first screening processing of the plurality of first users according to the historical hotness information and the video historical data of the plurality of first users, obtaining the video feature information of the plurality of second users, determining the transcoding value of the plurality of second users according to the historical hotness information and the video feature information, and sorting processing the plurality of second users according to the transcoding value to obtain the transcoding ranking information of the plurality of second users, the second screening processing of the plurality of second users can be performed according to the transcoding ranking information, and the target user for video transcoding processing is obtained, the first screening processing can effectively reduce the waste of video transcoding decision resources of users with low transcoding value, improve the video transcoding decision efficiency, and consider the transcoding value of the user from the historical hotness information and the video historical data multidimensionally, the video transcoding decision is no longer dependent on a single hotness, effectively reducing the situation that users with low video transcoding value are decided as users needing video transcoding, more comprehensively analyzing the video transcoding value of different anchors, reducing the waste of video transcoding resources, and improving the utilization rate of video transcoding resources. At the same time, the plurality of target users are distributed to the plurality of transcoding servers for transcoding processing, realizing the balanced allocation of video transcoding resources, and further improving the utilization efficiency of transcoding resources.
[0056] Figure 3 A structure diagram of a video transcoding decision device provided by an embodiment of the present application is given. Referring to Figure 3 The video transcoding decision device includes a first screening module 31, a value analysis module 32, a ranking processing module 33, and a second screening module 34.
[0057] The first screening module 31 is configured to obtain historical hotness information and video historical data of a plurality of first users, and perform first screening processing on the plurality of first users according to the historical hotness information and the video historical data to obtain a plurality of second users; the value analysis module 32 is configured to obtain video feature information of the plurality of second users, and determine transcoding values of the plurality of second users according to the historical hotness information and the video feature information; the ranking processing module 33 is configured to sort the plurality of second users according to the transcoding values to obtain transcoding ranking information of the plurality of second users; and the second screening module 34 is configured to perform second screening processing on the plurality of second users according to the transcoding ranking information to obtain a plurality of target users, and the target user is a user determined to perform video transcoding processing.
[0058] According to the historical heat information and the video historical data of the plurality of first users, the plurality of second users are obtained through the first screening processing of the plurality of first users, the video feature information of the plurality of second users is obtained, the transcoding value of the plurality of second users is determined according to the historical heat information and the video feature information, and the plurality of second users are sorted according to the transcoding value to obtain the transcoding ranking information of the plurality of second users. The plurality of second users can be second screened according to the transcoding ranking information, and the plurality of target users for video transcoding processing are obtained. The first screening processing can effectively reduce the waste of video transcoding decision resources of users with low transcoding value, improve the video transcoding decision efficiency, and consider the transcoding value of the user from multiple dimensions of the historical heat information and the video historical data. The video transcoding decision is no longer dependent on a single heat, effectively reducing the situation that users with low video transcoding value are decided as users needing video transcoding, more comprehensively analyzing the video transcoding value of different anchors, reducing the waste of video transcoding resources, and improving the utilization rate of video transcoding resources.
[0059] In one possible embodiment, the first screening module 31 obtains historical heat information and video historical data of a plurality of first users, and performs first screening processing on the plurality of first users according to the historical heat information and the video historical data to obtain a plurality of second users, and is configured to: obtain historical heat information of a plurality of first users, and a combination of one or more of video historical frame rate, video historical code rate and video historical resolution; perform first screening processing on the plurality of first users according to the historical heat information, the combination of one or more of the video historical frame rate, the video historical code rate and the video historical resolution to obtain a plurality of second users.
[0060] In one possible embodiment, the first screening module 31 obtains historical heat information and video historical data of a plurality of first users, and is configured to: obtain a plurality of heat information records and video historical records of the plurality of first users within a preset time length; determine the historical heat information and the video historical data of the plurality of first users according to the plurality of heat information records and the plurality of video historical records.
[0061] In one possible embodiment, the value analysis module 32 obtains video feature information of a plurality of second users, and is configured to: obtain a plurality of historical video frames of the plurality of second users, and perform feature extraction on the historical video frames to obtain the video feature information.
[0062] In one possible embodiment, the value analysis module 32 determines the transcoding value of the plurality of second users according to the historical heat information and the video feature information, and is configured to: The historical hotness information and the video feature information are input into the trained value determination model, the historical hotness information and the video feature information are analyzed and processed by the value determination model, and the transcoding values of the second users are obtained.
[0063] In one possible embodiment, the video transcoding decision apparatus further comprises a resource determination module configured to: acquire transcoding resource information; Correspondingly, the second screening module 34 performs second screening processing on the second users according to the transcoding ranking information to obtain target users, and is configured to: perform second screening processing on the second users according to the transcoding resource information and the transcoding ranking information to obtain target users.
[0064] In one possible embodiment, the video transcoding decision apparatus further comprises a user allocation module configured to: allocate the target users to the transcoding servers, and perform transcoding processing on the allocated target users by using the transcoding servers.
[0065] It is worth noting that, in the embodiments of the video transcoding decision apparatus described above, each unit and module included is only divided according to functional logic, but is not limited to the division described above, as long as the corresponding functions can be implemented; in addition, the specific names of each functional unit are only for the convenience of mutual differentiation, and do not limit the protection scope of the embodiments of the present application.
[0066] The embodiments of the present application also provide a video transcoding decision device, which can integrate the video transcoding decision apparatus provided by the embodiments of the present application. Figure 4 FIG. 1 is a structural schematic diagram of a video transcoding decision device provided by an embodiment of the present application. Referring to FIG. 1, Figure 4 The video transcoding decision device comprises an input apparatus 43, an output apparatus 44, a memory 42, and one or more processors 41; the memory 42 is used to store one or more programs; when the one or more programs are executed by the one or more processors 41, the one or more processors 41 implement the video transcoding decision method provided by the above-described embodiments. The video transcoding decision apparatus, device, and computer provided above can be used to execute the video transcoding decision method provided by any of the above-described embodiments, and have the corresponding functions and beneficial effects.
[0067] The embodiment of the present application further provides a nonvolatile storage medium storing computer executable instructions, which are used for executing the video transcoding decision method provided by the above embodiment when executed by a computer processor. Of course, the nonvolatile storage medium storing computer executable instructions provided by the embodiment of the present application is not limited to the video transcoding decision method provided above, and can also execute the related operations in the video transcoding decision method provided by any embodiment of the present application. The video transcoding decision device, equipment and storage medium provided in the above embodiment can execute the video transcoding decision method provided by any embodiment of the present application, and the technical details not described in detail in the above embodiment can be referred to the video transcoding decision method provided by any embodiment of the present application.
[0068] On the basis of the above embodiment, the embodiment of the present application further provides a computer program product, the technical solution of the present application or the whole or part of the contribution to the prior art can be embodied in the form of a software product, the computer program product is stored in a storage medium, and includes a plurality of instructions for making a computer device, a mobile terminal or a processor therein execute all or part of the steps of the video transcoding decision method provided by each embodiment of the present application.
Claims
1. A method of video transcoding decision, characterized in that, The method comprises the following steps: obtaining historical heat information and video history data of a plurality of first users, performing first screening processing on the plurality of first users according to the historical heat information and the video history data, and obtaining a plurality of second users; obtaining video feature information of the plurality of second users, determining the transcoding value of the plurality of second users according to the historical heat information and the video feature information; performing sorting processing on the plurality of second users according to the transcoding value to obtain transcoding ranking information of the plurality of second users; performing second screening processing on the plurality of second users according to the transcoding ranking information to obtain a plurality of target users, and the target users are users determined to be processed by video transcoding.
2. The method of claim 1, wherein, The method comprises the following steps: obtaining historical heat information and video history data of a plurality of first users, performing first screening processing on the plurality of first users according to the historical heat information and the video history data, and obtaining a plurality of second users; obtaining historical heat information and a combination of one or more of video history frame rate, video history code rate and video history resolution of a plurality of first users; 3. The method of claim 1, wherein, performing first screening processing on the plurality of first users according to the historical heat information, the combination of one or more of the video history frame rate, the video history code rate and the video history resolution to obtain a plurality of second users. The method comprises the following steps: obtaining a plurality of heat information records and video history records of a plurality of first users within a preset time length; 4. The method of claim 1, wherein, determining historical heat information and video history data of a plurality of first users according to a plurality of heat information records and a plurality of video history records. The method comprises the following steps:
5. The method of claim 1, wherein, obtaining a plurality of historical video frames of a plurality of second users, and performing feature extraction on the historical video frames to obtain video feature information. The method comprises the following steps:
6. The method of claim 1, wherein, inputting the historical heat information and the video feature information into a trained value determination model, and performing analysis processing on the historical heat information and the video feature information by the value determination model to obtain the transcoding value of the plurality of second users. Before the second screening processing on the plurality of second users according to the transcoding ranking information to obtain a plurality of target users, the method further comprises the following steps: obtaining transcoding resource information; Accordingly, the second screening processing on the plurality of second users according to the transcoding ranking information to obtain a plurality of target users comprises the following steps:
7. The method of claim 1-6, wherein, performing second screening processing on the plurality of second users according to the transcoding resource information and the transcoding ranking information to obtain a plurality of target users. After the second screening processing on the plurality of second users according to the transcoding ranking information to obtain a plurality of target users, the method further comprises the following steps:
8. A video transcoding decision apparatus characterized by comprising: allocating the plurality of target users to a plurality of transcoding servers, and performing transcoding processing on the allocated target users by using the transcoding servers. The method comprises a first screening module, a value analysis module, a ranking processing module and a second screening module. The first screening module is configured to acquire historical hotness information and video history data of a plurality of first users, perform first screening processing on the plurality of first users according to the historical hotness information and the video history data, and obtain a plurality of second users; The value analysis module is configured to acquire video feature information of the plurality of second users, determine a transcoding value of the plurality of second users according to the historical hotness information and the video feature information; The ranking processing module is configured to perform sorting processing on the plurality of second users according to the transcoding value, and obtain transcoding ranking information of the plurality of second users; The second screening module is configured to perform second screening processing on the plurality of second users according to the transcoding ranking information, and obtain a plurality of target users, the target users being users determined to perform video transcoding processing.
9. A video transcoding decision device, characterized by Comprise: A memory and one or more processors; The memory is used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the video transcoding decision method in any one of claims 1-7.
10. A non-volatile storage medium for storing computer-executable instructions, characterized in that, The computer executable instructions, when executed by a computer processor, are used to perform the video transcoding decision method in any one of claims 1-7.
11. A computer program product comprising a computer program, characterized in that, The computer program, when executed by a processor, implements the video transcoding decision method in any one of claims 1-7.