Video processing method and device, computer device and readable storage medium

Abstract

The application relates to a video processing method and device, computer equipment and a readable storage medium. The method comprises the following steps: acquiring gyroscope data corresponding to a panoramic video, determining the duration of the dynamic posture of a video subject in the panoramic video according to the gyroscope data, performing perspective conversion on the gyroscope data, and generating a special video according to the gyroscope data after perspective conversion and the duration of the dynamic posture of the video subject. The above method can generate a special video according to the gyroscope data of the panoramic video, so that the video processing method can be widely applied in the field of video processing, and the process can be realized through an automatic process, the realization process does not need manual participation, and the generation speed of the special video can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to a video processing method and device, computer equipment and readable storage medium. BACKGROUND

[0002] With the development of new media technology, it has gradually become a new way for people to record life, spread information and entertain and relax by shooting a video through the video function of a smart device. The type of the video shot is not limited to panoramic video, ordinary video, etc.

[0003] Taking a shot panoramic video as an example, in actual application, the original shot panoramic video needs to be processed to obtain a video with certain special effects. For example, a strange species video is a video in which a target in the original shot panoramic video is displayed in the form of an attacking giant.

[0004] However, the related art lacks a method for processing an original video to generate a strange species video.

[0005] It should be noted that the information disclosed in the above background section is only used to strengthen the understanding of the background of the present application, and therefore can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0006] Therefore, it is necessary to provide a video processing method, device, computer equipment and readable storage medium to generate a strange species video in view of the above technical problems.

[0007] In a first aspect, the present application provides a video processing method, comprising:

[0008] obtaining gyroscope data corresponding to a panoramic video, the panoramic video comprising a dynamic posture of a video subject;

[0009] determining a duration of the dynamic posture of the video subject in the panoramic video according to the gyroscope data;

[0010] performing perspective conversion on the gyroscope data, and processing the panoramic video to generate a strange species video according to the perspective converted gyroscope data and the duration of the dynamic posture of the video subject; the strange species video representing a video in which the dynamic posture of the video subject meets certain requirements.

[0011] In one embodiment, determining the duration of the dynamic posture of the video subject in the panoramic video according to the gyroscope data comprises:

[0012] obtaining target gyroscope data on a direction axis with the maximum acceleration from the gyroscope data;

[0013] The target gyroscope data is subjected to frequency domain transformation to obtain a transformation frequency and a transformation amplitude;

[0014] According to the transformation frequency and the transformation amplitude, a duration of a dynamic posture of the video subject in the panoramic video is determined.

[0015] In one of the embodiments, the duration of the dynamic posture includes a start time and an end time of the dynamic posture; and according to the transformation frequency and the transformation amplitude, the duration of the dynamic posture of the video subject in the panoramic video is determined, including:

[0016] If the transformation frequency is greater than a preset frequency threshold and the transformation amplitude is greater than a preset amplitude threshold, then according to a collection time of the target gyroscope data, the start time and the end time of the dynamic posture of the video subject in the panoramic video are determined.

[0017] In one of the embodiments, the perspective conversion of the gyroscope data includes:

[0018] The Euler angles of the shooting device corresponding to the panoramic video are obtained according to the gyroscope data;

[0019] The gyroscope data is subjected to perspective conversion according to the Euler angles of the shooting device and a target shooting perspective; the target shooting perspective includes a perspective toward an edge position of a lens of the shooting device, and the shooting device is the shooting device corresponding to the collection of the panoramic video.

[0020] In one of the embodiments, the perspective conversion of the gyroscope data according to the Euler angles of the shooting device and the target shooting perspective includes:

[0021] The initial shooting perspective of the panoramic video is determined according to the gyroscope data;

[0022] The initial shooting perspective of the panoramic video is adjusted to the target shooting perspective through the Euler angles to obtain adjustment perspective information;

[0023] The gyroscope data after the perspective conversion is determined according to the adjustment perspective information and the gyroscope data.

[0024] In one of the embodiments, the panoramic video is processed to generate a special video according to the gyroscope data after the perspective conversion and the duration of the dynamic posture of the video subject, including:

[0025] The video data of the panoramic video and the gyroscope data after the perspective conversion are subjected to fusion processing to obtain an initial special video;

[0026] The special video is generated according to the initial special video and the duration of the dynamic posture of the video subject.

[0027] In one of the embodiments, the generating of the odd-line species video includes:

[0028] The initial odd-line species video is cropped according to the duration of the dynamic posture of the video subject, to obtain the odd-line species video.

[0029] In one of the embodiments, the generating of the odd-line species video includes:

[0030] The candidate odd-line species video is generated by adding preset audio data with the same duration as the initial odd-line species video to the initial odd-line species video.

[0031] The candidate odd-line species video is cropped according to the duration of the dynamic posture of the video subject, to obtain the odd-line species video.

[0032] In one of the embodiments, the generating of the odd-line species video includes:

[0033] If the video subject has a dynamic posture in the candidate odd-line species video, the cropping time period is determined according to the duration of the dynamic posture of the video subject, and the candidate odd-line species video is cropped according to the cropping time period, to obtain the odd-line species video.

[0034] In one of the embodiments, the dynamic posture includes at least one of a running action and a fast walking action.

[0035] In a second aspect, the present application provides a video processing device, which includes:

[0036] The data acquisition module is configured to acquire gyroscope data corresponding to the panoramic video; the panoramic video includes a dynamic posture of a video subject.

[0037] The determination module is configured to determine the duration of the dynamic posture of the video subject in the panoramic video according to the gyroscope data.

[0038] The video processing module is configured to perform perspective conversion on the gyroscope data, and perform processing on the panoramic video according to the gyroscope data after perspective conversion and the duration of the dynamic posture of the video subject, to generate an odd-line species video; the odd-line species video represents a video in which the dynamic posture of the video subject meets a specific requirement.

[0039] In a third aspect, the present application further provides a shooting device, which includes a lens module, a gyroscope and a computer device.

[0040] The lens module is configured to acquire a panoramic video.

[0041] gyroscope, configured to acquire gyroscope data corresponding to the panoramic video;

[0042] a computer device, configured to execute the steps of any of the embodiments of the first aspect.

[0043] In a fourth aspect, the embodiments of the present application further provide a computer device, which comprises a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the following steps:

[0044] acquiring gyroscope data corresponding to the panoramic video, the panoramic video comprising a dynamic posture of a video subject;

[0045] determining a duration of the dynamic posture of the video subject in the panoramic video according to the gyroscope data;

[0046] performing perspective conversion on the gyroscope data, and processing the panoramic video according to the perspective-converted gyroscope data and the duration of the dynamic posture of the video subject to generate a special video; the special video represents a video in which the dynamic posture of the video subject meets a specific requirement.

[0047] In a fifth aspect, the embodiments of the present application further provide a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the following steps:

[0048] acquiring gyroscope data corresponding to the panoramic video, the panoramic video comprising a dynamic posture of a video subject;

[0049] determining a duration of the dynamic posture of the video subject in the panoramic video according to the gyroscope data;

[0050] performing perspective conversion on the gyroscope data, and processing the panoramic video according to the perspective-converted gyroscope data and the duration of the dynamic posture of the video subject to generate a special video; the special video represents a video in which the dynamic posture of the video subject meets a specific requirement.

[0051] In a sixth aspect, the embodiments of the present application further provide a computer program product, which comprises a computer program. The computer program is executed by a processor to implement the following steps:

[0052] acquiring gyroscope data corresponding to the panoramic video, the panoramic video comprising a dynamic posture of a video subject;

[0053] determining a duration of the dynamic posture of the video subject in the panoramic video according to the gyroscope data;

[0054] The gyro data is perspective converted, and the panoramic video is processed to generate a special video according to the perspective converted gyro data and the duration of the dynamic posture of the video subject.

[0055] The video processing method, device, computer device and readable storage medium provided by the embodiments of the present application include: acquiring gyro data corresponding to a panoramic video, determining the duration of a dynamic posture of a video subject in the panoramic video according to the gyro data, perspective converting the gyro data, and processing the panoramic video to generate a special video according to the perspective converted gyro data and the duration of the dynamic posture of the video subject. The above method can automatically edit and generate a special video according to the gyro data of the panoramic video, without manual editing by the user, thereby improving the user experience. BRIEF DESCRIPTION OF DRAWINGS

[0056] Figure 1 An application environment diagram of the video processing method in one embodiment;

[0057] Figure 2 A flowchart of the video processing method in one embodiment;

[0058] Figure 3 A display diagram of a video frame in the special video in one embodiment;

[0059] Figure 4 A flowchart of the video processing method in another embodiment;

[0060] Figure 5 A flowchart of the video processing method in another embodiment;

[0061] Figure 6 A side structural diagram of a shooting device in one embodiment;

[0062] Figure 7 A flowchart of the video processing method in another embodiment;

[0063] Figure 8 A flowchart of the video processing method in another embodiment;

[0064] Figure 9 A flowchart of the video processing method in another embodiment;

[0065] Figure 10 A structural block diagram of a video processing device in one embodiment;

[0066] Figure 11 An internal structural diagram of a computer device in one embodiment. DETAILED DESCRIPTION

[0067] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0068] In the field of new media, taking a video through the video function of a smart device has gradually become a new way for people to record life, spread information and entertain themselves. However, due to different application requirements, the original video needs to be processed to generate a video with certain special effects. For example, the original video is processed to generate a video with special lines. However, there is a lack of a method for processing the original video to generate a video with special lines in the related art. Based on this, the embodiments of the present application provide a video processing method which can process the original video to generate a video with special lines.

[0069] The video processing method provided by the embodiments of the present application can be applied to Figure 1 As shown in the video processing system, the video processing system includes a shooting device and a computer device peripheral to the shooting device, and the shooting device and the computer device are in communication connection. The communication mode can be Bluetooth, Wi-Fi, mobile network connection, etc. Optionally, the shooting device can be a mobile phone or a camera capable of shooting panoramic video. The camera can be a panoramic camera with a 360-degree field of view. The panoramic camera is a panoramic camera capable of generating a spherical panoramic picture or a spherical panoramic video. The computer device can be implemented by an independent server or a server cluster composed of multiple servers. It can also be, but is not limited to, various personal computers, notebook computers, electronic mobile phones and tablet computers. The specific form of the shooting device and the computer device is not limited in the embodiment. Wherein, Figure 1 is a schematic diagram of the computer device as a server.

[0070] In actual application, in order to simply fix the shooting device, the shooter can fix the shooting device through the fixing part of the shooting device. The fixing part can be understood as a holding part or a biting part. The holding part can be a part of the shooting device for the shooter to hold the shooting device with his hand to fix the shooting device. The biting part can be a part of the shooting device for the shooter to bite the shooting device with his mouth to fix the shooting device. Optionally, multiple cameras can be provided on the shooting device. However, in the embodiments of the present application, two cameras are provided on the shooting device, and each camera is a fisheye lens. Correspondingly, the shooting device can send the collected panoramic video to the computer device connected thereto, and the computer device receives the panoramic video and processes the panoramic video to generate a video with special lines.

[0071] Furthermore, the video processing method provided in this application embodiment can be applied to a shooting device that has a built-in computer. The shooting device acquires panoramic video with the photographer as the main subject and then sends the panoramic video to the computer, which processes the panoramic video to generate a different type of video. The following embodiments will specifically describe the specific process of the video processing method, using a computer as the executing entity.

[0072] like Figure 2 The diagram shown is a flowchart illustrating a video processing method provided in an embodiment of this application. The method may include the following steps:

[0073] S100. Obtain the gyroscope data corresponding to the panoramic video. The panoramic video includes the dynamic pose of the main subject.

[0074] When shooting panoramic video, the gyroscope inside the shooting device can simultaneously collect gyroscope data and store both the collected gyroscope data and panoramic video data into the video file.

[0075] In one embodiment, the panoramic video is a spherical panoramic video. The shooting device can send the panoramic video data in the video file and the gyroscope data of the shooting device that acquired the panoramic video to the computer device. Naturally, the computer device receives the panoramic video data and the corresponding gyroscope data sent by the shooting device.

[0076] It should be noted that the panoramic video data received by the computer device and the corresponding gyroscope data can be data collected in real time by the shooting device, or data collected by the shooting device within a historical time period. Furthermore, the total duration of the panoramic video can be any length, which is not limited in this embodiment of the application.

[0077] S200. Based on the gyroscope data, determine the duration of the dynamic posture of the main subject in the panoramic video.

[0078] Specifically, the computer device can classify the gyroscope data, determine the gyroscope data corresponding to the dynamic posture of the main subject in the panoramic video based on the classification results, and then determine the duration of the dynamic posture of the main subject in the panoramic video based on the gyroscope data corresponding to the dynamic posture of the main subject in the panoramic video.

[0079] Additionally, the computer device can pre-train an algorithm model, and then input the gyroscope data corresponding to the panoramic video into the algorithm model. The algorithm model outputs the duration of the dynamic pose of the main subject in the panoramic video. In the embodiments of this application, the dynamic pose includes at least one of the main subject's running motion and brisk walking motion.

[0080] S300, perspective conversion is performed on the gyroscope data, and the panoramic video is processed to generate a special video according to the perspective-converted gyroscope data and the duration of the dynamic posture of the video subject.

[0081] In the embodiments of the present application, in order to generate a special video, the assumed shooting perspective of the shooting device corresponding to each video frame in the special video needs to be set to the same shooting perspective, i.e., the target shooting perspective. Therefore, perspective conversion needs to be performed on the gyroscope data.

[0082] Specifically, the computer device can perform at least one of angle conversion, rotation, and translation on the gyroscope data to achieve perspective conversion of the gyroscope data. Further, the panoramic video is processed to generate a special video according to the perspective-converted gyroscope data and the duration of the dynamic posture of the video subject.

[0083] In an implementation manner, a special video generation network model can be pre-trained, and then the perspective-converted gyroscope data and the duration of the dynamic posture of the video subject are input into the special video generation network model. The special video generation network model outputs a special video corresponding to the panoramic video. Optionally, the special video generation network model can be composed of at least one of a convolutional neural network model, a fully connected neural network model, a recurrent neural network model, and a generative adversarial network model.

[0084] In another implementation manner, at least one of target detection, target editing, image cropping, and data fusion can be performed on each video frame in the panoramic video according to the perspective-converted gyroscope data and the duration of the dynamic posture of the video subject, to generate a special video carrying audio data or not carrying audio data.

[0085] The special video indicates a video in which the target dynamic posture of the video subject meets a specific requirement. The video subject can be a shooter of a fixed shooting device. The specific requirement can be understood as a requirement that the video subject is displayed in the form of a special giant in the video. Figure 3 As shown in FIG. 5, a special video is shown, Figure 3 The character in the video frame is the video subject, which is displayed in the form of a special giant.

[0086] The technical solution in the embodiment of the application obtains gyroscope data corresponding to the panoramic video, determines the duration of the dynamic posture of the video subject in the panoramic video according to the gyroscope data, performs perspective conversion on the gyroscope data, and generates the special video according to the perspective-converted gyroscope data and the duration of the dynamic posture of the video subject. The above method can determine the duration of the dynamic posture of the video subject in the panoramic video according to the gyroscope data of the panoramic video, perform perspective conversion on the gyroscope data, and then generate the special video according to the perspective-converted gyroscope data and the duration of the dynamic posture of the video subject, so that the video processing method can be widely applied in the field of video processing. The technical solution in the embodiment of the application simplifies the shooting process of the user, can automatically filter the dynamic posture segment in the video subject, and automatically edit the special video, thereby saving the time of manual editing of the user and improving the shooting success rate.

[0087] In actual application, the duration of the special video can be determined according to the duration of the target dynamic posture of the video subject in the panoramic video. The process of how to determine the duration of the dynamic posture of the video subject in the panoramic video is described below. In an embodiment, as shown in FIG. 2, the step of determining the duration of the dynamic posture of the video subject in the panoramic video according to the gyroscope data in S200 can be implemented in the following manner. Figure 4

[0088] S210, obtaining target gyroscope data on the direction axis with the maximum acceleration from the gyroscope data.

[0089] In the embodiment of the application, the gyroscope arranged in the shooting device can include a three-axis acceleration sensor and a three-axis angular velocity sensor. Naturally, the above gyroscope data can include gyroscope data on three direction axes, that is, acceleration of the shooting device on three direction axes and angular velocity of the shooting device on three direction axes.

[0090] The computer device can perform at least one of classification, feature extraction, and feature comparison on the gyroscope data of the shooting device to filter all target gyroscope data on the direction axis with the maximum acceleration from the gyroscope data of the shooting device.

[0091] In addition, the computer device can also pre-train a filtering network model, and then input the gyroscope data of the shooting device into the filtering network model. The filtering network model outputs all target gyroscope data after filtering the gyroscope data on the direction axis with the maximum acceleration from the gyroscope data of the shooting device. Optionally, the filtering network model can be composed of at least one of a residual network model, a self-attention network model, a dense connection network model, and a deep neural network model.

[0092] ​S220, performing frequency domain transformation on the target gyroscope data to obtain a transformed frequency and a transformed amplitude.

[0093] In actual application, for any target gyroscope data, the manner of performing frequency domain transformation on the target gyroscope data can be Fourier transformation processing on the target gyroscope data to obtain the transformed frequency and the transformed amplitude corresponding to the target gyroscope data.

[0094] S230, determining the duration of the dynamic posture of the video subject according to the transformed frequency and the transformed amplitude.

[0095] The computer device can analyze and process the transformed frequency and the transformed amplitude respectively, and determine the duration of the dynamic posture of the video subject in the candidate video according to the analysis result.

[0096] In one embodiment, the duration of the dynamic posture includes a start time of the dynamic posture and an end time of the dynamic posture; and the step of determining the duration of the dynamic posture of the video subject according to the transformed frequency and the transformed amplitude in S230 can include: if the transformed frequency is greater than a preset frequency threshold and the transformed amplitude is greater than a preset amplitude threshold, determining the start time and the end time of the dynamic posture of the video subject according to the collection time of the target gyroscope data.

[0097] In the embodiments of the present application, the computer device can determine whether the transformed frequency corresponding to the target gyroscope data is greater than a preset frequency threshold and whether the transformed amplitude corresponding to the target gyroscope data is greater than a preset amplitude threshold; if it is determined that the transformed frequency is greater than the preset frequency threshold and the transformed amplitude is greater than the preset amplitude threshold, it indicates that the target gyroscope data is the gyroscope data corresponding to the dynamic posture of the video subject, and then the start time and the end time of each dynamic posture of the video subject in the candidate video are determined according to the collection time of all target gyroscope data corresponding to the dynamic posture of the video subject.

[0098] In actual application, the gyroscope data can carry a data collection identifier for determining the collection time of the gyroscope data. Alternatively, the preset frequency threshold and the preset amplitude threshold can be user-defined or determined according to historical experience values.

[0099] The technical solution in the embodiments of the present application obtains the target gyroscope data on the direction axis with the maximum acceleration from the gyroscope data, performs frequency domain transformation on the target gyroscope data to obtain the transformed frequency and the transformed amplitude, and determines the duration of the dynamic posture of the video subject according to the transformed frequency and the transformed amplitude. The method can first obtain the target gyroscope data on the direction axis with the maximum acceleration, and then determine the duration of the dynamic posture of the video subject through the target gyroscope data, so that the accuracy of the duration of the dynamic posture of the video subject obtained is higher.

[0100] The process of converting the perspective of the gyroscope data is described below. In an embodiment, as shown in FIG. 3, the step of converting the perspective of the gyroscope data in S300 can be implemented in the following manner: Figure 5

[0101] S310, obtaining the Euler angle of the shooting device corresponding to the panoramic video according to the gyroscope data.

[0102] In an embodiment, the manner of obtaining the Euler angle of the shooting device corresponding to the panoramic video according to the gyroscope data can be processing the gyroscope data by an Euler angle solving method to obtain the Euler angle of the shooting device. Alternatively, the Euler angle can include a roll angle, a pitch angle, and a yaw angle.

[0103] In another embodiment, the manner of obtaining the Euler angle of the shooting device corresponding to the panoramic video according to the gyroscope data can also be pre-training an Euler angle calculation model, and then inputting the gyroscope data into the Euler angle calculation model, and the Euler angle calculation model outputs the Euler angle of the shooting device.

[0104] Alternatively, the above-mentioned Euler angle calculation model can be composed of at least one of a convolutional neural network model, a fully connected neural network model, a recurrent neural network model, a deep belief network model, a deep autoencoder, and a generative adversarial network model.

[0105] S320, converting the perspective of the gyroscope data according to the Euler angle of the shooting device and a target shooting perspective; the target shooting perspective includes a perspective towards the lens edge position of the shooting device, and the shooting device is the shooting device corresponding to the panoramic video.

[0106] In the embodiment of the present application, each gyroscope data has a corresponding Euler angle, so that the perspective of each gyroscope data can be converted by each Euler angle.

[0107] In the embodiment of the present application, the perspective of the lens edge position includes the perspective of the shooting device towards the fixed part. As shown in FIG. 4, the side structure of the shooting device is shown, Figure 6 Figure 6 The two ellipses on the top of the shooting device represent two cameras of the shooting device, and the area covered by the vertical line below the two cameras can be referred to as the perspective corresponding to the lens edge position.

[0108] ​​The technical solutions in the embodiments of the present application obtain the Euler angle of the panoramic video shooting device according to the gyroscope data, and perform perspective conversion on the gyroscope data according to the Euler angle of the shooting device and the target shooting perspective. Since the distortion of the picture under different perspectives of the camera is different, the distortion of the picture under the perspective of the camera facing the fixed part can make the person in the video main body present the posture of the giant race, so that in the embodiments of the present application, the giant race video can be automatically generated by changing the perspective of the spherical panoramic video.

[0109] The process of performing perspective conversion on the gyroscope data according to the Euler angle of the shooting device and the target shooting perspective is described below. In an embodiment, as shown in Figure 7 the steps in S320 can be implemented in the following manner:

[0110] S321, determining the initial shooting perspective of the panoramic video according to the gyroscope data.

[0111] Specifically, the computer device can use a sensor-based pose estimation algorithm, an image-based pose estimation algorithm, or a fusion pose estimation algorithm based on sensors and images to analyze and process the gyroscope data to obtain the shooting perspective corresponding to the shooting device for collecting the panoramic video, that is, the initial shooting perspective.

[0112] S322, adjusting the initial shooting perspective of the shooting device to the target shooting perspective through the Euler angle to obtain adjustment perspective information.

[0113] Specifically, the computer device can adjust the initial shooting perspective of the shooting device based on the Euler angle to adjust the initial shooting perspective of the shooting device to the target shooting perspective, and can obtain the adjustment perspective information according to the Euler angle and the initial shooting perspective of the shooting device. Optionally, the adjustment perspective information can include an adjustment perspective difference and an adjustment path of the adjustment perspective, and the adjustment path of the adjustment perspective can be a straight adjustment path, a polyline adjustment path, or a curve adjustment path, etc.

[0114] It should be noted here that if the adjustment path of the adjustment perspective is a straight adjustment path, the adjustment perspective difference can be obtained by subtracting the initial shooting perspective from the target shooting perspective; if the adjustment path of the adjustment perspective is a polyline adjustment path, the adjustment perspective difference can be determined according to the length of each polyline segment in the polyline adjustment path, the coordinates of the endpoints of the polyline segment, the target shooting perspective, and the initial shooting perspective; if the adjustment path of the adjustment perspective is a curve adjustment path, the adjustment perspective difference can be determined according to the curvature of the curve in the curve adjustment path, the length of the curve, the coordinates of the endpoints of the curve, the target shooting perspective, and the initial shooting perspective.

[0115] S323, determining the gyroscope data after perspective conversion according to the adjustment perspective information and the gyroscope data.

[0116] The computer device can process the gyroscope data according to the adjusted view angle information to obtain the gyroscope data after view angle conversion.

[0117] For example, the manner of processing the gyroscope data according to the adjusted view angle information can be searching for matched adjusted view angle information and gyroscope data in a mapping relationship, and then determining the view angle converted data corresponding to the matched adjusted view angle information and gyroscope data in the mapping relationship as the gyroscope data after view angle conversion. Optionally, the mapping relationship can include the adjusted view angle information, the gyroscope data, the gyroscope data after view angle conversion, and the corresponding relationship among the three.

[0118] For another example, the manner of processing the gyroscope data according to the adjusted view angle information can also be performing an arithmetic operation on the adjusted view angle difference in the adjusted view angle information and the gyroscope data to obtain the gyroscope data after view angle conversion. Optionally, the above arithmetic operation can include at least one of addition operation, subtraction operation, multiplication operation, division operation, exponential operation and logarithmic operation.

[0119] The technical solution in the embodiment of the present application determines the initial shooting view angle of the panoramic video according to the gyroscope data, adjusts the initial shooting view angle of the shooting device into a target shooting view angle through Euler angles to obtain the adjusted view angle information, and determines the gyroscope data after view angle conversion according to the adjusted view angle information and the gyroscope data. The above method can process the gyroscope data to obtain the gyroscope data after view angle conversion through a simple processing process, improves the speed and efficiency of the gyroscope data view angle conversion, and uses the obtained gyroscope data after view angle conversion as the basic information to process the original panoramic video to generate the odd-row video, so that the video processing method can be widely applied in the field of video processing.

[0120] The following describes the process of processing the panoramic video to generate the odd-row video according to the gyroscope data after view angle conversion and the duration of the dynamic posture of the video subject. In an embodiment, as shown in FIG. 3, the steps in the above 300 can be implemented in the following manner: Figure 8

[0121] S330, performing fusion processing on the video data of the panoramic video and the gyroscope data after view angle conversion to obtain an initial odd-row video.

[0122] ​The manner of fusing the video data of the panoramic video and the gyroscope data after the perspective conversion can be that a data fusion network model is pre-trained, and then the video data of the panoramic video and the gyroscope data after the perspective conversion are input into the data fusion network model, and the data fusion network model outputs the initial odd-line video. Alternatively, the data fusion network model can be composed of at least one of a feature pyramid network model, a convolutional neural network model, a spatial pyramid pooling network model, and a recurrent neural network model, and the like, and the embodiments of the present application are not limited in this regard.

[0123] In addition, the manner of fusing the video data of the panoramic video and the gyroscope data after the perspective conversion can also be that a fusion algorithm is used to fuse the video data of the panoramic video and the gyroscope data after the perspective conversion to obtain the initial odd-line video. Alternatively, the fusion algorithm can be a standardization fusion algorithm, a hierarchical fusion algorithm, or a hybrid fusion algorithm, and the like. In the embodiments of the present application, the initial odd-line video can be video data that does not carry audio data.

[0124] S340, generating an odd-line video according to the initial odd-line video and the duration of the dynamic posture of the video subject.

[0125] Specifically, the computer device can fuse the initial odd-line video and the audio data to generate an intermediate odd-line video, and then perform arbitrary paragraph video cropping on the intermediate odd-line video to obtain an odd-line video with the same duration as the duration of the dynamic posture of the video subject.

[0126] In addition, in some scenarios, an odd-line video that does not carry audio data can also be generated, based on which, in an embodiment, the process of generating an odd-line video according to the initial odd-line video and the duration of the dynamic posture of the video subject in S340 can include: cropping the initial odd-line video according to the duration of the dynamic posture of the video subject to obtain the odd-line video.

[0127] Alternatively, the computer device can determine the corresponding start position and end position in the initial odd-line video according to the duration of the dynamic posture of the video subject, then adjust the corresponding end position of the initial odd-line video to obtain a target end position, and then crop the initial odd-line video from the target end position to obtain a paragraph video between the corresponding start position and the target end position in the initial odd-line video, i.e., the odd-line video. Alternatively, the target end position can be located between the corresponding start position and the end position of the initial odd-line video, or can be located after the corresponding end position of the initial odd-line video.

[0128] In some scenarios, audio data can be automatically added to the initial odd-row video without carrying audio data, avoiding the process of manually adding audio data in the related art, which can reduce the complexity of adding audio data to the initial odd-row video and improve the speed of adding audio data to the initial odd-row video. Based on this, in one embodiment, as shown in Figure 9 The step of generating an odd-row video based on the initial odd-row video in S340 can be implemented in the following manner:

[0129] S341, add preset audio data with the same length as the initial odd-row video in the initial odd-row video to generate a candidate odd-row video.

[0130] In actual application, the computer device can perform fusion processing on the preset audio data with the same length as the initial odd-row video and the initial odd-row video to generate a candidate odd-row video.

[0131] Optionally, the preset audio data can be a whole piece of audio data made in advance, in which case the length of the whole piece of audio data is equal to the length of the initial odd-row video, or a part of a whole piece of audio data made in advance, in which case the length of the whole piece of audio data is greater than the length of the initial odd-row video, or audio data obtained by repeatedly superimposing a whole piece of audio data made in advance, in which case the length of the whole piece of audio data is less than the length of the initial odd-row video.

[0132] S342, according to the duration of the dynamic posture of the video subject, the candidate odd-row video is cropped to obtain an odd-row video.

[0133] The computer device can determine the corresponding start position and end position of the candidate odd-row video according to the duration of the dynamic posture of the video subject, then adjust the end position of the candidate odd-row video to obtain an adjusted end position, and then crop the candidate odd-row video from the adjusted end position to obtain the passage video between the corresponding start position and the adjusted end position of the candidate odd-row video, i.e., the odd-row video. Optionally, the adjusted end position can be located between the corresponding start position and end position of the candidate odd-row video, or after the corresponding end position of the candidate odd-row video.

[0134] The technical solutions in the embodiments of the present application fuse the video data of the panoramic video and the gyroscope data after perspective conversion to obtain initial odd-row video, and generate odd-row video based on the initial odd-row video. The above method can use the obtained gyroscope data after perspective conversion as basic information, fuse the video data of the panoramic video and the gyroscope data after perspective conversion to generate initial odd-row video data, and then add corresponding audio data to the initial odd-row video data to obtain odd-row video carrying audio signals. This process does not require manual intervention and can be completed automatically, which not only saves human resources but also improves the generation speed and efficiency of odd-row video. In addition, the method can generate odd-row video, so that the video processing method can be widely applied in the field of video processing.

[0135] The process of cutting the candidate odd-row video to obtain the odd-row video according to the duration of the dynamic posture of the video subject is described below. In an embodiment, the step in S342 can include: if the video subject in the candidate odd-row video has a dynamic posture, determining a cutting time period according to the duration of the dynamic posture of the video subject, and cutting the candidate odd-row video according to the cutting time period to obtain the odd-row video.

[0136] If it is determined that the video subject in the candidate odd-row video has a dynamic posture, the computer device can determine a cutting position, i.e., a cutting time period, according to the duration of the dynamic posture of the video subject, and then cut the candidate odd-row video according to the cutting time period to obtain the odd-row video.

[0137] Optionally, the duration of the dynamic posture of the video subject in the candidate odd-row video can be less than or equal to the total duration of the candidate odd-row video. The dynamic posture of the video subject in the candidate odd-row video can be a continuous dynamic posture or an intermittent dynamic posture, which is not limited in the embodiments of the present application.

[0138] If the video subject in the candidate odd-row video has a continuous dynamic posture, the time period corresponding to the continuous dynamic posture of the video subject in the candidate odd-row video can be adjusted according to the time period corresponding to the candidate odd-row video to obtain a cutting time period. If the video subject in the candidate odd-row video has an intermittent dynamic posture, the time period corresponding to the first dynamic posture of the video subject in the candidate odd-row video can be adjusted according to the time period corresponding to the candidate odd-row video to obtain a cutting time period.

[0139] It should be noted that the clipping time period can be less than or equal to the time period corresponding to the candidate odd species video, greater than, less than, or equal to the time period corresponding to the continuous dynamic posture in the candidate odd species video, of course, greater than, less than, or equal to the time period corresponding to the first dynamic posture in the candidate odd species video.

[0140] The technical solution in the embodiment of the application, when the video subject has a dynamic posture in the candidate odd species video, determines the clipping time period according to the duration of the dynamic posture of the video subject, and clips the candidate odd species video according to the clipping time period to obtain an odd species video with a time length required by actual application. This process is not only simple, but also can delete invalid videos from the candidate odd species video, so that the final obtained odd species video is all valid video, thereby being able to provide effective help for subsequent odd species video application, and also being able to reduce the data amount of subsequent odd species video application processing, and improve the accuracy of subsequent odd species video application processing.

[0141] In an embodiment, the embodiment of the application further provides a video processing method, which comprises the following processes:

[0142] (1) Obtain gyroscope data corresponding to a panoramic video, the panoramic video including a dynamic posture of a video subject; the dynamic posture including at least one of a running action and a fast walking action.

[0143] (2) Obtain target gyroscope data on a direction axis with the largest acceleration from the gyroscope data.

[0144] (3) Perform frequency domain transformation on the target gyroscope data to obtain a transformed frequency and a transformed amplitude.

[0145] (4) If the transformed frequency is greater than a preset frequency threshold and the transformed amplitude is greater than a preset amplitude threshold, determine a start time and an end time of the dynamic posture of the video subject in the panoramic video according to a collection time of the target gyroscope data.

[0146] (5) Obtain Euler angles of a shooting device corresponding to the panoramic video according to the gyroscope data.

[0147] (6) Determine an initial shooting angle of the panoramic video according to the gyroscope data.

[0148] (7) Adjust the initial shooting angle of the panoramic video to a target shooting angle through the Euler angles to obtain adjustment angle information; the target shooting angle including an angle of a lens edge position of the shooting device that collects the panoramic video.

[0149] (8) Determine gyroscope data after angle conversion according to the adjustment angle information and the gyroscope data.

[0150] (9) Fuse the video data of the panoramic video and the gyro data after the perspective conversion to obtain an initial odd-row video.

[0151] (10) Generate an odd-row video according to the initial odd-row video and the duration of the dynamic posture of the video subject; the odd-row video represents a video in which the dynamic posture of the video subject meets a specific requirement.

[0152] The step (10) can be implemented in two ways.

[0153] The first way includes the following steps.

[0154] (11) According to the duration of the dynamic posture of the video subject, the initial odd-row video is cropped to obtain the odd-row video.

[0155] The second way includes the following steps.

[0156] (12) Add preset audio data with the same duration as the initial odd-row video to the initial odd-row video to generate a candidate odd-row video.

[0157] (13) If the video subject has a dynamic posture in the candidate odd-row video, a clipping time period is determined according to the duration of the dynamic posture of the video subject, and the candidate odd-row video is cropped according to the clipping time period to obtain the odd-row video.

[0158] The execution process of (1) to (10) can be found in the description of the above embodiments, and the implementation principle and technical effects are similar, which will not be repeated here.

[0159] It should be understood that, although each step in the flowchart involved in the above embodiments is displayed in sequence according to the arrow, these steps are not necessarily executed in the order indicated by the arrow. Unless otherwise stated herein, the execution of these steps has no strict order limitation, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in the above embodiments can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps.

[0160] Based on the same inventive concept, the embodiments of the present application further provide a video processing apparatus for implementing the video processing method described above. The implementation scheme of the apparatus for solving the problem is similar to the implementation scheme described in the above method, so the specific limitations in one or more video processing apparatus embodiments provided below can refer to the limitations of the video processing method described above, which will not be repeated here.

[0161] In one embodiment, Figure 10 For the structural schematic diagram of the video processing apparatus in an embodiment of the present application, the video processing apparatus provided by the embodiments of the present application can be applied to a computer device. As shown in Figure 10 The video processing apparatus of the embodiments of the present application can include a data acquisition module 11, a determination module 12 and a video processing module 13, wherein:

[0162] The data acquisition module 11 is configured to acquire gyroscope data corresponding to a panoramic video; the panoramic video includes dynamic postures of a video subject;

[0163] The determination module 12 is configured to determine a duration of the dynamic postures of the video subject in the panoramic video according to the gyroscope data;

[0164] The video processing module 13 is configured to perform perspective conversion on the gyroscope data, and perform processing on the panoramic video according to the gyroscope data after perspective conversion and the duration of the dynamic postures of the video subject to generate a special video; the special video represents a video in which the dynamic postures of the video subject meet specific requirements.

[0165] The video processing apparatus provided by the embodiments of the present application can be used to execute the technical solutions in the above-mentioned video processing method embodiments of the present application, and the implementation principles and technical effects are similar, which will not be repeated here.

[0166] In one embodiment, the determination module 12 includes a target data acquisition unit, a frequency domain transformation unit and a determination unit, wherein:

[0167] The target data acquisition unit is configured to acquire target gyroscope data on a direction axis with the maximum acceleration from the gyroscope data;

[0168] The frequency domain transformation unit is configured to perform frequency domain transformation on the target gyroscope data to obtain a transformation frequency and a transformation amplitude;

[0169] The duration determination unit is configured to determine the duration of the dynamic postures of the video subject in the panoramic video according to the transformation frequency and the transformation amplitude;

[0170] The dynamic postures include at least one of a running action and a fast walking action.

[0171] The video processing apparatus provided in the embodiments of the present application can be used to execute the technical solutions in the video processing method embodiments of the present application, and the implementation principles and technical effects are similar, which will not be repeated here.

[0172] In one of the embodiments, the duration of the dynamic posture includes a start time and an end time of the dynamic posture; and the duration determining unit is specifically configured to:

[0173] When the transform frequency is greater than the preset frequency threshold and the transform amplitude is greater than the preset amplitude threshold, the start time and the end time of the dynamic posture of the video subject in the panoramic video are determined according to the collection time of the target gyroscope data.

[0174] The video processing apparatus provided in the embodiments of the present application can be used to execute the technical solutions in the video processing method embodiments of the present application, and the implementation principles and technical effects are similar, which will not be repeated here.

[0175] In one of the embodiments, the video processing module 13 includes an Euler angle obtaining unit and a view angle conversion unit, wherein:

[0176] The Euler angle obtaining unit is configured to obtain the Euler angle of the shooting device corresponding to the panoramic video according to the gyroscope data.

[0177] The view angle conversion unit is configured to perform view angle conversion on the gyroscope data according to the Euler angle of the shooting device and a target shooting view angle; and the target shooting view angle includes the view angle of the lens edge position of the shooting device that collects the panoramic video.

[0178] The video processing apparatus provided in the embodiments of the present application can be used to execute the technical solutions in the video processing method embodiments of the present application, and the implementation principles and technical effects are similar, which will not be repeated here.

[0179] In one of the embodiments, the view angle conversion unit includes an initial view angle determining subunit, a view angle adjusting subunit and a determining subunit, wherein:

[0180] The initial view angle determining subunit is configured to determine an initial shooting view angle of the panoramic video according to the gyroscope data.

[0181] The view angle adjusting subunit is configured to adjust the initial shooting view angle of the panoramic video to a target shooting view angle through the Euler angle to obtain adjustment view angle information.

[0182] The determining subunit is configured to determine the gyroscope data after view angle conversion according to the adjustment view angle information and the gyroscope data.

[0183] The video processing apparatus provided in the embodiments of the present application can be used to execute the technical solutions in the video processing method embodiments of the present application, and the implementation principles and technical effects are similar, which will not be repeated here.

[0184] In one of the embodiments, the video processing module 13 further comprises a fusion processing unit and a video generating unit, wherein:

[0185] The fusion processing unit is configured to perform fusion processing on the video data of the panoramic video and the gyro data converted in the perspective, to obtain an initial odd-row video.

[0186] The video generating unit is configured to generate the odd-row video according to the initial odd-row video and the duration of the dynamic posture of the video subject.

[0187] The video processing device provided by the embodiments of the present application can be used to execute the technical solutions in the video processing method embodiments of the present application, and the implementation principles and technical effects are similar, which will not be repeated here.

[0188] In one of the embodiments, the video generating unit comprises a first generating subunit, and the first generating subunit is specifically configured to:

[0189] According to the duration of the dynamic posture of the video subject, the initial odd-row video is cropped to obtain the odd-row video.

[0190] The video processing device provided by the embodiments of the present application can be used to execute the technical solutions in the video processing method embodiments of the present application, and the implementation principles and technical effects are similar, which will not be repeated here.

[0191] In one of the embodiments, the video generating unit comprises an audio data adding subunit and a cropping subunit, wherein:

[0192] The audio data adding subunit is configured to add preset audio data with the same duration as the initial odd-row video in the initial odd-row video to generate a candidate odd-row video.

[0193] The cropping subunit is configured to crop the candidate odd-row video according to the duration of the dynamic posture of the video subject to obtain the odd-row video.

[0194] The video processing device provided by the embodiments of the present application can be used to execute the technical solutions in the video processing method embodiments of the present application, and the implementation principles and technical effects are similar, which will not be repeated here.

[0195] In one of the embodiments, the cropping subunit is specifically configured to:

[0196] When the video subject has a dynamic posture in the candidate odd-row video, the cropping time period is determined according to the duration of the dynamic posture of the video subject, and the candidate odd-row video is cropped according to the cropping time period to obtain the odd-row video.

[0197] The video processing apparatus provided in the embodiments of the present application can be used to execute the technical solutions in the embodiments of the video processing method provided in the present application, and the implementation principles and technical effects are similar, which will not be repeated here.

[0198] The specific limitations on the video processing apparatus can be referred to the limitations on the video processing method, which will not be repeated here. Each module in the video processing apparatus can be realized by software, hardware and combination thereof in whole or in part. Each module can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to each module.

[0199] In one embodiment, a shooting device is provided, which comprises a lens module, a gyroscope and a computer device; an image acquisition device connected to a holder, wherein the computer device in the image acquisition device comprises the computer device in the above-mentioned embodiments;

[0200] The lens module is configured to acquire a panoramic video.

[0201] The gyroscope is configured to acquire gyroscope data corresponding to the panoramic video.

[0202] The computer device is configured to execute the steps of any one of the embodiments of the video processing method.

[0203] Optionally, the lens module can comprise at least two cameras, but in the embodiments of the present application, the lens module comprises two cameras, and the two cameras are symmetrically arranged.

[0204] It should be noted that each camera in the shooting device can shoot a super wide-angle image or video, and the super wide-angle image or video collected by each camera can be further stitched to obtain the panoramic video collected by the lens module. In actual application, the panoramic video and the gyroscope data can be collected synchronously.

[0205] The video processing system provided in the embodiments of the present application can be used to execute the technical solutions in the embodiments of the video processing method provided in the present application, and the implementation principles and technical effects are similar, which will not be repeated here.

[0206] In one embodiment, a computer device is provided, and the internal structure diagram of the computer device can be as shown in Figure 11As shown in the figure. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide processing capability. The memory of the computer device includes a non-volatile storage medium, an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store video files. The network interface of the computer device is used to communicate with the external terminal through the network connection. The computer program is executed by the processor to implement a video processing method.

[0207] Those skilled in the art can understand that, Figure 11 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0208] In one embodiment, a computer device is also provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the following steps:

[0209] Obtaining gyroscope data corresponding to the panoramic video, the panoramic video including dynamic gestures of a video subject;

[0210] According to the gyroscope data, determining the duration of the dynamic gestures of the video subject in the panoramic video;

[0211] Converting the perspective of the gyroscope data, and processing the panoramic video according to the gyroscope data after perspective conversion and the duration of the dynamic gestures of the video subject to generate a special video; the special video represents a video in which the dynamic gestures of the video subject meet specific requirements.

[0212] In one embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the following steps:

[0213] Obtaining gyroscope data corresponding to the panoramic video, the panoramic video including dynamic gestures of a video subject;

[0214] According to the gyroscope data, determining the duration of the dynamic gestures of the video subject in the panoramic video;

[0215] Converting the perspective of the gyroscope data, and processing the panoramic video according to the gyroscope data after perspective conversion and the duration of the dynamic gestures of the video subject to generate a special video; the special video represents a video in which the dynamic gestures of the video subject meet specific requirements.

[0216] In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the following steps:

[0217] Obtaining gyro data corresponding to the panoramic video, the panoramic video comprising dynamic postures of a video subject;

[0218] Determining a duration of the dynamic postures of the video subject in the panoramic video according to the gyro data;

[0219] Converting a perspective of the gyro data, and processing the panoramic video according to the gyro data after the perspective conversion and the duration of the dynamic postures of the video subject to generate a special video; the special video represents a video in which the dynamic postures of the video subject meet specific requirements.

[0220] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiments. Any reference to memory, storage, database or other medium used in the embodiments provided by the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0221] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0222] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, some modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of the patent of the present application should be subject to the appended claims.

Claims

1. A video processing method, characterized in that, The method includes: Obtain gyroscope data corresponding to the panoramic video, wherein the panoramic video includes the dynamic posture of the main subject in the video; Based on the gyroscope data, determine the duration of the dynamic posture of the main subject in the panoramic video; Based on the Euler angles of the shooting device corresponding to the panoramic video and the target shooting angle, the gyroscope data is transformed by angle conversion, and an odd-form video is generated based on the initial odd-form video and the duration of the dynamic posture of the video subject. The Euler angles are determined based on the gyroscope data; the target shooting angle includes the angle towards the edge of the lens of the shooting device, and the shooting device is the shooting device corresponding to the panoramic video; the initial odd-form video is obtained by fusing the video data of the panoramic video and the gyroscope data after the angle conversion; the odd-form video represents a video in which the dynamic posture of the main body in the video meets specific requirements.

2. The method according to claim 1, characterized in that, The step of determining the duration of the dynamic pose of the main subject in the panoramic video based on the gyroscope data includes: Obtain the target gyroscope data on the axis of maximum acceleration from the gyroscope data; The target gyroscope data is subjected to frequency domain transformation to obtain the transformed frequency and transformed amplitude; The duration of the dynamic pose of the main subject in the panoramic video is determined based on the transformation frequency and the transformation amplitude.

3. The method according to claim 2, characterized in that, The duration of the dynamic pose includes the start and end times of the dynamic pose; determining the duration of the dynamic pose of the main subject in the panoramic video based on the transformation frequency and the transformation amplitude includes: If the transformation frequency is greater than a preset frequency threshold and the transformation amplitude is greater than a preset amplitude threshold, then the start and end times of the dynamic posture of the video subject in the panoramic video are determined based on the acquisition time of the target gyroscope data.

4. The method according to claim 1, characterized in that, The step of performing perspective conversion on the gyroscope data based on the Euler angles of the shooting device corresponding to the panoramic video and the target shooting angle includes: Based on the gyroscope data, the initial shooting angle of the panoramic video is determined; By using the Euler angles, the initial shooting angle of the panoramic video is adjusted to the target shooting angle to obtain the adjusted angle information; Based on the adjusted viewing angle information and the gyroscope data, the gyroscope data after the viewing angle conversion is determined.

5. The method according to claim 1, characterized in that, The process of generating an odd-form video based on the initial odd-form video and the duration of the dynamic pose of the video subject includes: Based on the duration of the dynamic pose of the main subject in the video, the initial odd-form video is cropped to obtain the odd-form video.

6. The method according to claim 1, characterized in that, The process of generating an odd-form video based on the initial odd-form video and the duration of the dynamic pose of the video subject includes: Add preset audio data of the same duration as the initial odd-form video to the initial odd-form video to generate candidate odd-form videos; Based on the duration of the dynamic pose of the video subject, the candidate odd-form video is cropped to obtain the odd-form video.

7. The method according to claim 6, characterized in that, The step of cropping the candidate odd-form video based on the duration of the dynamic pose of the video subject to obtain the odd-form video includes: If the main subject in the candidate odd-form video has a dynamic pose, the cropping time period is determined based on the duration of the dynamic pose of the main subject, and the candidate odd-form video is cropped according to the cropping time period to obtain the odd-form video.

8. The method according to any one of claims 1-3, characterized in that, The dynamic posture includes at least one of running and brisk walking.

9. A video processing apparatus, characterized in that, The device includes: The data acquisition module is used to acquire gyroscope data corresponding to the panoramic video; the panoramic video includes the dynamic posture of the main subject in the video. The determination module is used to determine the duration of the dynamic posture of the main subject in the panoramic video based on the gyroscope data. The video processing module is used to perform perspective conversion on the gyroscope data based on the Euler angles of the shooting device corresponding to the panoramic video and the target shooting angle, and to generate an odd-form video based on the initial odd-form video and the duration of the dynamic posture of the video subject; the Euler angles are determined based on the gyroscope data; the target shooting angle includes the angle towards the edge of the lens of the shooting device, and the shooting device is the shooting device corresponding to the panoramic video; the initial odd-form video is obtained by fusing the video data of the panoramic video and the perspective-converted gyroscope data; the odd-form video represents a video in which the dynamic posture of the video subject meets specific requirements.

10. A shooting device, characterized in that, The shooting equipment includes: a lens module, a gyroscope, and a computer device; The lens module is used to capture panoramic video; The gyroscope is used to collect gyroscope data corresponding to the panoramic video; The computer device is used to perform the steps of the method according to any one of claims 1-8.

11. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-8.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-8.

Images