Method and apparatus for detecting highlight segments of live streaming resources, and computing device

Abstract

The application discloses a highlight segment detection method and device of a live broadcast resource and computing equipment. The method comprises the following steps: acquiring a play effect index value of a target live broadcast resource at different sampling times; generating a plurality of sliding time windows with different window lengths, each sliding time window dividing a resource interval of the target live broadcast resource into a plurality of time slices; calculating an index average value of each time slice according to the play effect index value in the time slice; determining a correction coefficient of each sliding time window; obtaining an effect evaluation value of the time slice according to the index average value of the time slice and the correction coefficient of the sliding time window to which the time slice belongs; and screening a target time slice according to the effect evaluation value, and determining a highlight segment in the target live broadcast resource according to the target time slice. The scheme does not need to perform a complex content recognition process, so that the highlight segment recognition efficiency is high, and the scheme has higher detection precision compared with a fixed time length detection method.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet, in particular to a highlight segment detection method and device for live streaming resources, computing equipment, computer storage medium and computer program product. BACKGROUND

[0002] With the continuous development of science and technology and society, the emergence of live streaming business greatly enriches people's work and life. In order to facilitate live streaming playback, secondary editing, data analysis, etc., it is usually necessary to detect the highlight period of live streaming resources.

[0003] The currently commonly used highlight segment detection method for live streaming resources includes two kinds:

[0004] The first is a fixed length detection method, that is, the highest point of indicators such as the number of viewers in the live streaming process is found, and the highest point is taken as the benchmark to expand a fixed length forward and backward to obtain the highlight segment. For example, in the prior art, it is determined that the number of viewers in the live streaming room is the most at T1, and the segment corresponding to the period of [T-2, T+3] is taken as the highlight period.

[0005] The second is a content detection method, that is, the image frames in the live streaming video are scored by image detection algorithms or AI models, and then the highlight period is determined according to the content score.

[0006] However, the inventors found in the implementation process that the prior art has the following defects: the fixed length detection method has insufficient detection accuracy of the highlight segment, and cannot adapt to the detection of highlight periods in different scenarios; and the content detection method has a large amount of calculation, consumes a lot of system resources, and has low detection efficiency. SUMMARY

[0007] In view of the above problems, the present application is proposed to provide a highlight segment detection method, device, computing equipment, computer storage medium and computer program product for live streaming resources, which overcomes the above problems or at least partially solves the above problems.

[0008] According to a first aspect of the present application, a highlight segment detection method for live streaming resources is provided, comprising:

[0009] Obtaining the play effect indicator value of the target live streaming resource at different sampling times;

[0010] Generating a plurality of sliding time windows with different window lengths, each sliding time window dividing the resource interval of the target live streaming resource into a plurality of time slices;

[0011] According to the play effect indicator value in each time slice, calculating the indicator average value of each time slice;

[0012] Determining the correction coefficient of each sliding time window;

[0013] For each time slice, an effect evaluation value of the time slice is obtained according to the index average value of the time slice and the correction coefficient of the sliding time window to which the time slice belongs;

[0014] A target time slice is screened according to the effect evaluation values of the time slices, and a highlight segment in the target live broadcast resource is determined according to the target time slice.

[0015] In an optional implementation, the method further comprises:

[0016] A fixed correction coefficient is assigned to the sliding time window with the minimum window length;

[0017] For any non-minimal sliding time window, a correction coefficient of the sliding time window relative to the previous sliding time window is calculated according to the window length of the sliding time window and an adjustment parameter, and a correction coefficient of the sliding time window is calculated according to the correction coefficient of the previous sliding time window and the correction coefficient of the sliding time window relative to the previous sliding time window.

[0018] In an optional implementation, the method further comprises:

[0019] The sum of the adjustment parameter, the window length of the sliding time window and -1 is calculated;

[0020] The ratio of the window length of the sliding time window to the sum is taken as the correction coefficient of the sliding time window relative to the previous sliding time window.

[0021] In an optional implementation, the adjustment parameter is determined by:

[0022] The editing lengths of the live highlight segment editing resources uploaded in history are counted;

[0023] A reference window length is determined according to the number of editing resources corresponding to different editing lengths;

[0024] A first editing resource number corresponding to the reference window length and a second editing resource number corresponding to a reference window length are determined; wherein the reference window length = the reference window length + 1;

[0025] The adjustment parameter is generated according to the first editing resource number and the second editing resource number.

[0026] In an optional implementation, before the plurality of sliding time windows with different window lengths are generated, the method further comprises:

[0027] statistically upload a clip length of a live highlight clip resource;

[0028] generate a window length value range according to the clip length;

[0029] The generating of the plurality of sliding time windows with different window lengths comprises generating the plurality of sliding time windows with different window lengths according to the window length value range.

[0030] In an optional implementation, the method further comprises: obtaining a system performance parameter;

[0031] The generating of the window length value range according to the clip length comprises generating the window length value range according to the clip length and the system performance parameter.

[0032] In an optional implementation, the filtering of the target time slice according to the effect evaluation value of each time slice, and the determining of the highlight clip in the target live resource according to the target time slice comprises:

[0033] sorting the time slices in a descending order of the effect evaluation value;

[0034] selecting the time slices with top N effect evaluation values as the target time slices;

[0035] aggregating the target time slices to obtain an aggregated time period;

[0036] taking a clip corresponding to the aggregated time period as the highlight clip.

[0037] According to a second aspect of the present application, a highlight clip detection device for live resources is provided, comprising:

[0038] an index obtaining module configured to obtain a play effect index value of a target live resource at different sampling times;

[0039] a window generating module configured to generate a plurality of sliding time windows with different window lengths, each sliding time window dividing a resource interval of the target live resource into a plurality of time slices;

[0040] an average value calculating module configured to calculate an index average value of each time slice according to the play effect index value in the time slice;

[0041] a correction coefficient determining module configured to determine a correction coefficient of each sliding time window;

[0042] an evaluation value calculating module configured to obtain an effect evaluation value of each time slice according to the index average value of the time slice and the correction coefficient of the sliding time window to which the time slice belongs.

[0043] The segment identifying module is configured to screen a target time slice according to the effect evaluation value of each time slice, and determine a highlight segment in the target live broadcast resource according to the target time slice.

[0044] In an optional implementation, the correction coefficient determining module is configured to assign a fixed correction coefficient to the sliding time window with the minimum window length.

[0045] For any non-minimal sliding time window, the correction coefficient determining module is configured to calculate a correction coefficient of the sliding time window relative to a previous sliding time window according to the window length of the sliding time window and the adjustment parameter, and calculate the correction coefficient of the sliding time window according to the correction coefficient of the previous sliding time window and the correction coefficient of the sliding time window relative to the previous sliding time window.

[0046] In an optional implementation, the correction coefficient determining module is configured to calculate a sum of the adjustment parameter, the window length of the sliding time window, and -1.

[0047] The ratio of the window length of the sliding time window to the sum is taken as the correction coefficient of the sliding time window relative to the previous sliding time window.

[0048] In an optional implementation, the device further includes an adjustment parameter generating module configured to count the clip lengths of the live highlight segment clip resources uploaded in history.

[0049] The reference window length is determined according to the number of clip resources corresponding to different clip lengths.

[0050] The first clip resource number corresponding to the reference window length and the second clip resource number corresponding to the reference window length are determined, where the reference window length = the reference window length + 1.

[0051] The adjustment parameter is generated according to the first clip resource number and the second clip resource number.

[0052] In an optional implementation, the device further includes a window length range determining module configured to count the clip lengths of the live highlight segment clip resources uploaded in history.

[0053] The window length value range is generated according to the clip lengths.

[0054] The multiple sliding time windows with different window lengths are generated according to the window length value range.

[0055] In an optional implementation, the window length range determining module is configured to obtain a system performance parameter, and generate the window length value range according to the clip lengths and the system performance parameter.

[0056] In an optional implementation, the segment identification module is configured to sort the time slices in descending order of the effect evaluation values;

[0057] Selecting time slices with top N effect evaluation values as target time slices;

[0058] Aggregating the target time slices to obtain an aggregated time period;

[0059] Taking a segment corresponding to the aggregated time period as the highlight segment.

[0060] According to a third aspect of the present application, a computing device is provided, comprising a processor, a memory, a communication interface and a communication bus, the processor, the memory and the communication interface are in communication with each other through the communication bus;

[0061] The memory is configured to store at least one executable instruction, and the executable instruction causes the processor to perform operations corresponding to the highlight segment detection method of the live resource.

[0062] According to a fourth aspect of the present application, a computer storage medium is provided, and the storage medium stores at least one executable instruction, and the executable instruction causes the processor to perform operations corresponding to the highlight segment detection method of the live resource.

[0063] According to a fifth aspect of the present application, a computer program product is provided, comprising at least one executable instruction, and the executable instruction causes the processor to perform operations corresponding to the highlight segment detection method of the live resource.

[0064] Embodiments of the present application divide the resource interval of the target live resource into time slices with different lengths by using multiple sliding time windows with different window lengths, and assign corresponding correction coefficients to the sliding time windows with different window lengths. The effect evaluation value of a time slice is obtained by correcting the average value of the index of the time slice using the correction coefficient of the sliding time window to which the time slice belongs. Then, the highlight segment is identified according to the effect evaluation value. By using the present solution, a complex content identification process is not required, so that the highlight segment identification efficiency is high. Moreover, compared with the fixed time length detection method, the present solution can have higher detection accuracy and wider application scenarios.

[0065] Embodiments of the present application generate the correction coefficient of any non-minimal sliding time window compared with the previous sliding time window, and then generate the correction coefficient of each sliding time window according to the correction coefficient, so as to improve the determination accuracy of the correction coefficient and further improve the detection accuracy of the highlight segment.

[0066] The embodiment of the present application determines the correction factor of the sliding time window compared with the previous sliding time window according to the adjustment parameter and the window length, so that the index mean value in the adjacent two sliding time windows with different window lengths can be accurately balanced through the correction factor.

[0067] The embodiment of the present application determines the adjustment parameter according to the editing length of the historical uploaded live highlight clip resource, so as to improve the determination accuracy of the adjustment parameter.

[0068] The embodiment of the present application generates the window length value range according to the editing length of the historical uploaded live highlight clip resource in advance, so that the generated multiple sliding time windows can be adapted to the actual situation, and the rationality of the generated sliding time window is improved.

[0069] The embodiment of the present application determines the window length value range by combining the historical editing length and the system performance parameter, so as to further improve the determination accuracy of the window length value range, avoid the adverse effect of the too large window length range on the system performance, and avoid the waste of system resources caused by the too small window length range.

[0070] The embodiment of the present application aggregates the target time slice to obtain an aggregated period, and takes the segment corresponding to the aggregated period as a highlight segment, which on the one hand improves the continuity of the highlight segment and improves the user experience, and on the other hand reduces the number of highlight segments and reduces the maintenance cost.

[0071] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the specific embodiments of the present application can be implemented according to the content of the specification, and in order to make the above and other purposes, characteristics and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0072] By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of illustrating the preferred embodiments and are not considered as limiting the present application. Moreover, the same reference symbols are used to represent the same components throughout the drawings. In the drawings:

[0073] Figure 1 A running environment schematic diagram provided for implementing at least one embodiment of the present application is shown;

[0074] Figure 2 A flowchart of a live resource highlight segment detection method provided by the embodiment of the present application is shown;

[0075] Figure 3 A highlight segment display schematic diagram provided by the embodiment of the present application is shown;

[0076] Figure 4 A flow chart of a highlight segment detection method of a live broadcast resource is shown according to an embodiment of the present application;

[0077] Figure 5 A three-point mean diagram is shown according to an embodiment of the present application;

[0078] Figure 6 A flow chart of a parameter adjustment method is shown according to an embodiment of the present application;

[0079] Figure 7 A structural diagram of a highlight segment detection device of a live broadcast resource is shown according to an embodiment of the present application;

[0080] Figure 8 A structural diagram of a computing device is shown according to an embodiment of the present application. DETAILED DESCRIPTION

[0081] Exemplary embodiments of the present application will be described herein below with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. Rather, these embodiments are provided so that the present application can be more thoroughly understood and the scope of the present application can be accurately conveyed to those skilled in the art.

[0082] It should be noted that the evaluation index data and / or live highlight segment clipping resources (including but not limited to data for analysis, storage, display, etc.) involved in the embodiments of the present application are authorized or fully authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation portal for selection of authorization or rejection.

[0083] Figure 1 A running environment diagram is shown according to an embodiment of the present application. The present application can be applied to an application environment including but not limited to a client 2, a server 4 and a network 6.

[0084] Wherein:

[0085] Server 4 can consist of one or more computing devices. These computing devices may include virtualized computing instances. Virtualized computing instances may include virtual machines, such as emulations of computer systems, operating systems, servers, etc. The computing devices can load virtual machines based on virtual images and / or other data that define specific software used for emulation (e.g., operating systems, dedicated applications, servers). As the demand for different types of processing services changes, different virtual machines can be loaded and / or terminated on one or more computing devices. A hypervisor can be implemented to manage the use of different virtual machines on the same computing device.

[0086] Server 4 can be configured to communicate with client 2, etc., via network 6. Network 6 includes various network devices such as routers, switches, multiplexers, hubs, modems, bridges, repeaters, firewalls, proxy devices, and / or similar devices. Network 6 may include physical links, such as coaxial cable links, twisted pair cable links, fiber optic links, or combinations thereof, or wireless links, such as cellular links, satellite links, Wi-Fi links, etc.

[0087] Server 4 can provide services such as storage, reading, downloading, writing, querying, and deleting, such as providing static resource download services to clients through multiple domain names.

[0088] Client 2 can be running Windows or Android. TM Electronic devices running operating systems such as iOS, including smartphones, tablets, laptops, virtual reality devices, gaming devices, set-top boxes, in-vehicle terminals, and smart TVs. Based on these operating systems, various applications, such as browsers, can run.

[0089] Example 1

[0090] Figure 2 A flowchart illustrating a method for detecting highlights in live streaming resources according to Embodiment 1 of this application is shown. The method provided in this embodiment can be applied to live streaming platforms.

[0091] Specifically, such as Figure 2 As shown, the method includes the following steps:

[0092] Step S210: Obtain the playback effect index values ​​of the target live streaming resource at different sampling times.

[0093] The target live stream resource is the live stream resource for which highlight segment detection is performed, that is, the highlight segment in the target live stream resource used in this application. The target live stream resource can be an entire live stream or a part of a live stream, and the target live stream resource can be video, audio, and / or images and text, etc.

[0094] In order to improve user experience and facilitate data analysis, the live broadcast platform records the playback effect indicator values at different sampling times during the live broadcast of the target live broadcast resource. The playback effect indicator value is a numerical value of at least one playback effect indicator, which includes but is not limited to: the amount of barrage, the amount of watching, the amount of new entry, and / or the amount of income, etc. The amount of barrage is the number of barrages detected at the corresponding sampling time; the amount of watching is the number of viewers in the live broadcast room detected at the corresponding sampling time; the amount of new entry is the number of people who newly enter the live broadcast room detected at the corresponding sampling time; and the amount of income is the anchor income value detected at the corresponding sampling time.

[0095] In an optional implementation, in order to improve the efficiency of highlight segment detection, when the sampling frequency is high, the playback effect indicator values at different sampling times can be clustered to obtain the playback effect indicator values of each sampling unit, and a sampling unit contains multiple sampling times. For example, the sampling frequency is 10 seconds, that is, the playback effect indicator values of the target live broadcast resource are detected every 10 seconds, and the sampling unit is 1 minute. The playback effect indicator values of multiple sampling times in the same minute are accumulated to obtain the playback effect indicator values of the sampling unit.

[0096] Step S220, a plurality of sliding time windows with different window lengths are generated, and each sliding time window divides the resource interval of the target live broadcast resource into a plurality of time slices.

[0097] A plurality of sliding time windows are generated, and each sliding time window has a different window length. The sliding time window can slide according to a preset step and a preset direction. For example, a sliding time window A with a window length of 3 minutes (indicating that the interval between the starting point and the ending point of the window is 3 minutes), a sliding time window B with a window length of 4 minutes (indicating that the interval between the starting point and the ending point of the window is 4 minutes), a sliding time window C with a window length of 5 minutes (indicating that the interval between the starting point and the ending point of the window is 5 minutes), and a sliding time window J with a window length of 10 minutes (indicating that the interval between the starting point and the ending point of the window is 10 minutes) can be generated.

[0098] For each sliding time window, the sliding time window can divide the resource interval of the target live resource into multiple time slices. The resource interval of the target live resource is the time interval between the start time and the end time of the target live resource. For any sliding time window, the interval located in the sliding time window in the start state is a time slice, and then the sliding time window slides according to the preset direction and the preset step length, and the interval in the sliding time window is taken as another time slice, and so on until the sliding time window reaches the end position and cannot slide any more, so as to realize the division of the resource interval by the sliding time window. The length of the time slice obtained by the same sliding time window is the same, and is equal to the window length of the sliding time window. For example, the resource interval is [1, 5], and the sliding time window A divides the resource interval to obtain time slices [1, 3], [2, 4] and [3, 5]. The sliding time window B divides the resource interval to obtain time slices [1, 4] and [2, 5].

[0099] In step S230, the average value of the index of each time slice is calculated according to the playing effect index value in each time slice.

[0100] For any time slice, the playing effect index value at the sampling time located in the time slice is obtained, and then the average value of the playing effect index value in the time slice is calculated, and the average value is taken as the average value of the index of the time slice.

[0101] In an optional embodiment, the embodiments of the present application can generate the average value of the index in any index dimension, and obtain the highlight segment in the index dimension through subsequent steps S240-S260. For example, the average value of the view count index of the time slice is calculated, and then the highlight segment in the view count index dimension is obtained subsequently; the average value of the barrage count index can also be calculated, and then the highlight segment in the barrage count index dimension is obtained, and so on.

[0102] In another optional embodiment, the average value of the comprehensive index can be obtained by synthesizing the average values of the indexes in multiple index dimensions, for example, the average value of the comprehensive index is obtained by weighted summation of the weights of different indexes and the corresponding average values of the indexes, and then the highlight segment in the comprehensive index dimension is obtained through subsequent steps.

[0103] In the calculation process of the average value of the index, the cumulative value of the playing effect index value located in the time slice can be calculated, and the ratio of the cumulative value to the length of the time slice is taken as the average value of the index; or the cumulative value of the playing effect index value of each sampling unit located in the time slice is calculated, and then the ratio of the cumulative value to the length of the time slice is taken as the average value of the index.

[0104] Step S240, determining a correction coefficient for each sliding time window.

[0105] In actual implementation, the playing effect index value in the resource playing process often has certain continuity. If the index average values of the time slices divided by different sliding time windows are directly compared, the highlight segments will not be smooth and large fragmentation will occur. In view of this, the embodiments of the present application assign a corresponding correction coefficient to each sliding time window, that is, each sliding time window has a corresponding correction coefficient. The correction coefficient is positively correlated with the window length of the sliding time window, that is, the larger the window length of the sliding time window, the larger the corresponding correction coefficient.

[0106] Step S250, for each time slice, obtaining an effect evaluation value of the time slice according to the index average value of the time slice and the correction coefficient of the sliding time window to which the time slice belongs.

[0107] For any time slice, the index average value of the time slice is corrected by using the correction coefficient of the sliding time window to which the time slice belongs, and the effect evaluation value of the time slice is obtained. For example, the product of the index average value of the time slice and the correction coefficient of the sliding time window to which the time slice belongs can be taken as the effect evaluation value of the time slice.

[0108] Step S260, screening a target time slice according to the effect evaluation values of the time slices, and determining a highlight segment in the target live resource according to the target time slice.

[0109] N time slices with high effect evaluation values are selected as target time slices. For example, the time slices are sorted in descending order of effect evaluation values, and the time slices with the top N effect evaluation values are selected as target time slices. Then, the segments corresponding to the target time slices are taken as highlight segments in the target live resource.

[0110] In an optional implementation, the segments corresponding to each target time slice can be taken as highlight segments, thereby simplifying the implementation process of the method.

[0111] In another optional embodiment, there can be overlaps between the obtained N target time slices, in order to improve the user's viewing experience of the highlight segment, the embodiment aggregates the target time slices to obtain an aggregated period, and takes the segment corresponding to the aggregated period as the highlight segment. Specifically, multiple target time slices with time overlaps and / or interval lower than a preset threshold can be aggregated into one aggregated period. For example, the target time slices are [3, 5] and [4, 7], and the two target time slices are aggregated to obtain an aggregated period [3, 7], and then the segment corresponding to the aggregated period [3, 7] is identified as the highlight segment, thereby simplifying the number of highlight segments and improving the continuity of the highlight segments.

[0112] In an optional embodiment, if the step S230 calculates the average value of the indicators in different indicator dimensions, and the step S250 obtains the effect evaluation values in different indicator dimensions, then for any indicator dimension, the highlight segment in the indicator dimension is obtained according to the effect average value in the indicator dimension. In addition, the indicator value curve in different indicator dimensions can be displayed in a visual manner, and the highlight segment in the indicator dimension can also be marked in the curve. Finally, the obtained highlight segment can be displayed to the user. As shown in Figure 3 The highlight segments (highlight segments) identified in different indicator dimensions can be displayed, and the segment duration and the corresponding segment interval of the highlight segments can be displayed. The user can select a highlight segment to play, and the position of the currently played highlight segment in the time axis is marked in the play time axis, thereby improving the user experience.

[0113] Therefore, the highlight segment detection method for live streaming resources provided by the embodiments of the present application divides the resource interval of the target live streaming resource into time slices with different lengths by using multiple sliding time windows with different window lengths, assigns corresponding correction coefficients to the sliding time windows with different window lengths, corrects the average value of the indicators of the time slice by using the correction coefficient of the sliding time window to which the time slice belongs, and then identifies the highlight segment according to the effect evaluation value. By using the present scheme, the complex content recognition process is not required, so that the highlight segment recognition efficiency is high, and the detection accuracy and the application scenarios are higher than those of the fixed length detection method.

[0114] Embodiment Two

[0115] Figure 4 A flowchart of a highlight segment detection method for live streaming resources provided by Embodiment Two of the present application is shown. The method provided by the embodiments of the present application can be applied to a live streaming platform.

[0116] Specifically, as Figure 4 shown, the method comprises the following steps:

[0117] Step S410, count the clip length of the live highlight clip resources uploaded in history.

[0118] In actual implementation, the live broadcast platform provides a clip uploading entrance for users (including audience users and anchor users), and the users can upload live highlight clips of their own clips through the entrance. The live highlight clip uploaded by the user is referred to as a live highlight clip resource in the embodiments of the application.

[0119] The clip length of the live highlight clip resources uploaded in history is determined, and the clip length is the resource duration of the live highlight clip resource. The number of live highlight clip resources corresponding to different clip lengths is counted respectively.

[0120] Step S420, generate a window length value range according to the clip length.

[0121] Through the implementation of step S410, the clip length distribution of each live highlight clip resource uploaded in history (i.e., the number of live highlight clip resources corresponding to different clip lengths) can be obtained, and then the window length value range is determined according to the clip length distribution. For example, according to the clip length distribution, it is determined that the clip length is mostly located in the interval [3, 10], and then the window length value range is determined as [3, 10]. The window length value range can be generated and stored in advance as configuration data.

[0122] Further optionally, the system performance parameters can also be obtained, which can include hardware performance parameters or CPU utilization and other parameters. Then, the window length value range is generated according to the clip length and the system performance parameters, that is, the window length value range is generated according to the clip length distribution and the system performance parameters. Specifically, the reference window length value range can be determined according to the clip length distribution, and then the performance score is generated according to the performance parameters. The reference window length value range is corrected by using the performance score to obtain the final window length value range. For example, the reference window length value range is [3, 10], the performance score is 90%, and then the window length value range can be [3, 9], so that the performance score is used to reduce the window length value range on the basis of the reference window length value range. The lower the performance score is, the smaller the generated window length value range is. In this way, the window length range can take into account the system performance, avoid the system processing efficiency from being reduced due to the window length range being set too long, and also avoid the system resource from being wasted due to the window length range being set too short.

[0123] In an optional implementation, the window length value range can be updated every corresponding period or after receiving an update instruction, that is, when the update condition is met, steps S410 and S420 are re-executed to obtain an updated window length value range.

[0124] Step S430, generating an adjustment parameter according to the clip length.

[0125] The adjustment parameter is the basis for generating the correction coefficient subsequently. Specifically, if [i, i+m-1] is a time slice corresponding to a highlight segment with a window length of m, the average value of the indicators of this time slice is called avg_m, and the indicator value of the i+m unit is a i+m When n*avg_m≤a i+m ≤avg_m, it indicates that a i+M is close to the average value of the indicators of the highlight period avg_m, so a i+m should also be part of the highlight segment, that is, a i+m is taken as an extension point of the highlight period. This n is the adjustment parameter, n<1 but close to 1, for example, n can be 0.8.

[0126] Take Figure 5 for example: the sampling unit is minute, and the indicator values of the i, i+1, i+2, i+3 minutes are a i , a i+1 , a i+2 , a i+3 respectively. [i, i+2] is a time slice corresponding to a highlight segment with a window length of 3 minutes, and the average value of the indicators of the [i, i+2] time slice is avg_3 (also called three-point average in Table 1). The indicator value of the i+3 minute (referred to as the fourth minute in Table 1) is a i+3 . The specific case satisfies the following Table 1:

[0127] Table 1

[0128]

[0129] Among them, since it is assumed in the present application that [i, i+2] is a time slice corresponding to a highlight segment with a window length of 3 minutes, the case that a i+3 > avg_3 does not hold; when a i+3 = avg_3, it indicates that the three-point average (the average value of the indicators in the [i, i+2] interval) is the same as the four-point average (the average value of the indicators in the [i, i+3] interval), so the fourth minute should be included in the highlight segment; when n*avg_3≤a i+3 , it indicates that the fourth minute is close to the average value of the indicators of the highlight segment, so the fourth minute is taken as an extension point of the highlight segment; when a i+3 <n*avg_3, it indicates that the fourth minute is significantly lower than the average value of the indicators of the highlight period, and should not be taken as an extension point of the highlight segment. Therefore, the present application takes n*avg_m as the judgment condition for whether the i+m is taken as an extension point of the highlight segment [i, i+m-1].

[0130] In the embodiments of the present application, in order to improve the rationality of the adjustment parameter configuration, the adjustment parameter is generated according to the clip length obtained in step S410. Specifically, the adjustment parameter can be generated by Figure 6 The adjustment parameter is obtained by the steps shown in the following table:

[0131] S431, determining the reference window length according to the number of clip resources corresponding to different clip lengths.

[0132] For each clip length, the number of live highlight clip resources corresponding to the clip length is counted, and the number is the number of clip resources corresponding to the clip length. According to the number of clip resources corresponding to each clip length, the reference window length is determined from the clip lengths. For example, the clip length corresponding to the largest number of clip resources is taken as the reference window length.

[0133] S432, determining the first number of clip resources of the reference window length and the second number of clip resources of the reference window length; wherein the reference window length = the reference window length + 1.

[0134] The number of clip resources corresponding to the reference window length is the first number of clip resources, and the reference window length = the reference window length + 1, where 1 represents the window length difference of one adjacent two sliding time windows. That is, the reference window length is the next window length of the reference window length, and the number of clip resources corresponding to the reference window length is the second number of clip resources.

[0135] For example, the number of clip resources of three minutes is the largest, then three minutes is the reference window length, four minutes is the reference window length, the number of clip resources corresponding to three minutes is the first number of clip resources, and the number of clip resources corresponding to four minutes is the second number of clip resources.

[0136] S433, generating the adjustment parameter according to the first number of clip resources and the second number of clip resources.

[0137] The adjustment parameter is generated according to the ratio of the second number of clip resources to the first number of clip resources.

[0138] In an optional embodiment, the adjustment parameter can be updated every corresponding period, that is, the updated adjustment parameter is obtained by re-executing steps S410 and S430 when the update period is reached.

[0139] Steps S410-S430 are executed asynchronously with the subsequent steps.

[0140] Step S440, obtaining the play effect index value of the target live resource at different sampling times.

[0141] Step S450, generating a plurality of sliding time windows with different window lengths according to the window length value range, each sliding time window dividing the resource interval of the target live resource into a plurality of time slices.

[0142] According to the window length value range, a plurality of sliding time windows with different window lengths are generated, so that the window length of the generated sliding time window is within the window length value range, and after the sliding time windows are sorted according to the window length of the sliding time window, the length difference of the adjacent two sliding time windows is the same, for example, the length difference can be 1 sampling unit.

[0143] Step S460, calculating the index average value of each time slice according to the play effect index value in each time slice.

[0144] Step S470, determining the correction coefficient of each sliding time window in combination with the adjustment parameter.

[0145] From the above description, it can be known that the present application takes n*avg_m as the determination condition of whether to take the i+m sampling unit as the extension point of the highlight segment [i, i+m-1], and a i+3 n*avg_m is the critical condition, and the index average value avg_m+1 of the window length m+1 is as follows Formula 1:

[0146] avg _m+1 = (n*avg _m +m*avg _m ) / (m+1) (Formula 1)

[0147] Formula 1 is transformed to obtain the following Formula 2:

[0148] avg _m+1 / avg _m =(n+m) / (m+1) (Formula 2)

[0149] Then the correction coefficient of the window length m+1 compared with the window length m is as shown in Formula 3:

[0150] k m+1→m =(m+1) / (n+m) (Formula 3)

[0151] That is, the index average values of the window length m+1 and the window length m can be made consistent by the following Formula 4:

[0152] avg _m+1 *k m+1→m =avg _m (Formula 4)

[0153] Therefore, the correction coefficient k of the window length m+1 compared with the window length m can be obtained by calculating the above Formula 3m+1→m , and a correction factor k of the window length m relative to the window length m-1 can be obtained accordingly m→m-1 . That is, the following formula 5 is obtained:

[0154] k m→m-1 = m / (n+m-1) (Formula 5)

[0155] wherein k m→m-1 represents a correction factor of the window length m relative to the window length m-1, which can also be referred to as a correction factor of the sliding time window m relative to the previous sliding time window m-1, m represents the window length of the sliding time window m, and m-1 represents the window length of the previous sliding time window m-1 (the sliding time window located in front of the sliding time window in the order of the sliding time windows is referred to as the previous sliding time window of the sliding time window).

[0156] Thus, for any sliding time window m, the correction factor k m→m-1 of the sliding time window m relative to the previous sliding time window m-1 is calculated according to the above formula 5. Specifically, the sum of the adjustment parameter n, the window length m of the sliding time window, and -1 (i.e., the denominator part in the formula 5) is calculated, and the ratio of the window length m of the sliding time window to the sum (n+m-1) is taken as the correction factor of the sliding time window relative to the previous sliding time window.

[0157] The application calculates the correction factor of each sliding time window according to the correction factor of each sliding time window relative to the previous sliding time window. Specifically, a fixed correction factor is assigned to the sliding time window with the smallest window length, such as a fixed correction factor 1; for any non-minimal sliding time window, the correction factor of the sliding time window relative to the previous sliding time window is calculated according to the window length of the sliding time window and the adjustment parameter; and then the correction factor of the sliding time window is calculated according to the correction factor of the previous sliding time window and the correction factor of the sliding time window relative to the previous sliding time window. For example, the correction factor of the sliding time window can be obtained by the following formula 6:

[0158] P m = P m-1 *k m→m-1 (Formula 6)

[0159] wherein P m represents the correction factor of the sliding time window m, P m-1 represents the correction factor of the previous sliding time window m-1 of the sliding time window m, and k m→m-1is the correction coefficient of the sliding time window m compared with the previous sliding time window m-1. That is, the correction coefficient of the sliding time window m is obtained according to the product of the correction coefficient of the previous sliding time window and the correction coefficient of the sliding time window m compared with the previous sliding time window.

[0160] The general calculation formula 7 of the correction coefficient can be obtained by combining the formula 5 and the formula 6.

[0161]

[0162] wherein, P m is the correction coefficient of the sliding time window m; m represents the window length of the sliding time window m; is the correction coefficient of the sliding time window with the minimum window length, which is also the fixed correction coefficient; m min is the window length of the sliding time window with the minimum window length; n represents the adjustment parameter.

[0163] In step S480, for each time slice, the effect evaluation value of the time slice is obtained according to the index average value of the time slice and the correction coefficient of the sliding time window to which the time slice belongs.

[0164] In step S490, the target time slice is screened according to the effect evaluation value of each time slice, and the highlight segment in the target live broadcast resource is determined according to the target time slice.

[0165] It can be seen that the highlight segment detection method of the live broadcast resource provided by the embodiment of the application can generate the window length value range according to the editing length of the historical uploaded live broadcast highlight segment editing resource, so that the window lengths of the generated multiple sliding time windows can be adapted to the actual situation, and the rationality of the generated sliding time windows is improved. Moreover, the adjustment parameter is determined according to the editing length of the historical uploaded live broadcast highlight segment editing resource, so that the determination accuracy of the adjustment parameter is improved. The correction coefficient of the sliding time window compared with the previous sliding time window is determined according to the adjustment parameter and the window length, so that the index average values in the adjacent two sliding time windows with different window lengths can be accurately balanced through the correction coefficient. Finally, the correction coefficient of each sliding time window is generated in combination with the correction coefficient, the determination accuracy of the correction coefficient is improved, and the detection accuracy of the highlight segment is further improved.

[0166] Embodiment three

[0167] Figure 7 The structure schematic diagram of a highlight segment detection device of a live broadcast resource provided by the embodiment three of the application is shown. As shown in Figure 7As shown, the apparatus 700 comprises an index obtaining module 710, a window generating module 720, an average value calculating module 730, a correction coefficient determining module 740, an evaluation value calculating module 750, and a segment identifying module 760.

[0168] The index obtaining module 710 is configured to obtain the playing effect index values of the target live resource at different sampling times.

[0169] The window generating module 720 is configured to generate a plurality of sliding time windows with different window lengths, each of which divides a resource interval of the target live resource into a plurality of time segments.

[0170] The average value calculating module 730 is configured to calculate an index average value of each time segment according to the playing effect index values in the time segment.

[0171] The correction coefficient determining module 740 is configured to determine a correction coefficient of each sliding time window.

[0172] The evaluation value calculating module 750 is configured to obtain, for each time segment, an effect evaluation value of the time segment according to the index average value of the time segment and the correction coefficient of the sliding time window to which the time segment belongs.

[0173] The segment identifying module 760 is configured to filter out a target time segment according to the effect evaluation values of the time segments, and determine a highlight segment in the target live resource according to the target time segment.

[0174] In an optional implementation, the correction coefficient determining module 740 is configured to assign a fixed correction coefficient to the sliding time window with the smallest window length.

[0175] For any sliding time window other than the smallest one, the correction coefficient determining module 740 is configured to calculate a correction coefficient of the sliding time window relative to the previous sliding time window according to the window length of the sliding time window and an adjustment parameter, and calculate the correction coefficient of the sliding time window according to the correction coefficient of the previous sliding time window and the correction coefficient of the sliding time window relative to the previous sliding time window.

[0176] In an optional implementation, the correction coefficient determining module 740 is configured to calculate a sum of the adjustment parameter, the window length of the sliding time window, and -1.

[0177] The correction coefficient determining module 740 is configured to take the ratio of the window length of the sliding time window to the sum as the correction coefficient of the sliding time window relative to the previous sliding time window.

[0178] In an optional implementation, the apparatus further comprises an adjustment parameter generating module (not shown in the figure) configured to count the clip lengths of the live highlight segment clip resources uploaded historically.

[0179] determine a reference window length according to the number of clip resources corresponding to different clip lengths;

[0180] determine a first number of clip resources of the reference window length and a second number of clip resources of the reference window length; wherein the reference window length = the reference window length + 1;

[0181] generate the adjustment parameter according to the first number of clip resources and the second number of clip resources.

[0182] In an optional embodiment, the apparatus further comprises a window length range determination module (not shown in the figure) configured to count clip lengths of live highlight clip resources uploaded historically;

[0183] generate a window length value range according to the clip lengths;

[0184] The generating of the plurality of sliding time windows with different window lengths comprises: generating a plurality of sliding time windows with different window lengths according to the window length value range.

[0185] In an optional embodiment, the window length range determination module is configured to: obtain a system performance parameter; and generate a window length value range according to the clip lengths and the system performance parameter.

[0186] In an optional embodiment, the clip identification module 760 is configured to: sort the time slices in descending order of the effect evaluation values;

[0187] select time slices with top N effect evaluation values as target time slices;

[0188] aggregate the target time slices to obtain an aggregated time period;

[0189] take a clip corresponding to the aggregated time period as the highlight clip.

[0190] It can be seen that the live resource highlight clip detection apparatus provided by the embodiments of the present application divides a resource interval of a target live resource into time slices with different lengths by using a plurality of sliding time windows with different window lengths, and assigns corresponding correction coefficients to the sliding time windows with different window lengths, corrects an index average value of a time slice by using a correction coefficient of a sliding time window to which the time slice belongs to obtain an effect evaluation value of the time slice, and then identifies a highlight clip according to the effect evaluation value. By using the present solution, a complex content identification process is not required, and the highlight clip identification efficiency is high; and compared with a fixed time length detection method, the present solution can have higher detection accuracy.

[0191] Embodiment Four

[0192] Figure 8 A structural schematic diagram of a computing device provided by Embodiment Four of the present application is shown, and the specific embodiments of the present application do not limit the specific implementation of the computing device.

[0193] As shown in Figure 8 the computing device can include a processor 802, a communications interface 804, a memory 806, and a communications bus 808.

[0194] The processor 802, the communications interface 804, and the memory 806 can communicate with each other through the communications bus 808. The communications interface 804 is configured to communicate with network elements such as clients or other servers. The processor 802 is configured to execute the program 810, and specifically can execute the related steps in the live resource highlight segment detection method for the computing device.

[0195] Specifically, the program 810 can include program code including computer operation instructions.

[0196] The processor 802 can be a central processing unit CPU, or an application specific integrated circuit ASIC, or one or more integrated circuits configured to implement the embodiments of the present application. The one or more processors of the computing device can be the same type of processor, such as one or more CPUs; or can be different types of processors, such as one or more CPUs and one or more ASICs.

[0197] The memory 806 is configured to store the program 810. The memory 806 can include a high-speed RAM memory, and can also include a non-volatile memory such as at least one disk memory. The program 810 can specifically be configured to cause the processor 802 to execute the steps in the above method embodiments.

[0198] Embodiment Five

[0199] Embodiment Five of the present application provides a non-volatile computer storage medium, which stores at least one executable instruction or computer program. The executable instruction or computer program can cause a processor to perform operations corresponding to the live resource highlight segment detection method in any of the above method embodiments.

[0200] Embodiment Six

[0201] The embodiment six of the present application provides a computer program product, which comprises at least one executable instruction or computer program, and the executable instruction or computer program can enable a processor to execute the operation corresponding to the highlight segment detection method of the live resource in any method embodiment.

[0202] In summary, according to the computing device, computer storage medium and computer program product provided by the embodiment, the target live resource interval is divided into time slices with different lengths by using a plurality of sliding time windows with different window lengths, and the sliding time windows with different window lengths are assigned with corresponding correction coefficients, the effect evaluation value of the time slice is obtained by correcting the average value of the index of the time slice by using the correction coefficient of the sliding time window to which the time slice belongs, and then the highlight segment is identified according to the effect evaluation value. By adopting the scheme, the complex content identification process is not required, and the highlight segment identification efficiency is high; and compared with the fixed time length detection method, the detection precision is higher.

[0203] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other apparatus. Various general purpose systems can be used with these teachings, or with modifications thereof. The structure for a variety of these systems will be apparent to those of ordinary skill in the art from the description above. In addition, the present embodiments are not described with reference to any particular programming language. It will be appreciated that a variety of programming languages can be used to implement the teachings of the present embodiments as described herein, and any references below to specific languages are provided for disclosure of enablement only.

[0204] In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the application can be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been described in detail in order to avoid obscuring the understanding of this description.

[0205] Similarly, it is to be understood that the above description is only illustrative of the application and certain examples thereof, and is subject to the prior art. Numerous modifications and changes can be devised by those skilled in the art without departing from the true spirit and scope of the application. It is intended that the scope of the application be defined by the following claims as interpreted according to the principles of patent law including 35 U.S.C. § 1 12, 121 and 132, and any equivalents thereof.

[0206] Those skilled in the art will appreciate that the modules in the apparatuses in the embodiments can be adapted and placed in one or more apparatuses other than the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and furthermore can be split into multiple sub-modules or sub-units or sub-components. Any combination of all the features disclosed in the present specification (including the accompanying claims, abstract and drawings), and any method or process or apparatus of any combination of the features disclosed in the present specification (including the accompanying claims, abstract and drawings) can be taken, except that at least some of such features and / or processes or units are mutually exclusive, unless specifically stated otherwise. Each feature disclosed in the present specification (including the accompanying claims, abstract and drawings) can be replaced by alternative features providing the same, equivalent or similar functions unless specifically stated otherwise.

[0207] Furthermore, those skilled in the art will appreciate that different embodiments of the application have different features and that the features of one embodiment can not be incorporated into other embodiments. For example, in the following claims, any of the claims can be used to enable a single claim, multiple claims or single and multiple claims in combination. The following claims are hereby incorporated into the description of the embodiments of the application by this reference.

[0208] The various component embodiments of the present application can be implemented in hardware, or as software modules running in one or more processors, or in combinations thereof. Those skilled in the art will appreciate that a microprocessor or digital signal processor (DSP) can be used in practice to implement some or all of the functionality of some or all of the components according to the embodiments of the present application. The present application can also be implemented as a program of apparatus or device (for example, a computer program and a computer program product) for performing part or all of the methods described herein. Such program implementing the present application can be stored on a computer readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, or provided on a carrier signal, or in any other form.

[0209] It should be noted that the above-mentioned embodiments illustrate rather than limit the application, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word 'comprising' does not exclude the presence of elements or steps other than those listed in a claim. The word 'a' or 'an' preceding an element does not exclude the presence of a plurality of such elements. The application can be implemented by means of both hardware and software, and any combination thereof. In a unitary claim, several devices, apparatuses or means can be listed, comprising means for carrying out a certain task. The use of the term'means' in a claim is intended to refer to a combination of devices, apparatuses or means for carrying out a task. The word 'first','second', 'third', etc. do not imply any order. The use of these terms is to be construed as names. The steps of any of the methods recited herein do not have to be performed in the exact order disclosed, unless a step is explicitly described as following or preceding another step.

Claims

1. A highlight clip detection method of a live resource, characterized in that, The method comprises the following steps: obtaining the play effect index values of the target live resource at different sampling times; generating a plurality of sliding time windows with different window lengths, each sliding time window dividing the resource interval of the target live resource into a plurality of time slices; calculating the average index value of each time slice according to the play effect index values in each time slice; determining the correction coefficient of each sliding time window; wherein a fixed correction coefficient is assigned to the sliding time window with the smallest window length; for any non-minimal sliding time window, the sum of an adjustment parameter, the window length of the sliding time window and -1 is calculated, and the ratio of the window length of the sliding time window to the sum is taken as the correction coefficient of the sliding time window compared with the previous sliding time window; the correction coefficient of the sliding time window is calculated according to the correction coefficient of the previous sliding time window and the correction coefficient of the sliding time window compared with the previous sliding time window; wherein the adjustment parameter is determined according to the clip length of the historical uploaded live highlight clip resource; for each time slice, obtaining the effect evaluation value of the time slice according to the average index value of the time slice and the correction coefficient of the sliding time window to which the time slice belongs; screening a target time slice according to the effect evaluation values of the time slices, and determining a highlight clip in the target live resource according to the target time slice.

2. The method of claim 1, wherein, The adjustment parameter is determined by the following method: statistically analyzing the clip length of the historical uploaded live highlight clip resource; determining the baseline window length according to the number of clip resources corresponding to different clip lengths; determining the first number of clip resources of the baseline window length and the second number of clip resources of the reference window length; wherein the reference window length = baseline window length + 1; generating the adjustment parameter according to the first number of clip resources and the second number of clip resources.

3. The method of claim 1, wherein, Before the step of generating a plurality of sliding time windows with different window lengths, the method further comprises the following steps: statistically analyzing the clip length of the historical uploaded live highlight clip resource; 4. The method of claim 3, wherein, generating the window length value range according to the clip length; The step of generating a plurality of sliding time windows with different window lengths comprises the following step:

5. The method according to any one of claims 1-4, characterized in that, generating a plurality of sliding time windows with different window lengths according to the window length value range. The method further comprises the following step: obtaining the system performance parameter; The step of generating the window length value range according to the clip length comprises the following step: generating the window length value range according to the clip length and the system performance parameter.

6. A device for detecting highlights in live streaming resources, characterized in that, The step of screening a target time slice according to the effect evaluation values of the time slices, and determining a highlight clip in the target live resource according to the target time slice comprises the following steps: sorting the time slices in descending order of effect evaluation values; selecting the time slices with the top N effect evaluation values as the target time slices; aggregating the target time slices to obtain an aggregated time period; taking the clip corresponding to the aggregated time period as the highlight clip. The method comprises the following steps: an index obtaining module, configured to obtain the play effect index values of the target live resource at different sampling times; The window generation module is configured to generate a plurality of sliding time windows with different window lengths, each of the sliding time windows dividing a resource interval of the target live resource into a plurality of time slices; The average value calculation module is configured to calculate an average value of the playing effect index in each time slice; The correction coefficient determination module is configured to determine a correction coefficient for each sliding time window; wherein a fixed correction coefficient is assigned to the sliding time window with the smallest window length; for any non-minimal sliding time window, a sum of an adjustment parameter, the window length of the sliding time window and -1 is calculated, and a ratio of the window length of the sliding time window to the sum is taken as a correction coefficient of the sliding time window compared with a previous sliding time window; the correction coefficient of the sliding time window is calculated according to the correction coefficient of the previous sliding time window and the correction coefficient of the sliding time window compared with the previous sliding time window; wherein the adjustment parameter is determined according to a clip length of a live highlight clip resource uploaded in history; The evaluation value calculation module is configured to obtain an effect evaluation value of each time slice according to the average value of the playing effect index in the time slice and the correction coefficient of the sliding time window to which the time slice belongs; The segment identification module is configured to screen a target time slice according to the effect evaluation values of the time slices, and determine a highlight segment in the target live resource according to the target time slice.

7. A computing device, comprising: comprise: a processor, a memory, a communication interface and a communication bus, the processor, the memory and the communication interface being in communication with each other through the communication bus; The memory is used to store at least one executable instruction, and the executable instruction makes the processor execute the operation corresponding to the highlight segment detection method of the live resource in any one of claims 1-5.

8. A computer storage medium, characterized in that, The storage medium stores at least one executable instruction, and the executable instruction makes the processor execute the operation corresponding to the highlight segment detection method of the live resource in any one of claims 1-5.

9. A computer program product, characterised in that, comprise at least one executable instruction, and the executable instruction makes the processor execute the operation corresponding to the highlight segment detection method of the live resource in any one of claims 1-5.

Images