A perspective prediction method, device, equipment, and storage medium

Abstract

Embodiments of the present disclosure provide a perspective prediction method, apparatus, device, and storage medium. The method includes: obtaining at least two perspective prediction methods and corresponding prediction interval ranges for each perspective prediction method; obtaining the current video cache length in the target panoramic video being currently played and the target segment length corresponding to the target video segment to be downloaded currently; determining the target perspective prediction interval corresponding to the target video segment according to the current video cache length and the target segment length; matching the target perspective prediction interval with the prediction interval ranges to determine the target prediction interval range in which the target perspective prediction interval is located; and performing perspective prediction on the target video segment according to the target perspective prediction method corresponding to the target prediction interval range to obtain the target prediction perspective corresponding to the target video segment, thereby performing perspective prediction using different perspective prediction methods for the dynamically changing perspective prediction intervals, effectively ensuring the accuracy of perspective prediction.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to computer technologies, and in particular, to a viewing angle prediction method, apparatus, device, and storage medium. Background Art

[0002] With the rapid development of computer technologies, panoramic videos with a 360-degree all-round view can be captured. When playing a panoramic video, a user can switch the viewing angle to view video frames from different viewing angles.

[0003] A playback end usually caches a certain length of video data to reduce the risk of playback jitter and ensure smooth viewing. At the same time, the playback end also predicts the user's future viewing angle, so that it can only request to download the video stream from that viewing angle, reducing the transmission bandwidth. The accuracy of viewing angle prediction directly affects the video image quality perceived by the user.

[0004] Currently, during video playback, a fixed viewing angle prediction method is usually used for viewing angle prediction. However, the video cache length changes dynamically due to reasons such as bitrate changes and network state changes, resulting in dynamic changes in the viewing angle prediction interval. The prediction performance of the single viewing angle prediction method used also changes accordingly, and it is unable to effectively ensure the accuracy of viewing angle prediction. Summary of the Invention

[0005] The present disclosure provides a viewing angle prediction method, apparatus, device, and storage medium, which use different viewing angle prediction methods for viewing angle prediction for a dynamically changing viewing angle prediction interval, so as to effectively ensure the accuracy of viewing angle prediction and thereby improve the user viewing experience.

[0006] In a first aspect, embodiments of the present disclosure provide a viewing angle prediction method, including:

[0007] Obtaining at least two viewing angle prediction methods and a prediction interval range corresponding to each of the viewing angle prediction methods;

[0008] Obtaining a current video cache length in a target panoramic video being currently played and a target segment length corresponding to a target video segment to be downloaded currently;

[0009] Determining a target viewing angle prediction interval corresponding to the target video segment according to the current video cache length and the target segment length;

[0010] Matching the target viewing angle prediction interval with the prediction interval range to determine a target prediction interval range in which the target viewing angle prediction interval is located;

[0011] Performing viewing angle prediction on the target video segment according to a target viewing angle prediction method corresponding to the target prediction interval range to obtain a target prediction viewing angle corresponding to the target video segment.

[0012] In a second aspect, an embodiment of the present disclosure further provides a perspective prediction device, including:

[0013] a prediction information acquisition module, configured to acquire at least two perspective prediction methods and a prediction interval range corresponding to each of the perspective prediction methods;

[0014] a length information acquisition module, configured to acquire a current video cache length in a target panoramic video being played and a target segment length corresponding to a target video segment to be downloaded currently;

[0015] a target perspective prediction interval determination module, configured to determine a target perspective prediction interval corresponding to the target video segment according to the current video cache length and the target segment length;

[0016] a target perspective prediction interval matching module, configured to match the target perspective prediction interval with the prediction interval range to determine a target prediction interval range in which the target perspective prediction interval is located;

[0017] a perspective prediction module, configured to perform perspective prediction on the target video segment according to a target perspective prediction method corresponding to the target prediction interval range to obtain a target prediction perspective corresponding to the target video segment.

[0018] In a third aspect, an embodiment of the present disclosure further provides an electronic device, where the electronic device includes:

[0019] one or more processors;

[0020] a storage device, configured to store one or more programs,

[0021] when the one or more programs are executed by the one or more processors, the one or more processors implement the perspective prediction method according to any one of the embodiments of the present disclosure.

[0022] In a fourth aspect, an embodiment of the present disclosure further provides a storage medium including computer-executable instructions, where the computer-executable instructions are used to execute the perspective prediction method according to any one of the embodiments of the present disclosure when being executed by a computer processor.

[0023] In the embodiments of the present disclosure, by obtaining at least two perspective prediction methods and the corresponding prediction interval ranges for each perspective prediction method, the current video buffer length in the target panoramic video being played and the target segment length corresponding to the target video segment to be downloaded are obtained; according to the current video buffer length and the target segment length, the target perspective prediction interval corresponding to the target video segment is determined; the target perspective prediction interval is matched with the prediction interval ranges corresponding to each perspective prediction method to determine the target prediction interval range in which the target perspective prediction interval is located, and according to the target perspective prediction method corresponding to the target prediction interval range, perspective prediction is performed on the target video segment. Thus, by using the target perspective prediction method that matches the target perspective prediction interval for perspective prediction, the target prediction perspective corresponding to the target video segment can be obtained more accurately. By performing perspective prediction using different perspective prediction methods for dynamically changing perspective prediction intervals, the prediction accuracy can be maximized, thereby effectively ensuring the accuracy of perspective prediction and further enhancing the user viewing experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In combination with the accompanying drawings and with reference to the following specific embodiments, the above and other features, advantages, and aspects of the embodiments of the present disclosure will become more apparent. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that the original and elements are not necessarily drawn to scale.

[0025] Figure 1 is a schematic flowchart of a perspective prediction method provided by an embodiment of the present disclosure;

[0026] Figure 2 is an example of a perspective prediction interval matching process involved in an embodiment of the present disclosure;

[0027] Figure 3 is another schematic flowchart of a perspective prediction method provided by an embodiment of the present disclosure;

[0028] Figure 4 is an example of a perspective prediction process involved in an embodiment of the present disclosure;

[0029] Figure 5 is a schematic structural diagram of a perspective prediction device provided by an embodiment of the present disclosure;

[0030] Figure 6 is a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Instead, these embodiments are provided to more thoroughly and completely understand the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not used to limit the protection scope of the present disclosure.

[0032] It should be understood that the various steps recited in the method embodiments of the present disclosure can be executed in a different order and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this regard.

[0033] The term "including" and its variations used herein are open-ended, that is, "including but not limited to". The term "based on" is "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0034] It should be noted that the concepts such as "first", "second", etc. mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order of functions executed by these devices, modules or units or their interdependent relationships.

[0035] It should be noted that the modifications of "one" and "multiple" mentioned in the present disclosure are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly stated in the context, it should be understood as "one or more".

[0036] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0037] It can be understood that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, usage scopes, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0038] For example, when responding to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested by the user will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, an application program, a server or a storage medium that executes the operations of the technical solutions of the present disclosure according to the prompt message.

[0039] As an optional but non-limiting implementation manner, in response to receiving an active request from a user, the manner of sending a prompt message to the user may be, for example, a pop-up window manner, and the prompt message may be presented in text in the pop-up window. In addition, the pop-up window may also carry a selection control for the user to select "agree" or "disagree" to provide personal information to the electronic device.

[0040] It can be understood that the above notification and the process of obtaining user authorization are only illustrative and do not constitute a limitation on the implementation manner of the present disclosure. Other manners that meet relevant laws and regulations can also be applied to the implementation manner of the present disclosure.

[0041] Figure 1 FIG. is a schematic flowchart of a view prediction method provided by an embodiment of the present disclosure. The embodiment of the present disclosure is applicable to the situation of predicting the viewing view of a shard in a panoramic video. This method can be executed by a view prediction device, and the device can be implemented in the form of software and / or hardware. Optionally, it is implemented by an electronic device, and the electronic device can be a mobile terminal, a PC terminal, a server, etc.

[0042] As Figure 1 shown, the view prediction method specifically includes the following steps:

[0043] S110. Obtain at least two view prediction manners and the prediction interval corresponding to each view prediction manner.

[0044] Among them, the perspective prediction method can be a method for predicting the viewing perspective in a panoramic video. For example, the perspective prediction method can include but is not limited to: at least two of the general linear regression method, the monotonic interval linear regression method, the weighted linear regression method, and the perspective prediction method based on the current playback perspective. Among them, the general linear regression method (LinearRegression) can be a method of obtaining a predicted perspective by fitting the historical perspective change trajectory through linear regression. The monotonic interval linear regression method (Truncated Linear Regression) can refer to a variant of Linear Regression, that is, filtering the input once before linear regression, and only using a monotonically increasing or decreasing interval closest to the current playback time to fit the method to obtain the predicted perspective. The weighted linear regression method (Weighted LinearRegression) can refer to another variant of Linear Regression, that is, when performing linear fitting, different observation points are given different weights to obtain the predicted perspective. The perspective prediction method based on the current playback perspective can be to use the weighted result of the current playback perspective direction and the initial perspective direction (such as the (0,0) direction) as the predicted perspective. Among them, the general linear regression method has strong robustness. The monotonic interval linear regression method is most accurate when predicting the viewing angle in the near future, such as the viewing angle in the first and second seconds in the future. The weighted linear regression method is most accurate when predicting the viewing angle in the distant future, such as the viewing angle in the fourth and fifth seconds in the future. The viewing angle prediction method based on the current playback viewing angle can be used to accurately predict the viewing angle in the distant future. Each viewing angle prediction method has different prediction performance at different viewing angle prediction intervals (i.e., the time interval between the viewing angle prediction moment and the current video playback moment).

[0045] Each perspective prediction method corresponds to a prediction interval. The prediction interval may consist of a minimum prediction interval and a maximum prediction interval. The minimum prediction interval may refer to the minimum time interval between the perspective prediction moment and the current playback moment. The maximum prediction interval may refer to the maximum time interval between the perspective prediction moment and the current playback moment. Different perspective prediction methods may correspond to different prediction intervals.

[0046] Specifically, at least two perspective prediction methods can be pre-configured in the server, and the prediction interval range corresponding to the best prediction performance of each perspective prediction method can be determined. Since the prediction performance of the perspective prediction method can be continuously optimized with the adjustment of prediction parameters, the prediction interval range corresponding to each perspective prediction method will also change accordingly. After the server determines the current optimal prediction interval range each time, it can actively send all the perspective prediction methods and the prediction interval range corresponding to each perspective prediction method to the playback end. Alternatively, in response to the startup operation of the playback end, the playback end can send an information acquisition request to the server. When the server receives this information acquisition request, it will send all the perspective prediction methods and the prediction interval range corresponding to each perspective prediction method to the playback end, so that the playback end can obtain multiple perspective prediction methods with the current optimal prediction parameters and the prediction interval range corresponding to each perspective prediction method sent by the server.

[0047] It should be noted that the server can directly send the prediction interval range corresponding to each perspective prediction method, or send the prediction interval threshold for distinguishing the prediction performance of every two adjacent perspective prediction methods, so that the playback end can determine the prediction interval range corresponding to each perspective prediction method based on the prediction interval threshold.

[0048] Exemplarily, S110 may include: obtaining at least two perspective prediction methods, the prediction method arrangement order, and the prediction interval threshold for distinguishing the prediction performance of every two adjacent perspective prediction methods; and determining the prediction interval range corresponding to each perspective prediction method based on the prediction method arrangement order and the prediction interval threshold.

[0049] Among them, the prediction method arrangement order can be obtained by sorting various perspective prediction methods according to the sequence of prediction intervals when the prediction performance is optimal. For example, Figure 2 gives an example of a perspective prediction interval matching process. As Figure 2 shown, 4 perspective prediction methods can be used in combination. The prediction method arrangement order corresponding to these 4 perspective prediction methods is: the monotonic interval linear regression method, the general linear regression method, the weighted linear regression method, and the perspective prediction method based on the current playback perspective. It should be noted that with the optimization of the prediction performance of the perspective prediction method, the prediction method arrangement order may change.

[0050] Among them, the prediction interval threshold can be the critical value for distinguishing the prediction performance of two adjacent perspective prediction methods. The number of prediction interval thresholds is determined based on the number of perspective prediction methods. The number of prediction interval thresholds is 1 less than the number of perspective prediction methods. For example, Figure 2 the 4 perspective prediction methods in [reference] correspond to 3 prediction interval thresholds, which are T1, T2, and T3 respectively.

[0051] Specifically, the playback end can divide the timeline according to all the prediction interval thresholds sent by the server, obtain the same number of prediction interval ranges as the number of view prediction methods, and match all the view prediction methods and all the divided prediction interval ranges one by one according to the arrangement order of the prediction methods, so as to determine the prediction interval range corresponding to each view prediction method. For example, as Figure 2 shown, by dividing the timeline using 3 prediction interval thresholds, 4 prediction interval ranges can be obtained, namely [0, T1], [T1, T2], [T2, T3] and [T3, Tmax], where "0" on the timeline represents the current playback moment of the panoramic video. "Tmax" represents the maximum prediction interval, that is, the time interval between the current playback moment of the panoramic video and the last video moment. According to the Figure 2 arrangement order of the prediction methods in, it can be determined that the prediction interval range corresponding to the monotonic interval linear regression method is [0, T1], the prediction interval range corresponding to the general linear regression method is [T1, T2], the prediction interval range corresponding to the weighted linear regression method is [T2, T3], and the prediction interval range corresponding to the view prediction method based on the current playback view is [T3, Tmax].

[0052] S120. Obtain the current video cache length in the currently played target panoramic video and the target segment length corresponding to the currently to-be-downloaded target video segment.

[0053] Among them, the target panoramic video may refer to the panoramic video currently being played. The panoramic video in this embodiment may be, but is not limited to, a VR (Virtual Reality) panoramic video. The target panoramic video can be downloaded and played in the form of video segments, that is to say, the target panoramic video includes multiple video segments. By downloading and playing each video segment one by one in sequence, the download and playback operations of the target panoramic video are completed. The current video cache length may refer to the length of the panoramic video that has been downloaded and is cached but not yet played in the playback end. The target video segment may refer to the panoramic video segment that needs to be downloaded at the current moment during the playback of the target panoramic video. The target segment length may refer to the video length of the target video segment. For example, if the target video segment is a 5-second video, the target segment length is 5 seconds. It should be noted that the segment lengths corresponding to each video segment may be the same or different. The current video cache length changes dynamically with the download operation of the video segment and the video playback operation of the playback end.

[0054] Specifically, when the playback end needs to download a video segment of the target panoramic video each time, it can obtain the current video cache length and the target segment length corresponding to the target video segment to be downloaded currently, so as to predict the playback perspective corresponding to the video segment to be downloaded this time based on the current video cache length and the target segment length obtained this time.

[0055] S130. Determine the target perspective prediction interval corresponding to the target video segment according to the current video cache length and the target segment length.

[0056] Among them, the target perspective prediction interval may refer to the time interval between the current playback moment of the target panoramic video and the perspective prediction moment of the target video segment. Among them, the perspective prediction moment may be a reference moment for predicting the perspective of the video segment. The predicted perspective corresponding to the entire target video segment is consistent.

[0057] Specifically, the start moment or the end moment of the target video segment can be used as the reference moment for predicting the perspective, that is, the perspective prediction moment. Thus, when the start moment of the target video segment is used as the perspective prediction moment, the current video cache length can be directly used as the target perspective prediction interval corresponding to the target video segment. When the end moment of the target video segment is used as the perspective prediction moment, the current video cache length and the target segment length can be added together, and the obtained addition result is used as the target perspective prediction interval corresponding to the target video segment.

[0058] Exemplarily, S130 may include: determining an intermediate length corresponding to the target segment length; adding the current video cache length and the intermediate length together, and determining the obtained addition result as the target perspective prediction interval corresponding to the target video segment.

[0059] Specifically, half of the target segment length can be used as the intermediate length corresponding to the target segment length, and the current video cache length and the intermediate length are added together, and the obtained addition result is determined as the perspective prediction interval, so that the perspective prediction of the front and back parts of the target video segment can be taken into account at the same time, making the perspective prediction result of the entire target video segment more accurate. For example, if the current video cache length is 10 seconds and the target segment length is 5 seconds, then the target perspective prediction interval corresponding to the target video segment is determined to be 12.5 seconds.

[0060] S140. Match the target perspective prediction interval with the prediction interval range to determine the target prediction interval range where the target perspective prediction interval is located.

[0061] Specifically, the target perspective prediction interval can be matched with the prediction interval range corresponding to each perspective prediction method to determine the target prediction interval range where the target perspective prediction interval is located. For example, Figure 2As shown, if the target perspective prediction interval is within the prediction interval [T1, T2], the target prediction interval is determined to be [T1, T2].

[0062] S150. According to the target perspective prediction method corresponding to the target prediction interval, perform perspective prediction on the target video segment to obtain the target prediction perspective corresponding to the target video segment.

[0063] Specifically, after determining the target prediction interval in which the target perspective prediction interval is located, the target perspective prediction method corresponding to the target prediction interval can be selected as the perspective prediction method that is optimal for predicting the perspective of the target video segment. Thus, the target video segment can be more accurately predicted using the target perspective prediction method. When predicting the perspective of each video segment, the optimal perspective prediction method can be dynamically selected for perspective prediction, thereby maximizing the prediction accuracy, effectively ensuring the accuracy of the perspective prediction, and further enhancing the user viewing experience.

[0064] The technical solution of the embodiments of the present disclosure obtains at least two perspective prediction methods and the prediction interval corresponding to each perspective prediction method, obtains the current video cache length in the currently played target panoramic video and the target segment length corresponding to the target video segment to be downloaded; determines the target perspective prediction interval corresponding to the target video segment according to the current video cache length and the target segment length; matches the target perspective prediction interval with the prediction interval corresponding to each perspective prediction method to determine the target prediction interval in which the target perspective prediction interval is located, and performs perspective prediction on the target video segment according to the target perspective prediction method corresponding to the target prediction interval. Thus, the target prediction perspective corresponding to the target video segment can be more accurately obtained by using the target perspective prediction method that matches the target perspective prediction interval. By performing perspective prediction using different perspective prediction methods for the dynamically changing perspective prediction interval, the prediction accuracy can be maximized, thereby effectively ensuring the accuracy of the perspective prediction and further enhancing the user viewing experience.

[0065] Based on the above technical solution, S110 may include: in response to a play trigger operation of the target panoramic video, sending the target video identifier corresponding to the target panoramic video to the server, so that the server determines the target video type corresponding to the target panoramic video based on the target video identifier, and sends at least two perspective prediction methods corresponding to the target video type and the prediction interval corresponding to each perspective prediction method; receiving at least two perspective prediction methods and the prediction interval corresponding to each perspective prediction method sent by the server.

[0066] Among them, the video types can be classified according to the change situation of the video content. For example, if the video content changes little, the user may randomly switch perspectives when watching the video. At this time, the user's perspective is not easily predictable, and this type of video can be classified into one category, such as the first video type. If the video content changes greatly and there is an attractive target that remains stationary or moves at a constant speed, the user's perspective is relatively easy to predict at this time, and this type of video can be classified into another category, such as the second video type. For different video types, different prediction interval ranges corresponding to each perspective prediction method are determined.

[0067] Specifically, when the user wants to watch the target panoramic video, a play trigger operation can be performed on the target panoramic video, such as clicking the start play button corresponding to the target panoramic video. When the playback device detects the play trigger operation of the target panoramic video, it can send the target video identifier corresponding to the target panoramic video to the server. Based on the correspondence relationship between the video identifier and the video type determined in advance, the server can obtain the target video type corresponding to the target video identifier, and send all the perspective prediction methods and the corresponding prediction interval ranges corresponding to the target video type to the playback device, so that the playback device can perform perspective prediction on the target panoramic video based on all the perspective prediction methods and the corresponding prediction interval ranges corresponding to the target video type, thereby enabling targeted perspective prediction in the video dimension and further improving the accuracy of perspective prediction.

[0068] Based on the above technical solutions, S110 may further include: in response to the startup operation of the playback device, sending the target user identifier corresponding to the video playback user to the server, so that the server determines the target user group based on the target user identifier, and sends at least two perspective prediction methods corresponding to the target user group and the prediction interval range corresponding to each perspective prediction method; receiving at least two perspective prediction methods and the prediction interval range corresponding to each perspective prediction method sent by the server.

[0069] Among them, the user group can be a set of users obtained by classifying all users according to the user's perspective switching habits. For example, users with the same perspective switching habits are classified into one user group, so that each user in the user group can share the same perspective prediction method and prediction interval range. For different user groups, different prediction interval ranges corresponding to each perspective prediction method can be determined.

[0070] Specifically, when a user wants to open a playback end for video playback, a startup operation can be triggered on the playback end, such as clicking on the application icon corresponding to the playback end. When the startup operation of the playback end is detected, the target user identifier corresponding to the video playback user can be sent to the server. Based on the correspondence relationship between the user identifier and the user group determined in advance, the server can obtain the target user group corresponding to the target user identifier, and send all the perspective prediction methods and the corresponding prediction interval ranges corresponding to the target user group to the playback end, so that the playback end can perform perspective prediction on each panoramic video played after startup based on all the perspective prediction methods and the corresponding prediction interval ranges corresponding to the target user group, thereby enabling targeted perspective prediction at the user dimension and further improving the accuracy of perspective prediction.

[0071] Based on the above technical solutions, in this embodiment, the video dimension and the user dimension can also be combined to perform targeted perspective prediction in both the video dimension and the user dimension, further improving the accuracy of perspective prediction. For example, in response to a playback trigger operation of a target panoramic video, the target video identifier corresponding to the target panoramic video and the target user identifier corresponding to the video playback user are sent to the server, so that the server can determine at least two perspective prediction methods corresponding to the target panoramic video and the prediction interval range corresponding to each perspective prediction method based on the target video identifier and the target user identifier and send them. Thereby, the playback end performs perspective prediction on the target panoramic video based on all the perspective prediction methods and the corresponding prediction interval ranges under the target video identifier and the target user identifier, further improving the accuracy of perspective prediction.

[0072] Figure 3 The flowchart of another perspective prediction method provided by an embodiment of the present disclosure details the process of updating the prediction parameters and the corresponding prediction interval ranges in each perspective prediction method based on the above disclosed embodiments. The explanations of the same or corresponding terms as those in the above disclosed embodiments are not repeated here.

[0073] As Figure 3 shown, the perspective prediction method specifically includes the following steps:

[0074] S310. Obtain at least two perspective prediction methods and the prediction interval range corresponding to each perspective prediction method.

[0075] S320. Obtain the current video cache length in the currently played target panoramic video and the target segment length corresponding to the target video segment to be downloaded currently.

[0076] S330. Determine the target perspective prediction interval corresponding to the target video segment according to the current video cache length and the target segment length.

[0077] S340. Match the target perspective prediction interval with the prediction interval range to determine the target prediction interval range where the target perspective prediction interval is located.

[0078] S350. According to the target perspective prediction method corresponding to the target prediction interval range, perform perspective prediction on the target video segment to obtain the target prediction perspective corresponding to the target video segment.

[0079] S360. Obtain the actual playback perspective corresponding to the target video segment.

[0080] Among them, the real-time playback perspective can refer to the actual viewing perspective of the user when watching the target video segment.

[0081] Specifically, after obtaining the target prediction perspective corresponding to the target video segment, the playback end can perform segmented download of the target video segment based on the target prediction perspective, so as to obtain the target video segment under the target prediction perspective and perform caching. When the user watches the target video segment, the user may switch to the target prediction perspective and watch the target video segment cached under the target prediction perspective, or may switch to the target video segment under other playback perspectives different from the target prediction perspective. Therefore, when the user actually watches the target video segment, the actual playback perspective corresponding to the target video segment can be obtained.

[0082] For example, Figure 4 An example of a perspective prediction process is given. As Figure 4 shown, the perspective prediction module in the playback end sends the current video cache length to the perspective prediction module. The perspective prediction module performs perspective prediction according to the current video cache length and the target segment length, and after determining the target prediction perspective corresponding to the target video segment, it can send a video segment download request corresponding to the target video segment to the video segment processing module in the server. The video segment processing module can send the target video segment under the target prediction perspective to the video playback module for caching and playback.

[0083] S370. Send the actual playback perspective corresponding to the target video segment to the server, so that the server stores the actual playback perspective in the perspective database, and based on the perspective database, updates the prediction parameters and the corresponding prediction interval range in each perspective prediction method.

[0084] Among them, the perspective database can be used to store the actual playback perspective information of the user when watching the target panoramic video. Specifically, as Figure 4As shown in the figure, the perspective prediction module in the playback end can report the actual playback perspective of each video segment in the target panoramic video to the perspective information collection module in the server, and the perspective information collection module stores the actual playback perspective of each video segment in the perspective database. When there is enough real user perspective data stored in the perspective database, the prediction parameters and the corresponding prediction interval ranges in each perspective prediction method can be iteratively updated based on the actual playback perspective trajectory information corresponding to the panoramic video in the perspective database. Thus, the prediction parameters and the prediction interval ranges can be updated more accurately in a real-data-driven manner, further improving the accuracy of the predicted perspective. In this embodiment, the prediction parameters and the corresponding prediction interval ranges in each perspective prediction method can be updated through online evaluation or offline evaluation. For online evaluation and update, the perspective prediction interval corresponding to each video segment in the panoramic video, the perspective prediction method used, the predicted perspective, and the actual viewing perspective need to be stored in the perspective database, so as to directly optimize the prediction parameters in the perspective prediction method and adjust the corresponding prediction interval ranges based on the actual perspective prediction situation and the actual viewing situation of the panoramic video. For offline evaluation and update, various perspective prediction situations of the panoramic video can be simulated, and based on the simulated various predicted perspective trajectory information and the actual viewing and actual playback perspective trajectory information, the prediction parameters in the perspective prediction method can be more accurately optimized and updated, and the corresponding prediction interval ranges can be adjusted and updated.

[0085] Exemplarily, S370 may include: iteratively updating the prediction parameters and the corresponding prediction interval ranges in each perspective prediction method offline based on the actual playback perspective trajectory information corresponding to the panoramic video in the perspective database; if the current preset convergence condition is satisfied, stop the iterative update and obtain the perspective prediction method containing the current prediction parameters and the corresponding current prediction interval range.

[0086] Specifically, as Figure 4As shown in the figure, the offline update module in the server can iteratively update the prediction parameters and the corresponding prediction interval ranges in each perspective prediction method offline according to the actual playback perspective trajectory information corresponding to the panoramic video in the perspective database. For example, the prediction parameters can be updated first, and then the prediction interval range can be updated based on the updated prediction parameters. For example, the prediction interval range can be updated by updating the prediction interval threshold, and then the prediction parameters can be updated again, and so on, iteratively updating in a loop. For example, through the performance analysis simulation platform, various viewing situations of the panoramic video are simulated, and based on the actual playback perspective trajectory information corresponding to the panoramic video, the performance of each perspective prediction method is evaluated offline, and the parameters of each perspective prediction method are optimized to confirm the optimal parameter settings of each perspective prediction method. Based on the performance comparison of the perspective prediction method with the optimized parameters at the current time in different prediction interval ranges, the prediction interval range corresponding to the optimal performance is determined. For example, a time point on the time axis that can more accurately distinguish the performance advantages and disadvantages of different perspective prediction methods is used as the prediction interval threshold. After each iterative update of the prediction parameters and the prediction interval range, it can be detected whether the change in the prediction interval range tends to be stable or whether the number of iterations is equal to the preset number. If not, the prediction parameters and the prediction interval range are continued to be iteratively updated. If so, it indicates that the current preset convergence condition is satisfied, and at this time, the iterative update can be stopped to obtain the perspective prediction method including the current optimal current prediction parameters and the corresponding current optimal current prediction interval range.

[0087] It should be noted that since the online users and the videos in the video library are updated in real time, it is necessary to dynamically update the prediction parameters and the prediction interval range in the perspective prediction method to optimize the current overall prediction accuracy. In this embodiment, the threshold parameters can be updated at a fixed frequency and stored in the server. When performing targeted perspective prediction in the video dimension or the user dimension, the prediction parameters and the prediction interval range can be dynamically updated based on the sample data in the video dimension or the user dimension, so as to obtain the prediction parameters and the prediction interval range in each dimension, further ensuring the accuracy of perspective prediction.

[0088] The technical solution of the embodiment of the present disclosure sends the actual playback perspective corresponding to the target video shard to the server, so that the server stores the actual playback perspective in the perspective database, and based on the perspective database, updates the prediction parameters and the corresponding prediction interval range in each perspective prediction method, so that the prediction parameters and the prediction interval range can be more accurately dynamically updated in a real data-driven manner, further improving the accuracy of the predicted perspective.

[0089] Figure 5 The structural schematic diagram of a perspective prediction device provided by the embodiment of the present disclosure is as Figure 5As shown in the figure, the device specifically includes: a prediction information acquisition module 410, a length information acquisition module 420, a target perspective prediction interval determination module 430, a target perspective prediction interval matching module 440, and a perspective prediction module 450.

[0090] Among them, the prediction information acquisition module 410 is configured to acquire at least two perspective prediction methods and the prediction interval ranges corresponding to each of the perspective prediction methods; the length information acquisition module 420 is configured to acquire the current video cache length in the currently played target panoramic video and the target segment length corresponding to the currently to-be-downloaded target video segment; the target perspective prediction interval determination module 430 is configured to determine the target perspective prediction interval corresponding to the target video segment according to the current video cache length and the target segment length; the target perspective prediction interval matching module 440 is configured to match the target perspective prediction interval with the prediction interval ranges to determine the target prediction interval range in which the target perspective prediction interval is located; the perspective prediction module 450 is configured to perform perspective prediction on the target video segment according to the target perspective prediction method corresponding to the target prediction interval range to obtain the target prediction perspective corresponding to the target video segment.

[0091] The technical solution provided by the embodiments of the present disclosure obtains at least two perspective prediction methods and the prediction interval ranges corresponding to each perspective prediction method, and acquires the current video cache length in the currently played target panoramic video and the target segment length corresponding to the currently to-be-downloaded target video segment; determines the target perspective prediction interval corresponding to the target video segment according to the current video cache length and the target segment length; matches the target perspective prediction interval with the prediction interval ranges corresponding to each perspective prediction method to determine the target prediction interval range in which the target perspective prediction interval is located, and performs perspective prediction on the target video segment according to the target perspective prediction method corresponding to the target prediction interval range, so as to perform perspective prediction using the target perspective prediction method that matches the target perspective prediction interval, and can more accurately obtain the target prediction perspective corresponding to the target video segment. By performing perspective prediction using different perspective prediction methods for the dynamically changing perspective prediction interval, the prediction accuracy can be maximized, thereby effectively ensuring the accuracy of perspective prediction and further improving the user viewing experience.

[0092] Based on the above technical solution, the prediction information acquisition module 410 is specifically configured to:

[0093] acquire at least two perspective prediction methods, the prediction method arrangement order, and the prediction interval threshold for distinguishing the prediction performances of every two adjacent perspective prediction methods; and determine the prediction interval range corresponding to each of the perspective prediction methods based on the prediction method arrangement order and the prediction interval threshold.

[0094] Based on the above technical solutions, the target perspective prediction interval determination module 430 is specifically configured to:

[0095] Determine the intermediate length corresponding to the target segment length; add the current video buffer length and the intermediate length, and determine the obtained addition result as the target perspective prediction interval corresponding to the target video segment.

[0096] Based on the above technical solutions, the device further includes:

[0097] An actual playback perspective acquisition module, configured to acquire the actual playback perspective corresponding to the target video segment after obtaining the target prediction perspective corresponding to the target video segment;

[0098] An actual playback perspective sending module, configured to send the actual playback perspective corresponding to the target video segment to the server, so that the server stores the actual playback perspective in the perspective database, and based on the perspective database, updates the prediction parameters and the corresponding prediction interval range in each perspective prediction method.

[0099] Based on the above technical solutions, the connection update module in the server is specifically configured to:

[0100] Based on the actual playback perspective trajectory information of the panoramic video in the perspective database, iteratively update the prediction parameters and the corresponding prediction interval range in each of the perspective prediction methods offline; if the current preset convergence condition is satisfied, stop the iterative update, and obtain the perspective prediction method including the current prediction parameters and the corresponding current prediction interval range.

[0101] Based on the above technical solutions, the prediction information acquisition module 410 is specifically configured to:

[0102] In response to the playback trigger operation of the target panoramic video, send the target video identifier corresponding to the target panoramic video to the server, so that the server determines the target video type corresponding to the target panoramic video based on the target video identifier, and sends at least two perspective prediction methods corresponding to the target video type and the prediction interval range corresponding to each perspective prediction method;

[0103] Receive at least two perspective prediction methods sent by the server and the prediction interval range corresponding to each of the perspective prediction methods.

[0104] Based on the above technical solutions, the prediction information acquisition module 410 is specifically configured to:

[0105] In response to the startup operation of the playback end, send the target user identifier corresponding to the video playback user to the server, so that the server determines the target user group based on the target user identifier, and sends at least two perspective prediction methods corresponding to the target user group and the prediction interval range corresponding to each perspective prediction method;

[0106] Receive at least two perspective prediction methods sent by the server and the prediction interval range corresponding to each perspective prediction method.

[0107] Based on the above technical solutions, the perspective prediction methods include at least two of: general linear regression method, monotonic interval linear regression method, weighted linear regression method, and perspective prediction method based on the current playback perspective.

[0108] The perspective prediction device provided by the embodiments of the present disclosure can execute the perspective prediction method provided by any embodiment of the present disclosure, and has the corresponding functional modules and beneficial effects for executing the perspective prediction method.

[0109] It should be noted that the various units and modules included in the above device are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the embodiments of the present disclosure.

[0110] Figure 6 It is a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure. Refer to the following Figure 6 , which shows a schematic structural diagram of an electronic device (such as Figure 6 the terminal device or server in) 500 suitable for implementing the embodiments of the present disclosure. The terminal device in the embodiments of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The electronic device shown is only an example and should not bring any limitation to the functions and usage scope of the embodiments of the present disclosure.

[0111] Such as Figure 6As shown, the electronic device 500 may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 501, which may perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage device 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An editing / output (I / O) interface 505 is also connected to the bus 504.

[0112] Generally, the following devices may be connected to the I / O interface 505: an input device 506 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 507 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 508 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 509. The communication device 509 may allow the electronic device 500 to communicate with other devices wirelessly or wiredly to exchange data. Although Figure 6 an electronic device 500 with various devices is shown, it should be understood that it is not required to implement or have all the shown devices. Instead, more or fewer devices may be implemented or had.

[0113] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart may be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes program codes for performing the method shown in the flowchart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication device 509, or installed from the storage device 508, or installed from the ROM 502. When the computer program is executed by the processing device 501, the above functions defined in the method of the embodiment of the present disclosure are executed.

[0114] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0115] The electronic device provided by the embodiment of the present disclosure and the perspective prediction method provided by the above embodiment belong to the same inventive concept. Technical details not described in detail in this embodiment may be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0116] The embodiment of the present disclosure provides a computer storage medium, on which a computer program is stored, and when the program is executed by a processor, the perspective prediction method provided by the above embodiment is implemented.

[0117] It should be noted that the above-mentioned computer-readable medium in the present disclosure may be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium that contains or stores a program, and the program may be used by or in conjunction with an instruction execution system, apparatus, or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, and the computer-readable signal medium may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0118] In some embodiments, the client and the server may communicate using any currently known or future-developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include local area networks ("LANs"), wide area networks ("WANs"), the Internet (e.g., the Internet), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0119] The above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or it may exist separately and not be assembled into the electronic device.

[0120] The above-mentioned computer-readable medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device is caused to:

[0121] The above computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: obtain at least two perspective prediction methods and the corresponding prediction interval ranges for each of the perspective prediction methods; obtain the current video cache length in the currently played target panoramic video and the target segment length corresponding to the target video segment to be downloaded currently; determine the target perspective prediction interval corresponding to the target video segment according to the current video cache length and the target segment length; match the target perspective prediction interval with the prediction interval ranges to determine the target prediction interval range in which the target perspective prediction interval is located; and perform perspective prediction on the target video segment according to the target perspective prediction method corresponding to the target prediction interval range to obtain the target prediction perspective corresponding to the target video segment.

[0122] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages or combinations thereof. The programming languages include, but are not limited to, object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0123] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combinations of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system for performing the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0124] The units involved in the embodiments of the present disclosure can be implemented in software or in hardware. Among them, the name of the unit does not constitute a limitation on the unit itself in some cases. For example, the first acquisition unit can also be described as "the unit for acquiring at least two Internet protocol addresses".

[0125] The functions described above herein can be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that can be used include: Field Programmable Gate Array (FPGA), Application Specific Integrated Circuit (ASIC), Application Specific Standard Product (ASSP), System on Chip (SOC), Complex Programmable Logic Device (CPLD), and so on.

[0126] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media would include electrical connections based on one or more wires, portable computer disks, hard disks, Random Access Memory (RAM), Read Only Memory (ROM), Erasable Programmable Read Only Memory (EPROM or Flash Memory), optical fibers, portable compact disc read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0127] According to one or more embodiments of the present disclosure, [Example 1] provides a perspective prediction method, including:

[0128] Acquiring at least two perspective prediction methods and a prediction interval range corresponding to each of the perspective prediction methods;

[0129] Acquiring a current video cache length in a target panoramic video being currently played and a target segment length corresponding to a target video segment to be downloaded currently;

[0130] Determining a target perspective prediction interval corresponding to the target video segment according to the current video cache length and the target segment length;

[0131] Matching the target perspective prediction interval with the prediction interval range to determine a target prediction interval range in which the target perspective prediction interval is located;

[0132] Performing perspective prediction on the target video segment according to the target perspective prediction method corresponding to the target prediction interval range, to obtain the target prediction perspective corresponding to the target video segment.

[0133] According to one or more embodiments of the present disclosure, [Example 2] provides a perspective prediction method, further comprising:

[0134] Optionally, the obtaining at least two perspective prediction methods and the prediction interval range corresponding to each perspective prediction method includes:

[0135] Obtaining at least two perspective prediction methods, the prediction method arrangement order, and a prediction interval threshold for distinguishing the prediction performances of every two adjacent perspective prediction methods;

[0136] Based on the prediction method arrangement order and the prediction interval threshold, determining the prediction interval range corresponding to each perspective prediction method.

[0137] According to one or more embodiments of the present disclosure, [Example 3] provides a perspective prediction method, further comprising:

[0138] Optionally, the determining the target perspective prediction interval corresponding to the target video segment according to the current video cache length and the target segment length includes:

[0139] Determining an intermediate length corresponding to the target segment length;

[0140] Adding the current video cache length and the intermediate length, and determining the obtained addition result as the target perspective prediction interval corresponding to the target video segment.

[0141] According to one or more embodiments of the present disclosure, [Example 4] provides a perspective prediction method, further comprising:

[0142] Optionally, after obtaining the target prediction perspective corresponding to the target video segment, further comprising:

[0143] Obtaining the actual playback perspective corresponding to the target video segment;

[0144] Sending the actual playback perspective corresponding to the target video segment to the server, so that the server stores the actual playback perspective in the perspective database, and updates the prediction parameters and the corresponding prediction interval ranges in each perspective prediction method based on the perspective database.

[0145] According to one or more embodiments of the present disclosure, [Example 5] provides a perspective prediction method, further comprising:

[0146] Optionally, updating the prediction parameters and corresponding prediction interval ranges in each perspective prediction method based on the perspective database includes:

[0147] Based on the actual playback perspective trajectory information corresponding to the panoramic video in the perspective database, iteratively updating the prediction parameters and corresponding prediction interval ranges in each of the perspective prediction methods offline;

[0148] If the current preset convergence condition is satisfied, stop the iterative update to obtain the perspective prediction method including the current prediction parameters and the corresponding current prediction interval range.

[0149] According to one or more embodiments of the present disclosure, [Example Six] provides a perspective prediction method, further including:

[0150] Optionally, obtaining at least two perspective prediction methods and the prediction interval range corresponding to each of the perspective prediction methods includes:

[0151] In response to a playback trigger operation of a target panoramic video, sending the target video identifier corresponding to the target panoramic video to a server, so that the server determines the target video type corresponding to the target panoramic video based on the target video identifier, and sends at least two perspective prediction methods corresponding to the target video type and the prediction interval range corresponding to each perspective prediction method;

[0152] Receiving at least two perspective prediction methods and the prediction interval range corresponding to each of the perspective prediction methods sent by the server.

[0153] According to one or more embodiments of the present disclosure, [Example Seven] provides a perspective prediction method, further including:

[0154] Optionally, obtaining at least two perspective prediction methods and the prediction interval range corresponding to each of the perspective prediction methods includes:

[0155] In response to a startup operation of a playback terminal, sending the target user identifier corresponding to the video playback user to a server, so that the server determines a target user group based on the target user identifier, and sends at least two perspective prediction methods corresponding to the target user group and the prediction interval range corresponding to each perspective prediction method;

[0156] Receiving at least two perspective prediction methods and the prediction interval range corresponding to each of the perspective prediction methods sent by the server.

[0157] According to one or more embodiments of the present disclosure, [Example Eight] provides a perspective prediction method, further including:

[0158] Optionally, the perspective prediction method includes at least two of the following: general linear regression method, monotonic interval linear regression method, weighted linear regression method, and perspective prediction method based on the current playing perspective.

[0159] According to one or more embodiments of the present disclosure, [Example Nine] provides a perspective prediction device, including:

[0160] A prediction information acquisition module, configured to acquire at least two perspective prediction methods and the prediction interval corresponding to each of the perspective prediction methods;

[0161] A length information acquisition module, configured to acquire the current video cache length in the current target panoramic video being played and the target segment length corresponding to the target video segment to be downloaded currently;

[0162] A target perspective prediction interval determination module, configured to determine the target perspective prediction interval corresponding to the target video segment according to the current video cache length and the target segment length;

[0163] A target perspective prediction interval matching module, configured to match the target perspective prediction interval with the prediction interval to determine the target prediction interval in which the target perspective prediction interval is located;

[0164] A perspective prediction module, configured to perform perspective prediction on the target video segment according to the target perspective prediction method corresponding to the target prediction interval to obtain the target prediction perspective corresponding to the target video segment.

[0165] The above description is only a preferred embodiment of the present disclosure and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosure concept. For example, the technical solutions formed by mutually replacing the above features with the (but not limited to) technical features having similar functions disclosed in the present disclosure.

[0166] In addition, although the operations are depicted in a specific order, this should not be construed as requiring the operations to be performed in the specific order shown or in sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the present disclosure. Certain features described in the context of a single embodiment can also be implemented in combination in a single embodiment. Conversely, the various features described in the context of a single embodiment can also be implemented separately or in any suitable sub-combination in multiple embodiments.

[0167] Although the subject matter has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. On the contrary, the specific features and acts described above are merely example forms for implementing the claims.

Claims

1. A perspective prediction method, characterized in that, it includes: Obtain at least two perspective prediction methods and the corresponding prediction interval ranges for each of the perspective prediction methods; Obtain the current video buffer length in the current target panoramic video being played and the target segment length corresponding to the target video segment to be downloaded currently; Determine the target perspective prediction interval corresponding to the target video segment according to the current video buffer length and the target segment length; Match the target perspective prediction interval with the prediction interval ranges to determine the target prediction interval range where the target perspective prediction interval is located; Perform perspective prediction on the target video segment according to the target perspective prediction method corresponding to the target prediction interval range to obtain the target prediction perspective corresponding to the target video segment; The step of determining the target perspective prediction interval corresponding to the target video segment according to the current video buffer length and the target segment length includes: Determine the intermediate length corresponding to the target segment length; Add the current video buffer length and the intermediate length, and determine the obtained addition result as the target perspective prediction interval corresponding to the target video segment.

2. The perspective prediction method according to claim 1, characterized in that, the step of obtaining at least two perspective prediction methods and the corresponding prediction interval ranges for each of the perspective prediction methods includes: Obtain at least two perspective prediction methods, the prediction method arrangement order, and the prediction interval threshold for distinguishing the prediction performances of every two adjacent perspective prediction methods; Based on the prediction method arrangement order and the prediction interval threshold, determine the prediction interval range corresponding to each of the perspective prediction methods.

3. The perspective prediction method according to claim 1, characterized in that, after obtaining the target prediction perspective corresponding to the target video segment, it further includes: Obtain the actual playing perspective corresponding to the target video segment; Send the actual playing perspective corresponding to the target video segment to the server, so that the server stores the actual playing perspective in the perspective database, and based on the perspective database, update the prediction parameters and the corresponding prediction interval ranges in each perspective prediction method.

4. The perspective prediction method according to claim 3, characterized in that, the step of updating the prediction parameters and the corresponding prediction interval ranges in each perspective prediction method based on the perspective database includes: Based on the actual playing perspective trajectory information corresponding to the panoramic video in the perspective database, perform offline iterative update on the prediction parameters and the corresponding prediction interval ranges in each of the perspective prediction methods; If the current preset convergence condition is satisfied, stop the iterative update to obtain the perspective prediction method including the current prediction parameters and the corresponding current prediction interval range.

5. The perspective prediction method according to claim 1, characterized in that, the step of obtaining at least two perspective prediction methods and the corresponding prediction interval ranges for each of the perspective prediction methods includes: In response to a play trigger operation of a target panoramic video, send a target video identifier corresponding to the target panoramic video to a server, so that the server determines a target video type corresponding to the target panoramic video based on the target video identifier, and sends at least two perspective prediction methods corresponding to the target video type and a prediction interval range corresponding to each perspective prediction method; Receive at least two perspective prediction methods sent by the server and a prediction interval range corresponding to each perspective prediction method.

6. The perspective prediction method according to claim 1, wherein, the obtaining at least two perspective prediction methods and a prediction interval range corresponding to each perspective prediction method includes: In response to a start operation of a playback end, send a target user identifier corresponding to a video playback user to a server, so that the server determines a target user group based on the target user identifier, and sends at least two perspective prediction methods corresponding to the target user group and a prediction interval range corresponding to each perspective prediction method; Receive at least two perspective prediction methods sent by the server and a prediction interval range corresponding to each perspective prediction method.

7. The perspective prediction method according to any one of claims 1-6, wherein, the perspective prediction methods include at least two of: a general linear regression method, a monotonic interval linear regression method, a weighted linear regression method, and a perspective prediction method based on a current playback perspective.

8. A perspective prediction device, wherein, it includes: a prediction information acquisition module, configured to acquire at least two perspective prediction methods and a prediction interval range corresponding to each perspective prediction method; a length information acquisition module, configured to acquire a current video cache length in a currently played target panoramic video and a target segment length corresponding to a currently to-be-downloaded target video segment; a target perspective prediction interval determination module, configured to determine a target perspective prediction interval corresponding to the target video segment according to the current video cache length and the target segment length; a target perspective prediction interval matching module, configured to match the target perspective prediction interval with the prediction interval range to determine a target prediction interval range where the target perspective prediction interval is located; a perspective prediction module, configured to perform perspective prediction on the target video segment according to a target perspective prediction method corresponding to the target prediction interval range to obtain a target prediction perspective corresponding to the target video segment; the target perspective prediction interval determination module is specifically configured to: determine an intermediate length corresponding to the target segment length; add the current video cache length and the intermediate length, and determine the obtained addition result as the target perspective prediction interval corresponding to the target video segment.

9. An electronic device, wherein, the electronic device includes: one or more processors; a storage device, configured to store one or more programs, when the one or more programs are executed by the one or more processors, enabling the one or more processors to implement the perspective prediction method according to any one of claims 1-7.

10. A storage medium containing computer-executable instructions, It is characterized in that when the computer-executable instructions are executed by a computer processor, they are used to execute the perspective prediction method described in any one of claims 1-7.

Images