Video editing method and apparatus, electronic device, and medium

Abstract

The present disclosure relates to a video editing method and apparatus, an electronic device, and a medium. The video editing method comprises: displaying a current video frame, and, in response to a selection of a user, determining a first target point of interest; on the basis of the first target point of interest, performing playback, and recording a playback path; on the basis of the playing path, editing a video. The present disclosure permits a user to participate in the editing process on the basis of personal preferences, and the user can determine the first target point of interest by means of interaction with an electronic device, and perform playback. In addition, the playback path is recorded, and the process of performing playback on the basis of the first target point of interest allows the user to directly see a corresponding editing effect, i.e., what you see is what you get, so that a generated edited video contains the first target point of interest that the user cares about, and the efficiency of the video editing process; the use experience of the user is improved, and the individualization of the edited video and the satisfaction regarding same are also improved.

Description

Video editing method, device, electronic device and medium Technical Field

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

[0002] Video editing technology is widely used today. Users can use the automatic editing function in video editing applications to edit raw videos. While this function is simple to use, the resulting edited videos may not always reflect the segments the user is interested in. This can result in missing segments that the user considers interesting, preventing the user from obtaining the desired video content.

[0003] Summary of the Invention

[0004] To overcome the problems existing in the related art, the present disclosure provides a video editing method, device, electronic device and medium.

[0005] According to a first aspect of an embodiment of the present disclosure, a video editing method is provided, comprising:

[0006] displaying a current video frame and determining a first target point of interest in response to a user selection;

[0007] Playing based on the first target point of interest, and recording the playback path;

[0008] The video is edited based on the playback path.

[0009] In some exemplary embodiments of the present disclosure, displaying the current video frame and determining the first target point of interest in response to a user's selection includes:

[0010] At least one candidate interest point of the current video frame is displayed, and a first target interest point is determined in response to a user's selection.

[0011] In some exemplary embodiments of the present disclosure, playing based on the first target point of interest and recording the playback path includes:

[0012] Tracking playback is performed based on the first target point of interest, and the playback path is recorded.

[0013] In some exemplary embodiments of the present disclosure, playing based on the first target point of interest and recording the playback path includes:

[0014] Determine a playback sequence corresponding to the first target point of interest, play based on the playback sequence, and record a playback path.

[0015] In some exemplary embodiments of the present disclosure, the first target point of interest is an interest point determined from the at least one candidate point of interest or an interest point determined from other non-candidate points of interest in the current video frame.

[0016] In some exemplary embodiments of the present disclosure, the method further includes:

[0017] Determining a key video frame in a subsequent video frame of the current video frame, and determining a second target interest point in the key video frame;

[0018] The playing based on the first target point of interest and recording the playing path includes:

[0019] Play the video based on the first target point of interest and the second target point of interest, and record the playback path.

[0020] In some exemplary embodiments of the present disclosure, the playing based on the first target point of interest and the second target point of interest and recording the playing path includes:

[0021] A viewing angle of an intermediate video frame between the current video frame and the key video frame is determined, where the viewing angle is used to gradually transition from the first target interest point to the second target interest point.

[0022] In some exemplary embodiments of the present disclosure, the video editing based on the first target point of interest includes:

[0023] Perform tracking playback based on the first target point of interest and record the playback path;

[0024] The video is edited based on the playback path.

[0025] In some exemplary embodiments of the present disclosure, performing tracking playback based on the first target point of interest and recording the playback path includes:

[0026] determining a viewing angle of each video frame according to the first target point of interest, wherein the viewing angle is used to gradually transition the first target point of interest to the center of a viewing screen and maintain the first target point of interest at the center of the viewing screen;

[0027] Play each video frame based on the determined viewing angle and record the playback path.

[0028] In some exemplary embodiments of the present disclosure, the method further includes:

[0029] A forward path identifier corresponding to each candidate interest point is displayed on the current video frame.

[0030] In some exemplary embodiments of the present disclosure, the forward path identifier is used to indicate the moving direction corresponding to the point of interest at the next moment.

[0031] In some exemplary embodiments of the present disclosure, the forward path identifier is used to indicate the subsequent moving direction of the point of interest.

[0032] In some exemplary embodiments of the present disclosure, displaying a forward path identifier corresponding to each candidate point of interest includes:

[0033] Based on the interest point tracking model, identifying the motion trajectory of each candidate interest point;

[0034] Based on the motion trajectory, displaying a forward path identifier corresponding to each candidate point of interest;

[0035] Wherein, the interest point tracking model uses a target tracking algorithm.

[0036] In some exemplary embodiments of the present disclosure, performing tracking playback based on the first target point of interest and recording the playback path includes:

[0037] Tracking playback is performed based on the moving direction indicated by the forward path identifier corresponding to the first target point of interest, and the playback path is recorded.

[0038] In some exemplary embodiments of the present disclosure, performing tracking playback based on the first target point of interest and recording the playback path includes:

[0039] During the tracking playback based on the first target interest point, upon receiving a pause or rewind instruction from the user, pausing or rewinding to a video frame selected by the user, using the paused or rewound video frame as a new current video frame, and displaying the candidate interest point on the current video frame;

[0040] In response to the user selecting a second target point of interest from the candidate points of interest for tracking playback, a playback path of each video frame is recorded.

[0041] In some exemplary embodiments of the present disclosure, performing tracking playback based on the first target point of interest and recording the playback path includes:

[0042] During the tracking playback based on the first target point of interest, upon receiving a pause or rewind instruction from the user, pausing or rewinding to a video frame selected by the user, using the paused or rewound video frame as a new current video frame, and displaying the candidate point of interest and the corresponding forward path identifier on the current video frame;

[0043] In response to a second target point of interest selected by the user from the candidate points of interest, tracking playback is performed based on the moving direction indicated by the corresponding forward path identifier, and the playback path of each video frame is recorded.

[0044] In some exemplary embodiments of the present disclosure, the method further includes:

[0045] In response to a user's preview instruction for a candidate point of interest, a preview playback screen of the candidate point of interest is displayed, and the preview playback screen can be used to preview a motion trajectory screen of the candidate point of interest within a preset time period.

[0046] In some exemplary embodiments of the present disclosure, the method further includes:

[0047] In response to a frame skipping instruction from the user, the video stops at a new video frame selected by the user, takes the new video frame as the current video frame, and displays candidate interest points in the current video frame.

[0048] In some exemplary embodiments of the present disclosure, the method further includes:

[0049] The camera's forward direction indicator is displayed on the current video frame, where the forward direction indicator is used to indicate the camera's moving direction.

[0050] In some exemplary embodiments of the present disclosure, the method further includes:

[0051] A forward viewing angle mark is displayed on the current video frame.

[0052] In some exemplary embodiments of the present disclosure, the method further includes:

[0053] Receive an instruction to play based on a forward perspective, play based on the forward perspective, and record the playback path.

[0054] In some exemplary embodiments of the present disclosure, performing tracking playback based on the first target point of interest and recording the playback path includes:

[0055] When tracking playback is performed based on the first target point of interest, playback is paused when it is detected that the first target point of interest is lost, and the video frame where the paused playback stops is used as the new current video frame. Candidate points of interest are displayed in the new current video frame and the user is prompted to select a new point of interest.

[0056] In some exemplary embodiments of the present disclosure, displaying at least one candidate interest point in the current video frame includes:

[0057] Determining a viewing picture currently displayed on a display, where the viewing picture is a partial area in the current video frame;

[0058] At least one of the candidate points of interest is displayed in the viewing screen.

[0059] In some exemplary embodiments of the present disclosure, displaying at least one candidate interest point on the current video frame includes:

[0060] identifying at least one candidate interest point on the current video frame based on an interest point detection model;

[0061] At least one candidate interest point on the current video frame is displayed.

[0062] In some exemplary embodiments of the present disclosure, the interest point detection model uses at least one of the following algorithms: a target detection algorithm, a saliency detection algorithm, and a video highlight detection algorithm.

[0063] In some exemplary embodiments of the present disclosure, the video editing based on the playback path includes:

[0064] The target video frames involved in the editing are determined, and the editing is performed according to the recorded playback path of each of the target video frames to generate an edited video, where the edited video is a planar video.

[0065] In some exemplary embodiments of the present disclosure, the current video frame is a video frame of a panoramic video.

[0066] According to a second aspect of an embodiment of the present disclosure, a video editing device is provided, comprising:

[0067] a display module for displaying the current video frame and determining a first target point of interest in response to a user's selection;

[0068] A processing module, configured to play the video based on the first target point of interest and record the playback path;

[0069] The processing module is further configured to edit the video based on the playback path.

[0070] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising a memory, a processor and a display, wherein the memory stores a computer program, and when the computer program instructions are executed by the processor, the display is configured to display a current video frame, and the processor is configured to determine a first target point of interest in response to a user's selection; play based on the first target point of interest and record the playback path; and edit the video based on the playback path.

[0071] In some exemplary embodiments of the present disclosure, the display is configured to display at least one candidate point of interest of the current video frame, and the processor is configured to determine a first target point of interest in response to a user's selection.

[0072] In some exemplary embodiments of the present disclosure, the processor is configured to perform tracking playback based on the first target point of interest and record the playback path.

[0073] In some exemplary embodiments of the present disclosure, the processor is configured to determine a playback sequence corresponding to the first target point of interest, play based on the playback sequence, and record a playback path.

[0074] In some exemplary embodiments of the present disclosure, the first target point of interest is an interest point determined from the at least one candidate point of interest or an interest point determined from other non-candidate points of interest in the current video frame.

[0075] In some exemplary embodiments of the present disclosure, the processor is configured to determine a key video frame in a subsequent video frame of the current video frame, and determine a second target point of interest in the key video frame; play based on the first target point of interest and the second target point of interest, and record the playback path.

[0076] In some exemplary embodiments of the present disclosure, the processor is configured to determine a viewing angle of an intermediate video frame between the current video frame and the key video frame, and the viewing angle is used to gradually transition from the first target point of interest to the second target point of interest.

[0077] In some exemplary embodiments of the present disclosure, the processor is configured to determine a viewing angle of each video frame based on the first target point of interest, and the viewing angle is used to gradually transition the first target point of interest to the center of the viewing screen and maintain it in the center of the viewing screen; play is performed based on the determined viewing angle of each video frame, and the playback path is recorded.

[0078] In some exemplary embodiments of the present disclosure, the display is configured to display a forward path indicator corresponding to each candidate point of interest on the current video frame.

[0079] In some exemplary embodiments of the present disclosure, the forward path identifier is used to indicate the moving direction corresponding to the point of interest at the next moment.

[0080] In some exemplary embodiments of the present disclosure, the forward path identifier is used to indicate the subsequent moving direction of the point of interest.

[0081] In some exemplary embodiments of the present disclosure, the processor is configured to identify the motion trajectory of each candidate point of interest based on a point of interest tracking model; the processor is configured to display the forward path identification corresponding to each candidate point of interest based on the motion trajectory; wherein the point of interest tracking model uses a target tracking algorithm.

[0082] In some exemplary embodiments of the present disclosure, the processor is configured to perform tracking playback based on the moving direction indicated by the forward path identifier corresponding to the first target point of interest, and record the playback path.

[0083] In some exemplary embodiments of the present disclosure, the processor is configured to, when receiving a pause or rewind instruction from the user during tracking playback based on the first target point of interest, pause or rewind to the video frame selected by the user, use the paused or rewinded video frame as the new current video frame, and display the candidate points of interest on the current video frame; in response to the user selecting a second target point of interest from the candidate points of interest for tracking playback, record the playback path of each video frame.

[0084] In some exemplary embodiments of the present disclosure, the processor is configured to, during the tracking playback based on the first target point of interest, pause or rewind to the video frame selected by the user when receiving a pause or rewind instruction from the user, use the paused or rewinded video frame as the new current video frame, and display the candidate points of interest and the corresponding forward path identifier on the current video frame; in response to the second target point of interest selected by the user from the candidate points of interest, perform tracking playback based on the moving direction indicated by the corresponding forward path identifier, and record the playback path of each video frame.

[0085] In some exemplary embodiments of the present disclosure, the display is configured to display a preview playback screen of the candidate point of interest in response to a user's preview instruction for the candidate point of interest, and the preview playback screen can be used to preview the motion trajectory screen of the candidate point of interest within a preset time length.

[0086] In some exemplary embodiments of the present disclosure, the processor is configured to respond to a user's frame skip instruction, stop at a new video frame selected by the user, and use the new video frame as the current video frame, and the display is configured to display candidate points of interest in the current video frame.

[0087] In some exemplary embodiments of the present disclosure, the display is configured to display a forward direction indicator of the camera on the current video frame, where the forward direction indicator is used to indicate a moving direction of the camera.

[0088] In some exemplary embodiments of the present disclosure, the display is configured to display a forward viewing angle mark on the current video frame.

[0089] In some exemplary embodiments of the present disclosure, the processor is configured to receive an instruction to play based on a forward perspective, play based on the forward perspective, and record the playback path.

[0090] In some exemplary embodiments of the present disclosure, the processor is configured to pause playback when detecting that the first target point of interest is lost during tracking playback based on the first target point of interest, use the video frame where the playback is paused as the new current video frame, display candidate points of interest in the new current video frame, and prompt the user to select a new point of interest.

[0091] In some exemplary embodiments of the present disclosure, the processor is configured to determine a viewing screen currently displayed by the display, where the viewing screen is a partial area in the current video frame; and the display is configured to display at least one candidate point of interest in the viewing screen.

[0092] In some exemplary embodiments of the present disclosure, the processor is configured to identify at least one candidate interest point on the current video frame based on an interest point detection model; and the display is configured to display at least one candidate interest point on the current video frame.

[0093] In some exemplary embodiments of the present disclosure, the interest point detection model uses at least one of the following algorithms: a target detection algorithm, a saliency detection algorithm, and a video highlight detection algorithm.

[0094] In some exemplary embodiments of the present disclosure, the processor is configured to determine target video frames involved in editing, perform editing according to the recorded playback path of each target video frame, and generate an edited video, where the edited video is a planar video.

[0095] In some exemplary embodiments of the present disclosure, the current video frame is a video frame of a panoramic video.

[0096] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in the first aspect are implemented.

[0097] The technical solution provided by the embodiments of the present disclosure may include the following beneficial effects: the display shows the current video frame, allowing the user to participate in the editing process according to personal preferences, the user can determine the first target point of interest through interaction with the electronic device, and play it, while recording the playback path. The process of playing based on the first target point of interest is to allow the user to directly see the corresponding editing effect, that is, what you see is what you get, so that the generated edited video contains the first target point of interest that the user cares about, which improves the personalization and satisfaction of the edited video, while improving the efficiency of the video editing process and the user experience.

[0098] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0099] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0100] Fig. 1 is a flowchart showing a video editing method according to an exemplary embodiment.

[0101] Fig. 2 is a flowchart showing a video editing method according to an exemplary embodiment.

[0102] Fig. 3a is a schematic diagram showing a panoramic video according to an exemplary embodiment.

[0103] Fig. 3b is a schematic diagram showing a viewing screen according to an exemplary embodiment.

[0104] Fig. 3c is a schematic diagram showing a viewing screen according to an exemplary embodiment.

[0105] Fig. 3d is a schematic diagram showing a viewing screen according to an exemplary embodiment.

[0106] Fig. 4 is a schematic diagram showing an interactive interface according to an exemplary embodiment.

[0107] Fig. 5a is a schematic diagram showing a forward path identification according to an exemplary embodiment.

[0108] Fig. 5b is a schematic diagram showing a forward path identification according to an exemplary embodiment.

[0109] Fig. 6 is a schematic diagram showing a forward direction mark according to an exemplary embodiment.

[0110] Fig. 7 is a schematic diagram showing a forward viewing angle mark according to an exemplary embodiment.

[0111] Fig. 8 is a flowchart showing a video editing method according to an exemplary embodiment.

[0112] Fig. 9 is a block diagram showing a video editing apparatus according to an exemplary embodiment.

[0113] Fig. 10 is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0114] Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. In the following description, when referring to the drawings, like numbers in different figures represent like or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present invention, as detailed in the appended claims.

[0115] Video editing technology is widely used today. Users can use the automatic editing function in video editing applications to edit raw videos. While this function is simple to use, the resulting edited videos may not always reflect the segments the user is interested in. This can result in missing segments that the user considers interesting, preventing the user from obtaining the desired video content.

[0116] In order to solve the above-mentioned problems, the present disclosure provides a video editing method, in which a display shows the current video frame, allowing the user to participate in the editing process according to personal preferences. The user can determine the first target interest point by interacting with the electronic device, and the electronic device performs video editing based on the first target interest point, so that the generated edited video contains the first target interest point that the user is concerned about, thereby improving the personalization and satisfaction of the edited video, and at the same time improving the efficiency of the video editing process and the user experience.

[0117] The exemplary embodiments of the present disclosure provide a video editing method for use in electronic devices, such as smart devices such as smartphones, tablets, laptops, and computers. This method can also be integrated into video editing applications, providing users with an interactive editing interface and functioning as a video editing feature. Users can use this video editing method to edit videos based on their personal needs and visual preferences.

[0118] As shown in FIG1 , the video editing method shown in the present disclosure includes:

[0119] S101, displaying the current video frame, and determining a first target point of interest in response to a user's selection;

[0120] S102, playing based on the first target point of interest, and recording the playing path;

[0121] S103: Edit the video based on the playback path.

[0122] In step S101, the original video used for video editing can be a video with a large field of view angle, such as a panoramic video, an ultra-wide-angle video, a single wide-angle video, a single fisheye video, etc. The original video can be captured by a camera device corresponding to the electronic device, or it can be captured by other camera devices such as a panoramic camera and acquired based on data transmission, or it can be acquired by stitching together two-dimensional videos shot by multiple ordinary cameras, etc.

[0123] When the original video is a panoramic video, since the panoramic video contains 360-degree all-round picture content, and the display of the electronic device is mostly a flat display, in order to ensure that the electronic device can normally display the picture content of the original video, the electronic device needs to reproject the original video, such as using stereographic projection (Stereographic Projection), equidistant azimuthal projection (Azimuthal equidistant), etc., to convert the panoramic video into a flat video. It is understandable that if the electronic device has a curved wide-angle display, the panoramic video does not need to be reprojected. When the original video is an ultra-wide-angle video, a single wide-angle video, or a single fisheye video, since the display interface of the electronic device is limited, the original video can be cropped or reprojected.

[0124] The electronic device loads the original video and displays the current video frame of the original video on the display. There are points of interest in the picture content of the current video frame. The points of interest can be the area where the preset target object, significant object or object with wonderful performance is located in the current video frame, or a certain point in the area. For example, the point of interest can be a living thing such as a person, animal, plant, or the area where non-living things such as a car, building, mountain, etc. The point of interest can also refer to the center point of a person, animal, car, building, etc. The user can select the point of interest of his or her interest in the current video frame as the first target point of interest based on personal interests or clips. When the display screen of the electronic device is a touch screen, the user can select the first target point of interest by clicking on the touch screen; when the display screen of the electronic device is a non-touch screen, the user can use an auxiliary device such as a mouse to select the first target point of interest.

[0125] In step S102, after determining the first target point of interest, the display of the electronic device will start playing the video frame by frame based on the first target point of interest, starting from the current video frame. Playing based on the first target point of interest can include two situations: one is tracking playback based on the first target point of interest, and the other is playing based on the playback sequence corresponding to the first target point of interest.

[0126] In some embodiments, tracking playback is performed based on the first target point of interest, and the playback path is recorded.

[0127] The electronic device tracks and plays the first target point of interest in real time, determines the viewing angle of each video frame played, and presents the viewing screen corresponding to the viewing angle to the user on the display. The playback path is composed of the viewing angle of each video frame. By recording the viewing angle corresponding to each video frame, it is known which area of ​​the video frame to crop later. When the first target point of interest is a non-static object or the camera device moves relative to the first target point of interest, the position of the first target point of interest on the viewing screen of the display will move. In order to highlight the first target point of interest that the user is concerned about, the electronic device tracks and plays the first target point of interest, that is, the viewing screen of the display will change with the movement of the first target point of interest, ensuring that the first target point of interest always exists on the viewing screen. At the same time, the electronic device records the playback path of all video frames containing the first target point of interest in the viewing screen.

[0128] In some embodiments, a playback sequence corresponding to the first target point of interest is determined, playback is performed based on the playback sequence, and the playback path is recorded.

[0129] The electronic device can pre-track all points of interest in the current video frame or the entire original video in the background, that is, pre-track and identify each point of interest in the video, and obtain a tracking sequence corresponding to each point of interest. The user can directly select the first target point of interest of interest from the current video frame, or preview the tracking sequence of each point of interest and then select the point of interest of interest as the first target point of interest. After the user selects a point of interest as the first target point of interest, the display directly plays the corresponding tracking sequence and records the playback path corresponding to the tracking sequence.

[0130] In step S103, the electronic device edits the video according to the playback path, and saves the picture actually played on the display as the final edited video, so that the final edited video contains the video frame of the first target interest point.

[0131] In the present disclosure, the display shows the current video frame, allowing the user to participate in the editing process according to personal preferences. The user can determine the first target point of interest by interacting with the electronic device, and the electronic device performs video editing based on the first target point of interest, so that the generated edited video contains the first target point of interest that the user is concerned about, thereby improving the personalization and satisfaction of the edited video, and at the same time improving the efficiency of the video editing process and the user experience.

[0132] In some embodiments, presenting the current video frame and determining a first target point of interest in response to a user selection includes:

[0133] At least one candidate interest point of the current video frame is displayed, and a first target interest point is determined in response to a user's selection.

[0134] To facilitate user selection of the first target point of interest and to provide a prompt to the user, at least one candidate point of interest in the current video frame may be highlighted as an icon. For example, when the point of interest is a point within the area where an object is located, the point of interest may be displayed as an icon to prompt the user of the candidate point of interest, as shown in FIG3b , where the point of interest is displayed as a circle filled with a pattern. When the point of interest is within the area where an object is located, the point of interest may be displayed as an outline to prompt the user of the candidate point of interest, as shown in FIG3d , where the point of interest is displayed with a gray shading.

[0135] In one embodiment, after the user selects and determines the first target point of interest, during the playback of the first target point of interest, the display does not highlight other candidate points of interest in the form of icons, outlines, etc., to avoid visual impact on the user's editing process.

[0136] In the present disclosure, displaying candidate interest points in the current video frame can provide users with editing ideas, reduce the user's selection time for the editing subject, and improve editing efficiency.

[0137] In some embodiments, the first target point of interest is an interest point determined from at least one candidate point of interest or an interest point determined from other non-candidate points of interest in the current video frame.

[0138] The display displays at least one candidate point of interest in the current video frame, and the user can select and determine a first target point of interest from the candidate points of interest based on their interest. However, since the candidate points of interest are determined by the electronic device detecting the current video frame, it is possible that none of the displayed candidate points of interest are of interest to the user. In this case, the user can freely select an area of ​​interest, that is, determine the first target point of interest from other non-candidate points of interest in the current video frame. For example, the user can select the first target point of interest in the current video frame by selecting a box or dotting it.

[0139] In some embodiments, the method further comprises:

[0140] Determining a key video frame in a subsequent video frame of the current video frame, and determining a second target interest point in the key video frame;

[0141] Play based on the first target point of interest and record the playback path, including:

[0142] Play the video based on the first target point of interest and the second target point of interest, and record the playback path.

[0143] Since the original video is composed of many video frames, the picture content of each video frame displayed on the monitor is different, and the points of interest in the video frames will also change. If the user wants to edit a video containing multiple target points of interest, the user can determine the key video frame in the subsequent video frames of the current video frame, and select another target point of interest as the second target point of interest by marking in the key video frame, that is, select the playback path from one point of interest to another.

[0144] After determining the first target point of interest and the second target point of interest, the display plays based on the first target point of interest and the second target point of interest, and records the playback path.

[0145] In some embodiments, playing based on the first target point of interest and the second target point of interest and recording the playback path includes:

[0146] A viewing angle of an intermediate video frame between the current video frame and the key video frame is determined, where the viewing angle is used to gradually transition from the first target interest point to the second target interest point.

[0147] Since the display can only show part of the original video at a time, that is, the user can only see part of the area, the user's current visible area is usually called the viewing angle. The playback path is the viewing angle recorded for each video frame, that is, the regional position of the partial area played. The user can plan the playback path by determining the first target point of interest in the current video frame and the second target point of interest in the key video frame. In this way, the playback path will gradually transition from the first target point of interest to the second target point of interest. The playback path can be planned automatically or manually by the user, that is, the user manually sets the transition path.

[0148] In another embodiment, playback is performed based on the first target point of interest, including: obtaining a second target point of interest selected by the user and a transition path from the first target point of interest to the second target point of interest determined by the user, and playing based on the first target point of interest, the second target point of interest, and the transition path from the first target point of interest to the second target point of interest.

[0149] In order to avoid the transition process of the target point of interest being too abrupt and affecting the visual effect of the edited video, when playing the first target point of interest and the second target point of interest, the viewing angle of the intermediate video frame between the current video frame and the key video frame will gradually transition from the first target point of interest to the second target point of interest. For example, the current video frame is the first frame, the user selects the person in the current video frame as the first target point of interest, the key video frame is the eighth frame, and the user selects the puppy in the key video frame as the second target point of interest, then during playback, the viewing angle will gradually transition from the person in the first frame to the puppy in the eighth frame, and the generated edited video will include the viewing angle of the first frame, the viewing angle from the person to the puppy, and the viewing angle of the eighth frame. The gradual transition of the viewing angle from the first target point of interest to the second target point of interest can improve the picture continuity of the edited video, prevent the picture from feeling fragmented, and give the user a good visual experience.

[0150] In some embodiments, tracking playback based on the first target point of interest and recording the playback path includes:

[0151] Determining a viewing angle of each video frame according to the first target point of interest, the viewing angle being used to gradually transition the first target point of interest to the center of the viewing screen and maintain it at the center of the viewing screen;

[0152] Play each video frame based on the determined viewing angle and record the playback path.

[0153] Because the original video can be panoramic, wide-angle, or other videos with a wide field of view, and the display size is limited, the viewing angle of each video frame may change. Since the first target point of interest is selected by the user and represents the user's interest, to highlight the first target point of interest, during tracking playback of the first target point of interest, the viewing angle of each video frame is determined based on the first target point of interest's position within the display's viewing frame, so that the first target point of interest gradually transitions to the center of the viewing frame and remains there. For example, if the first target point of interest selected by the user in the current video frame is located on the right side of the display, to gradually transition the first target point of interest to the center of the viewing frame, the display's viewing angle shifts rightward until the first target point of interest is located at the center of the viewing frame. This position is maintained throughout the subsequent tracking playback process. The electronic device records the playback path of the first target point of interest gradually transitioning to the center of the viewing frame and remaining there, and the subsequently generated video clip stores this change in viewing angle.

[0154] In the present disclosure, the viewing angle of each video frame will change according to the first target point of interest. Keeping the first target point of interest at the center of the viewing screen can emphasize the first target point of interest, meet the user's editing needs, and the generated edited video is in line with the user's intentions.

[0155] According to an exemplary embodiment, as shown in FIG2 , the video editing method in this embodiment includes:

[0156] S201, displaying at least one candidate interest point in the current video frame;

[0157] S202: Displaying a forward path identifier corresponding to each candidate point of interest on the current video frame;

[0158] S203, determining a first target point of interest in response to the user's selection;

[0159] S204: Tracking and playing based on the first target point of interest, and recording the playback path;

[0160] S205: Edit the video based on the playback path.

[0161] Among them, steps S201, S203, S204, and S205 are the same as those in the above embodiment and will not be repeated here.

[0162] In step S202, to facilitate user selection of a first target POI from multiple candidate POIs, a forward path indicator for each candidate POI is displayed on the current video frame to indicate the user's motion trajectory. This disclosure does not limit the specific display format of the forward path indicator; for example, the forward path indicator may be displayed as an arrow, a shape representing a motion trend, such as a finger pattern, or a symbol. The forward path indicator corresponding to each candidate POI appears only during user interaction with the display screen and is not displayed during actual playback.

[0163] In some embodiments, the forward path identifier is used to indicate the moving direction corresponding to the point of interest at the next moment.

[0164] The forward path marker can be used to indicate the moving direction of the interest point at the next moment. As shown in Figure 5a, the forward path marker is a black arrow. There are multiple candidate interest points in the current video frame, and each candidate interest point has a corresponding left forward path marker, which means that at the next moment, the moving direction of all candidate interest points is to the left.

[0165] In some embodiments, the forward path indicator is used to indicate the subsequent movement direction of the point of interest.

[0166] The forward path identifier can be used to indicate the overall moving trend direction of the interest point in the subsequent process. As shown in Figure 5b, the forward path identifier is a gray arrow. There are multiple candidate interest points in the current video frame, and each candidate interest point has a corresponding forward path identifier. The forward path identifiers of different candidate interest points are different. Among them, the forward path identifiers of candidate interest point 1, candidate interest point 2, candidate interest point 3, and candidate interest point 4 indicate that the subsequent overall moving trend direction of the mountain and the tree is diagonally forward to the left, and the forward path identifier of candidate interest point 5 indicates that the subsequent overall moving trend direction of the person is to the left and then forward.

[0167] Of course, different forward path indicators can also be displayed using the same form of indicators. For example, a black arrow as shown in FIG5a can be used to indicate the next moving direction of the point of interest, or to indicate the overall moving trend of the point of interest. For another example, a gray arrow as shown in FIG5b can be used to indicate the next moving direction of the point of interest, or to indicate the overall moving trend of the point of interest.

[0168] In the present disclosure, displaying the forward path identifier can provide users with more information about candidate points of interest, allowing users to know the moving direction of each candidate point of interest. Displaying the forward path identifier during the video editing process facilitates users to select the first target point of interest, meets the user's editing needs, and effectively improves the efficiency of editing.

[0169] In some embodiments, displaying a forward path indicator corresponding to each candidate point of interest includes:

[0170] Based on the interest point tracking model, the motion trajectory of each candidate interest point is identified;

[0171] Based on the motion trajectory, the forward path identification corresponding to each candidate point of interest is displayed;

[0172] Among them, the point of interest tracking model uses the target tracking algorithm.

[0173] The electronic device has a built-in point of interest tracking model, which identifies the motion trajectory of each candidate point of interest. Based on the motion trajectory of each candidate point of interest, the electronic device displays the corresponding forward path identification for each candidate point of interest. The point of interest tracking model uses a target tracking algorithm, which can be an existing Yolo v1, v2, v3, v4, v5, X, SSD, RetinaNet, etc. To improve recognition accuracy and efficiency, video samples with marked points of interest and forward path identification can also be used to self-train the existing network. The point of interest tracking model uses a self-trained target tracking algorithm.

[0174] In the present disclosure, the point of interest tracking model can accurately identify the motion trajectory of each candidate point of interest, and the displayed forward path identification can reflect the motion trend of the candidate point of interest. The target point of interest selected in combination with the forward path identification is more in line with the user's ideas.

[0175] In some embodiments, tracking playback based on the first target point of interest and recording the playback path includes:

[0176] Tracking playback is performed based on the moving direction indicated by the forward path identifier corresponding to the first target point of interest, and the playback path is recorded.

[0177] To ensure that the edited video includes the first target point of interest selected by the user, the electronic device tracks and plays the video based on the movement direction indicated by the forward path identifier corresponding to the first target point of interest, and records the playback path of the viewing image on the display. The electronic device then edits the video according to the playback path, so that the final edited video includes the image of the first target point of interest moving along the forward path identifier. For example, if the forward path identifier corresponding to the first target point of interest indicates that the subsequent movement direction of the first target point of interest is to the left, the electronic device will track the leftward movement of the first target point of interest and play the video, so that the display will appear to be moving to the left.

[0178] In some embodiments, tracking playback based on the first target point of interest and recording the playback path includes:

[0179] During the tracking playback based on the first target point of interest, upon receiving a pause or rewind command from the user, the video is paused or rewound to the video frame selected by the user, the paused or rewound video frame is used as the new current video frame, and the candidate point of interest is displayed on the current video frame;

[0180] In response to the user selecting a second target point of interest from the candidate points of interest for tracking playback, a playback path of each video frame is recorded.

[0181] As shown in Figure 4, in addition to the viewing screen, the display also includes multiple virtual buttons, meaning the viewing screen also serves as an interactive interface. These virtual buttons include a pause button, a rewind button, a save button, and a frame skip button. During the tracking playback of the first target point of interest, if the user believes that the first target point of interest or the movement trajectory of the first target point of interest in the viewing screen of the display screen does not meet expectations, the user can click the pause button or the rewind command to adjust the tracked point of interest.

[0182] When the user clicks the pause button, triggering the pause command, the display stops tracking the first target point of interest and pauses at the video frame corresponding to the time the pause command was received, with the video frame corresponding to the time the pause command was received being the new current video frame. When the user clicks the rewind button, triggering the rewind command, the display rewinds to at least one frame before the current video frame. The user can specify a specific number of frames to rewind, for example, if the rewind command is clicked multiple times, the display will rewind to the number of frames corresponding to the number of clicks, with the rewinded video frame being the new current video frame. In addition, the rewind command can also be implemented by dragging the video progress bar.

[0183] In the new current video frame, in order to enable the user to reselect the point of interest, the candidate points of interest are displayed on the current video frame, and the user can select a second target point of interest from the candidate points of interest, wherein the second target point of interest can be the same point of interest as the first target point of interest or a different point of interest. The display tracks and plays the second target point of interest, and records the playback path of each frame of the video frame to ensure that the final generated clip video contains the second target point of interest. After the second target point of interest is selected, in order to make the camera movement smooth, it will gradually transition from the first target point of interest to the second target point of interest as the center. The specific transition method can adopt an intelligent recommendation camera movement method, or the transition path can be manually set. In one embodiment, the transition can be achieved by interpolating frames based on two points of interest. The transition method is not limited here.

[0184] In the present disclosure, the electronic device provides a new current video frame to the user in response to a pause instruction and a rewind instruction. The user can select other target points of interest on the new current video frame for tracking and playback, so that the final generated clip video contains multiple points of interest, is rich in content, and avoids a monotonous video screen.

[0185] In some embodiments, tracking and playing the video based on the forward path identifier corresponding to the first target point of interest and recording the playback path include:

[0186] During the tracking playback based on the first target point of interest, upon receiving a pause or rewind command from the user, the video is paused or rewound to the video frame selected by the user, the paused or rewound video frame is used as the new current video frame, and the candidate point of interest and the corresponding forward path identifier are displayed on the current video frame;

[0187] In response to a second target point of interest selected by the user from the candidate points of interest, tracking playback is performed based on the moving direction indicated by the corresponding forward path identifier, and the playback path of each video frame is recorded.

[0188] In order to enable users to know the movement trajectory of each candidate point of interest, after receiving a pause or rewind command, the candidate points of interest and the corresponding forward path identifiers are displayed on the paused or rewound video frame. Users can consider selecting a second target point of interest based on the moving direction indicated by the forward path identifier of each candidate point of interest.

[0189] In some embodiments, the video editing method further includes:

[0190] In response to a user's preview instruction for a candidate point of interest, a preview playback screen of the candidate point of interest is displayed, and the preview playback screen can be used to preview a motion trajectory screen of the candidate point of interest within a preset time period.

[0191] In order to avoid the impact of accidental touch on the editing process, the preview instruction can be set to long press the candidate point of interest, for example, long press for n seconds (n≥2). When the user triggers the preview instruction, the electronic device responds to the preview instruction and displays the motion trajectory of the candidate point of interest within a preset time length on the display. It should be noted that under the preview instruction, the electronic device does not record the motion trajectory of the candidate point of interest as a playback path. If the motion trajectory picture of the candidate point of interest meets the user's expectations, the user can select the candidate point of interest as the first target point of interest. If the motion trajectory picture of the candidate point of interest does not meet the user's expectations, the user can select other candidate points of interest as the first target point of interest.

[0192] Since the preview playback screen can intuitively show the user the motion trajectory of each candidate point of interest, and the preview playback screen will not be recorded in the playback path, the user does not need to perform a backoff operation when the motion trajectory of the previewed candidate point of interest does not meet expectations. The user can directly select other candidate points of interest as the first target point of interest. The preview instruction can effectively avoid frequent backoff operations caused by incorrect selection of the first target point of interest, thereby reducing the processing pressure of the electronic device.

[0193] In some embodiments, the video editing method further includes:

[0194] In response to the user's frame skipping instruction, the video stops at a new video frame selected by the user, takes the new video frame as the current video frame, and displays the candidate interest points in the current video frame.

[0195] Frame skipping means not playing the original video frame by frame in the order of each video frame, but jumping from the current video frame to other video frames that are not adjacent to the current video frame. During the video editing process, if the user is not interested in the video content of the time period corresponding to the current video frame, the user can apply a frame skip instruction to the picture displayed on the display according to personal editing needs. For example, the video time corresponding to the current video frame is 5:00, and the user has no editing requirements for this part of the content, then a frame skip instruction can be applied to jump directly to the video frame corresponding to 5:05. In response to the user's frame skip instruction, the electronic device displays the new video frame selected by the user on the display, and uses the new video frame as the current video frame, and displays candidate points of interest in the current video frame. The user can select the first target point of interest from the displayed candidate points of interest.

[0196] Among them, the frame skip instruction can be a touch operation acting on the video progress bar, and the user can implement the frame skip operation by moving the video progress bar; the frame skip instruction can also be a touch operation acting on the frame skip button, as shown in Figure 4. If the frame skip button is clicked once, the display will display the video frame one frame later. If the user clicks the frame skip button three times, the display will stay and display the video frame three frames later. If the frame skip button is clicked once, the video frame one second later will be displayed. If the user clicks the frame skip button three times, the display will stay and display the video frame three seconds later.

[0197] In the present disclosure, users can apply frame skipping instructions to select video frames of interest. Frame skipping instructions improve the flexibility of video editing, avoid including content that the user is not interested in in the generated edited video, and improve user satisfaction.

[0198] In some embodiments, the video editing method further includes:

[0199] The camera's forward direction indicator is displayed on the current video frame. The forward direction indicator is used to indicate the camera's movement direction.

[0200] The forward direction indicator indicates the direction of camera movement. By viewing the camera's forward direction indicator, the user can understand the display changes of all objects in the next video frame. As shown in Figure 6, the forward direction indicator 6 can be displayed as a white swallowtail arrow. In the current video frame, the arrow of the forward direction indicator points vertically upward, indicating that the camera is moving upward. Since the camera moves away from the object being photographed, all objects in the next video frame will be displayed smaller than in the current video frame.

[0201] In the present disclosure, the forward direction indicator indicates the direction of movement of the camera. The display of the forward direction indicator provides users with more editing ideas. When the displayed screen does not have the points of interest that the user is interested in, the user can complete the editing of the current video frame based on the forward direction indicator to enrich the video content.

[0202] In some embodiments, the video editing method further includes:

[0203] Displays the forward view marker on the current video frame.

[0204] The forward perspective refers to the perspective in the forward direction of camera movement. The image contained within the forward perspective marker represents the forward perspective of the camera as it is displayed on the display screen. As shown in Figure 7, the forward perspective marker 7 can be displayed as a black dashed box, indicating that the forward perspective represents a distant mountain. The forward perspective marker provides additional consideration when selecting a first target POI. For example, the user can combine the forward perspective marker with the candidate POI contained within the forward perspective marker to select the first target POI.

[0205] In some embodiments, the video editing method further includes:

[0206] Receive instructions for playing based on a forward perspective, play based on the forward perspective, and record the playback path.

[0207] The user triggers the forward-perspective playback command through touch, gesture, or voice, for example, by double-clicking the forward-perspective marker, clicking the corresponding virtual button, or typing "play forward-perspective." Upon receiving the forward-perspective playback command, the electronic device plays the video based on the forward perspective, and the display's viewing angle changes based on the forward perspective. The electronic device records the playback path of the video frames, and the resulting video clip, generated based on the playback path, includes the video content from the perspective of the camera's forward direction.

[0208] For example, the user applies a pause command, and the viewing angle of the current video frame is shown in Figure 7. If the user completes the forward perspective playback instruction for the forward perspective mark 7, since the forward perspective mark 7 is a distant mountain in the current video frame, the electronic device will change the viewing angle from the viewing angle of the current video frame to the picture contained in the forward perspective mark 7 when playing in response to the forward perspective playback instruction. The final generated clip includes a video clip with the mountain contained in the forward perspective mark 7 as the viewing angle.

[0209] In the present disclosure, the forward perspective identifier provides users with another editing idea. The editing video generated based on the forward perspective identifier has rich viewing perspectives to highlight different picture contents and give users a good editing and viewing experience.

[0210] In some embodiments, tracking playback based on the first target point of interest and recording the playback path includes:

[0211] When tracking playback is performed based on the first target point of interest, playback is paused when it is detected that the first target point of interest is lost, and the video frame where the paused playback stops is used as the new current video frame. The candidate points of interest are displayed in the new current video frame and the user is prompted to select a new point of interest.

[0212] During the display's tracking and playback of the first target point of interest, if the first target point of interest cannot be detected in the viewing screen of the display, that is, the first target point of interest is lost, if the playback continues and the playback path is recorded, it cannot be guaranteed that the final generated clip video can contain the content that the user is concerned about. Therefore, when it is detected that the first target point of interest is lost, the original video is paused, and the video frame where the playback is paused is used as the new current video frame. The candidate points of interest are displayed in the new current video frame, and the user can be prompted to select a new point of interest in the form of text, sound, vibration, etc. for subsequent editing.

[0213] In addition, when the user applies a pause command, a rewind command, or the first target point of interest is lost, the video editing process can be completed through a save command. When the user triggers the save command, the electronic device responds to the save command and no longer records the subsequent playback path, but only generates a clipped video based on the recorded playback path before responding to the save command.

[0214] In the present disclosure, when the first target point of interest is lost and cannot be tracked, the electronic device promptly reminds the user to select a new point of interest, so that each video frame of the generated clipped video has a picture focus, avoiding the existence of meaningless video frames.

[0215] In some embodiments, displaying at least one candidate interest point in the current video frame includes:

[0216] Determine the viewing image currently displayed on the display, where the viewing image is a portion of the current video frame;

[0217] At least one candidate point of interest is displayed in the viewing screen.

[0218] Due to display specifications, when the original video is a panoramic video, an ultra-wide-angle video, or other video with a large field of view, the display cannot fully display a single frame of the original video at once during the actual display process. Therefore, the viewing screen currently displayed by the display is a partial area of ​​the current video frame, and at least one candidate point of interest is displayed in the viewing screen. When the original video is initially loaded, the display screen can display the partial area of ​​the current video frame at a default viewing angle, for example, using the camera's forward direction as the default viewing angle. When the viewing screen currently displayed by the display does not contain the user's desired point of interest, the user can interactively change the viewing screen currently displayed by the display, for example, by dragging the viewing screen to move the viewing angle, or by touching and clicking a virtual button to move the viewing angle, so that the viewing screen currently displayed by the display is another partial area of ​​the current video frame. In one embodiment, to save computing resources, only the candidate points of interest contained in the currently displayed viewing screen are detected and then displayed. After the viewing screen is changed, detection and display are performed again for the changed viewing screen.

[0219] For example, Figure 3a is a frame of a panoramic video. The electronic device uses the panoramic video corresponding to Figure 3a as the original video for editing. Due to the display's specifications, the viewing screen currently displayed on the display is a partial area of ​​the current video frame. As shown in Figure 3b, five candidate points of interest are displayed in the viewing screen. Because the viewing screen currently displayed in Figure 3b does not include the user's desired point of interest, the user drags the viewing angle to change the viewing screen currently displayed on the display to that shown in Figure 3c, which displays seven candidate points of interest.

[0220] In the present disclosure, users can change the viewing screen of the display by interacting with the electronic device, pay attention to all candidate points of interest in each video frame, avoid omissions, ensure that users can select the objects of greatest interest, and make the edited video consistent with the user's ideas.

[0221] In some embodiments, displaying at least one candidate interest point on the current video frame includes:

[0222] Based on the interest point detection model, identifying at least one candidate interest point on the current video frame;

[0223] Display at least one candidate interest point on the current video frame.

[0224] The electronic device pre-stores a POI detection model. After acquiring the original video, the electronic device can use the POI detection model to identify candidate POIs. After the electronic device uses the POI detection model to identify the candidate POIs, it can display the candidate POIs on a display with a preset identifier to prompt the user to select a first target POI.

[0225] For electronic devices with better performance and higher computing power, interest point detection can be performed on the entire area of ​​each frame of the original video, and all candidate interest points on each frame of the video can be identified. When the viewing angle changes, there is no need to repeat interest point detection for the same frame of the video. For electronic devices with average performance and ordinary computing power, interest point detection can be performed only on each frame of the original video and on part of the area displayed on the display, and some candidate interest points on each frame of the video can be identified. When the viewing angle changes, interest point detection needs to be performed again on another part of the area of ​​the same frame of the video.

[0226] In the present disclosure, the interest point detection model can accurately identify the candidate interest points of the current video frame without the need for manual identification by the user, which can effectively improve the efficiency of video editing. At the same time, the display of candidate interest points also provides users with a variety of possible video editing options, thereby improving user satisfaction with video editing.

[0227] In some embodiments, the interest point detection model uses at least one of the following algorithms: an object detection algorithm, a saliency detection algorithm, and a video highlight detection algorithm.

[0228] Target detection algorithms can identify target objects from a single image frame. Common target detection algorithms include: R-CNN (Region with Convolutional Neural Networks), SPP-Net (Spatial Pyramid Pooling Networks), Fast R-CNN (Fast Regions with Convolutional Neural Networks), R-FCN (Region-based fully convolutional network), OverFeat network, YOLOv1 (You Only Look Once v1) network, YOLOv3 (You Only Look Once v3) network, SSD (Single Shot Detection), and RetinaNet (Retina network). Electronic devices can directly use these target detection algorithms as algorithms for point of interest detection models. To improve the recognition accuracy of the point of interest detection model and meet user needs, these target detection algorithms can also be pre-trained using sample videos marked with points of interest, and the pre-trained target detection algorithms can be used as the algorithms for the point of interest detection model.

[0229] Saliency detection algorithms can detect salient regions within an image frame as quickly as the human eye. Common saliency detection algorithms include the Itti algorithm, the FT (Frequency-tuned) algorithm, the SR (Spectral Residual) algorithm, the HC (Histogram-based Contrast) algorithm, the CA (Context-Aware) algorithm, and the GR (Graph-Regularized) algorithm. Similarly, to improve salient region detection accuracy and meet user needs, these algorithms can be pre-trained using sample videos marked with salient regions. This pre-trained algorithm can then be used as the algorithm for the interest point detection model.

[0230] The video highlight detection algorithm can detect objects with wonderful performance from video clips. The video highlight detection algorithm obtains the optical flow of the video frame corresponding to each sampling moment through optical flow calculation, combines the optical flow and video frame to obtain the corresponding comprehensive features of each sampling moment, further calculates the wonderful score based on the comprehensive features, and then determines the objects with wonderful performance based on the wonderful score.

[0231] In some embodiments, video editing based on the playback path includes:

[0232] The target video frames involved in the editing are determined, and the editing is performed according to the recorded playback path of each target video frame to generate an edited video, which is a planar video.

[0233] The target video frame refers to the video frame involved in the editing. The target video frame does not include the video frame skipped by the frame skip instruction, the video frame that is not played after the save instruction, etc. For example, the total length of the original video is 10 minutes. The user applies the frame skip instruction to make the display directly display the video frame corresponding to 3 minutes, and determines the first target point of interest on the candidate points of interest displayed by the video frame corresponding to 3 minutes. The display tracks the first target point of interest and plays it to 5 minutes. At 5 minutes, the save instruction is clicked to end the video editing. Then all video frames from 3 minutes to 5 minutes are target video frames involved in the editing, and the remaining video frames are non-target video frames.

[0234] The electronic device records the playback path of each target video frame, and performs editing according to the playback path of each target video frame to generate an edited video, that is, the screen content of each target video frame on the display interface of the display is saved as an edited video. When the display is a flat display, the target video frame is converted into a flat video through operations such as reprojection or cropping and displayed on the display screen, so the final edited video is also a flat video. When the display is a curved display, in order to improve the applicability of the final edited video and ensure that the edited video can be transmitted to other electronic devices for display and re-editing, the generated edited video is converted into a flat video through operations such as reprojection or cropping.

[0235] In the present disclosure, the video generated based on the playback path clipping of the target video frame includes tracking video frames of all target points of interest, without missing any video clips that the user thinks are wonderful, thereby improving the user's satisfaction with the clipped video.

[0236] In some embodiments, the current video frame is a video frame of a panoramic video.

[0237] When the original video is a panoramic video, the electronic device loads the panoramic video and displays the panoramic video frame on the display. The electronic device uses the panoramic video frame as the current video frame and displays the candidate points of interest in the current video frame. Based on personal preferences and needs, the user can select a first target point of interest from the candidate points of interest in the current video frame and perform tracking playback of the first target point of interest to edit the panoramic video.

[0238] According to an exemplary embodiment, as shown in FIG8 , the video editing method in this embodiment includes:

[0239] S801, performing a reprojection operation on the panoramic video, loading the panoramic video, and displaying a partial area of ​​the current video frame on the display;

[0240] S802, identifying at least one candidate interest point in the current video frame based on the interest point detection model;

[0241] S803, displaying at least one candidate interest point in the current video frame;

[0242] S804: Identify the motion trajectory of each candidate point of interest based on the point of interest tracking model;

[0243] S805: Displaying a forward path identifier corresponding to each candidate point of interest on the current video frame;

[0244] S806, in response to the user's preview instruction for the candidate point of interest, displaying a preview playback screen of the candidate point of interest;

[0245] S807: Determine a first target point of interest in response to the user's selection;

[0246] S808: Track and play based on the forward path identifier corresponding to the first target point of interest, and record the playback path;

[0247] S809: Detecting that the first target point of interest is lost, pausing playback, setting the video frame where playback is paused as a new current video frame, displaying candidate points of interest in the new current video frame, and prompting the user to select a new point of interest;

[0248] S810, determining a second target point of interest in response to a user selection;

[0249] S811, determining a viewing angle of each video frame according to the second target interest point;

[0250] S812: Play each video frame based on the determined viewing angle, and record the playback path;

[0251] S813: Receive a rewind instruction from the user, rewind to the video frame selected by the user, and use the rewinded video frame as the new current video frame;

[0252] S814, displaying a forward view angle mark on the current video frame;

[0253] S815: Receive an instruction to play based on a forward perspective, play based on the forward perspective, and record the playback path;

[0254] S816: Determine target video frames to be edited, and perform editing according to the recorded playback path of each target video frame to generate an edited video.

[0255] The above steps are the same as those in the previous embodiment and will not be described again here.

[0256] An exemplary embodiment of the present disclosure provides a video editing device, as shown in FIG9 , which is a block diagram of a video editing device in the present disclosure.

[0257] The block diagram includes: a display module 91, which is used to display the current video frame and determine the first target point of interest in response to a user's selection; and a processing module 92, which is used to play the video based on the first target point of interest and record the playback path. The processing module 92 is also used to edit the video based on the playback path.

[0258] In an exemplary embodiment of the present disclosure, the display module 91 is further configured to: display at least one candidate point of interest of the current video frame, and determine a first target point of interest in response to a user's selection.

[0259] In an exemplary embodiment of the present disclosure, the processing module 92 is further configured to: perform tracking playback based on the first target point of interest and record the playback path.

[0260] In an exemplary embodiment of the present disclosure, the processing module 92 is further configured to: determine a playback sequence corresponding to the first target point of interest, play the video based on the playback sequence, and record the playback path.

[0261] In an exemplary embodiment of the present disclosure, the first target interest point is an interest point determined from at least one candidate interest point or an interest point determined from other non-candidate interest points in the current video frame.

[0262] In an exemplary embodiment of the present disclosure, the processing module 92 is also used to: determine a key video frame in a subsequent video frame of the current video frame, and determine a second target point of interest in the key video frame; play based on the first target point of interest and the second target point of interest, and record the playback path.

[0263] In an exemplary embodiment of the present disclosure, the processing module 92 is further configured to determine a viewing angle of an intermediate video frame between the current video frame and the key video frame, the viewing angle being used to gradually transition from the first target point of interest to the second target point of interest.

[0264] In an exemplary embodiment of the present disclosure, the processing module 92 is also used to: determine the viewing angle of each video frame based on the first target point of interest, and the viewing angle is used to gradually transition the first target point of interest to the center of the viewing screen and maintain it in the center of the viewing screen; play based on the determined viewing angle of each video frame and record the playback path.

[0265] In an exemplary embodiment of the present disclosure, the display module 91 is further configured to display a forward path identifier corresponding to each candidate point of interest on the current video frame.

[0266] In an exemplary embodiment of the present disclosure, the forward path identifier is used to indicate the moving direction corresponding to the point of interest at the next moment.

[0267] In an exemplary embodiment of the present disclosure, the forward path indicator is used to indicate the subsequent moving direction of the point of interest.

[0268] In an exemplary embodiment of the present disclosure, the display module 91 is also used to: identify the motion trajectory of each candidate point of interest based on the point of interest tracking model; and display the forward path identification corresponding to each candidate point of interest based on the motion trajectory; wherein the point of interest tracking model uses a target tracking algorithm.

[0269] In an exemplary embodiment of the present disclosure, the processing module 92 is further configured to: perform tracking playback based on the moving direction indicated by the forward path identifier corresponding to the first target point of interest, and record the playback path.

[0270] In an exemplary embodiment of the present disclosure, the processing module 92 is also used to: during tracking playback based on the first target point of interest, upon receiving a pause or rewind instruction from the user, pause or rewind to the video frame selected by the user, use the paused or rewinded video frame as the new current video frame, and display the candidate points of interest on the current video frame; in response to the user selecting a second target point of interest from the candidate points of interest, track playback is performed and the playback path of each video frame is recorded.

[0271] In an exemplary embodiment of the present disclosure, the processing module 92 is also used to: during tracking playback based on the first target point of interest, when a pause or rewind instruction from the user is received, pause or rewind to the video frame selected by the user, use the paused or rewinded video frame as the new current video frame, and display the candidate points of interest and the corresponding forward path identifiers on the current video frame; in response to the second target point of interest selected by the user from the candidate points of interest, track playback based on the moving direction indicated by the corresponding forward path identifier, and record the playback path of each video frame.

[0272] In an exemplary embodiment of the present disclosure, the processing module 92 is also used to: respond to the user's preview instruction for the candidate point of interest, display a preview playback screen of the candidate point of interest, and the preview playback screen can be used to preview the motion trajectory screen of the candidate point of interest within a preset time length.

[0273] In an exemplary embodiment of the present disclosure, the processing module 92 is further configured to: in response to a frame skip instruction from the user, pause at a new video frame selected by the user, use the new video frame as the current video frame, and display candidate interest points in the current video frame.

[0274] In an exemplary embodiment of the present disclosure, the display module 91 is further configured to: display a forward direction indicator of the camera on the current video frame, where the forward direction indicator is configured to indicate a moving direction of the camera.

[0275] In an exemplary embodiment of the present disclosure, the display module 91 is further configured to display a forward viewing angle mark on the current video frame.

[0276] In an exemplary embodiment of the present disclosure, the processing module 92 is further configured to: receive an instruction to play based on a forward perspective, play based on the forward perspective, and record a playback path.

[0277] In an exemplary embodiment of the present disclosure, the processing module 92 is also used to: during tracking playback based on the first target point of interest, pause playback when it is detected that the first target point of interest is lost, use the video frame where the paused playback stops as the new current video frame, display candidate points of interest in the new current video frame, and prompt the user to select a new point of interest.

[0278] In an exemplary embodiment of the present disclosure, the display module 91 is further configured to: determine a viewing screen currently displayed on the display, where the viewing screen is a partial area in the current video frame; and display at least one candidate point of interest in the viewing screen.

[0279] In an exemplary embodiment of the present disclosure, the display module 91 is further configured to: identify at least one candidate interest point on the current video frame based on an interest point detection model; and display at least one candidate interest point on the current video frame.

[0280] In an exemplary embodiment of the present disclosure, the interest point detection model uses at least one of the following algorithms: a target detection algorithm, a saliency detection algorithm, and a video highlight detection algorithm.

[0281] In an exemplary embodiment of the present disclosure, the processing module 92 is further used to: determine the target video frames involved in editing, perform editing according to the recorded playback path of each target video frame, and generate an edited video, where the edited video is a planar video.

[0282] In an exemplary embodiment of the present disclosure, the current video frame is a video frame of a panoramic video.

[0283] Regarding the video editing device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0284] Each module in the aforementioned camera device may be implemented in whole or in part through software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor in an electronic device in hardware form, or may be stored in a memory in an electronic device in software form, so that the processor can call and execute the corresponding operations of each module.

[0285] In an exemplary embodiment, an electronic device is provided, including a memory, a processor, and a display, wherein the memory stores a computer program. When the computer program instructions are executed by the processor, the display is configured to display at least one candidate point of interest in a current video frame, and the processor is configured to determine a first target point of interest in response to a user's selection; perform tracking playback based on the first target point of interest and record the playback path; and edit the video based on the playback path.

[0286] In an exemplary embodiment of the present disclosure, the display is configured to display at least one candidate point of interest of the current video frame, and the processor is configured to determine a first target point of interest in response to a user's selection.

[0287] In an exemplary embodiment of the present disclosure, the processor is configured to perform tracking playback based on the first target point of interest and record the playback path.

[0288] In an exemplary embodiment of the present disclosure, the processor is configured to determine a playback sequence corresponding to the first target point of interest, play based on the playback sequence, and record the playback path.

[0289] In an exemplary embodiment of the present disclosure, the first target interest point is an interest point determined from at least one candidate interest point or an interest point determined from other non-candidate interest points in the current video frame.

[0290] In an exemplary embodiment of the present disclosure, the processor is configured to determine a key video frame in a subsequent video frame of the current video frame, determine a second target point of interest in the key video frame; play based on the first target point of interest and the second target point of interest, and record the playback path.

[0291] In an exemplary embodiment of the present disclosure, the processor is configured to determine a viewing angle of an intermediate video frame between a current video frame and a key video frame, where the viewing angle is used to gradually transition from a first target point of interest to a second target point of interest.

[0292] In an exemplary embodiment of the present disclosure, the processor is configured to determine the viewing angle of each video frame based on the first target point of interest, and the viewing angle is used to gradually transition the first target point of interest to the center of the viewing screen and maintain it in the center of the viewing screen; play is performed based on the determined viewing angle of each video frame, and the playback path is recorded.

[0293] In an exemplary embodiment of the present disclosure, the display is configured to display a forward path indicator corresponding to each candidate point of interest on the current video frame.

[0294] In an exemplary embodiment of the present disclosure, the forward path identifier is used to indicate the moving direction corresponding to the point of interest at the next moment.

[0295] In an exemplary embodiment of the present disclosure, the forward path indicator is used to indicate the subsequent moving direction of the point of interest.

[0296] In an exemplary embodiment of the present disclosure, the processor is configured to identify the motion trajectory of each candidate point of interest based on a point of interest tracking model; the processor is configured to display a forward path identifier corresponding to each candidate point of interest based on the motion trajectory; wherein the point of interest tracking model uses a target tracking algorithm.

[0297] In an exemplary embodiment of the present disclosure, the processor is configured to perform tracking playback based on the moving direction indicated by the forward path identifier corresponding to the first target point of interest, and record the playback path.

[0298] In an exemplary embodiment of the present disclosure, the processor is configured to, when receiving a pause or rewind instruction from the user during tracking playback based on the first target point of interest, pause or rewind to the video frame selected by the user, use the paused or rewinded video frame as the new current video frame, and display the candidate points of interest on the current video frame; in response to the user selecting a second target point of interest from the candidate points of interest, track playback is performed and the playback path of each video frame is recorded.

[0299] In an exemplary embodiment of the present disclosure, the processor is configured to, when receiving a pause or rewind instruction from the user during tracking playback based on the first target point of interest, pause or rewind to the video frame selected by the user, use the paused or rewinded video frame as the new current video frame, and display the candidate points of interest and the corresponding forward path identifiers on the current video frame; in response to the user selecting a second target point of interest from the candidate points of interest, track playback is performed based on the moving direction indicated by the corresponding forward path identifier, and the playback path of each video frame is recorded.

[0300] In an exemplary embodiment of the present disclosure, the display is configured to display a preview playback screen of the candidate point of interest in response to a user's preview instruction for the candidate point of interest. The preview playback screen can be used to preview the motion trajectory screen of the candidate point of interest within a preset time length.

[0301] In an exemplary embodiment of the present disclosure, the processor is configured to respond to a user's frame skip instruction, stop at a new video frame selected by the user, and use the new video frame as the current video frame. The display is configured to display candidate points of interest in the current video frame.

[0302] In an exemplary embodiment of the present disclosure, the display is configured to display a forward direction indicator of the camera on the current video frame, where the forward direction indicator is used to indicate the movement direction of the camera.

[0303] In an exemplary embodiment of the present disclosure, the display is configured to display a forward viewing angle mark on the current video frame.

[0304] In an exemplary embodiment of the present disclosure, the processor is configured to receive an instruction to play based on a forward perspective, play based on the forward perspective, and record the playback path.

[0305] In an exemplary embodiment of the present disclosure, the processor is configured to pause playback when it detects that the first target point of interest is lost during tracking playback based on the first target point of interest, use the video frame where the playback is paused as the new current video frame, display candidate points of interest in the new current video frame, and prompt the user to select a new point of interest.

[0306] In an exemplary embodiment of the present disclosure, the processor is configured to determine a viewing screen currently displayed by the display, where the viewing screen is a partial area in the current video frame; the display is configured to display at least one candidate point of interest in the viewing screen.

[0307] In an exemplary embodiment of the present disclosure, the processor is configured to identify at least one candidate interest point on the current video frame based on an interest point detection model; and the display is configured to display at least one candidate interest point on the current video frame.

[0308] In an exemplary embodiment of the present disclosure, the interest point detection model uses at least one of the following algorithms: a target detection algorithm, a saliency detection algorithm, and a video highlight detection algorithm.

[0309] In an exemplary embodiment of the present disclosure, the processor is configured to determine target video frames involved in editing, perform editing according to the recorded playback path of each target video frame, and generate an edited video, where the edited video is a planar video.

[0310] In an exemplary embodiment of the present disclosure, the current video frame is a video frame of a panoramic video.

[0311] Regarding the electronic device in the above embodiment, the specific manner in which the processor and the display perform operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0312] In one exemplary embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the computer program implements the steps of any of the above-described video editing methods. The computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, or the like.

[0313] In an exemplary embodiment, a computer program product is provided, including a computer program, which implements the steps of any of the above-mentioned video editing methods when executed by a processor.

[0314] With reference to FIG10 , a block diagram of an electronic device that can be used as the present disclosure will now be described. The electronic device includes a computing unit 1001 that can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1002 or a computer program loaded from a storage unit 1008 into a random access memory (RAM) 1003. Various programs and data required for the operation of the electronic device 1000 can also be stored in the RAM 1003. The computing unit 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0315] Multiple components in the electronic device 1000 are connected to the I / O interface 1005, including: an input unit 1006, an output unit 1007, a storage unit 1008, and a communication unit 1009. The input unit 906 can be any type of device that can input information to the electronic device 900. The input unit 1006 can receive input digital or character information and generate key signal input related to user settings and / or function control of the electronic device 1000, and can include but is not limited to a mouse, a keyboard, a touch screen, a trackpad, a trackball, a joystick, a microphone, and / or a remote control. The output unit 1007 can be any type of device that can present information, and can include but is not limited to a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 1008 can include but is not limited to a magnetic disk and an optical disk. The communication unit 1009 allows the electronic device 1000 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and may include but is not limited to a modem, a network card, an infrared communication device, a wireless communication transceiver and / or a chipset, such as a Bluetooth™ device, a WiFi device, a WiMax device, a cellular communication device and / or the like.

[0316] The computing unit 1001 can be a variety of general-purpose and / or specialized processing components with processing and computing capabilities. Some examples of the computing unit 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1001 performs the various methods and processes described above, such as the video editing method. For example, in some embodiments, the video editing method can be implemented as a computer software program that is tangibly contained in a machine-readable medium, such as the storage unit 1008. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 1000 via the ROM 1002 and / or the communication unit 1009. When the computer program is loaded into the RAM 1003 and executed by the computing unit 1001, one or more steps of the video editing method described above can be performed. Alternatively, in other embodiments, the computing unit 1001 can be configured to perform the video editing method by any other appropriate means (e.g., by means of firmware).

[0317] The electronic device 1000 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned video editing method.

[0318] Other embodiments of the present invention will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the invention being indicated by the following claims.

[0319] It should be understood that the present invention is not limited to the exact construction described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A video editing method, characterized in that: include: displaying a current video frame and determining a first target point of interest in response to a user selection; Playing based on the first target point of interest, and recording the playback path; The video is edited based on the playback path.

2. The method according to claim 1, characterized in that The displaying of the current video frame and determining the first target point of interest in response to the user's selection includes: At least one candidate interest point of the current video frame is displayed, and a first target interest point is determined in response to a user's selection.

3. The method according to claim 1, characterized in that The playing based on the first target point of interest and recording the playing path includes: Tracking playback is performed based on the first target point of interest, and the playback path is recorded.

4. The method according to claim 1, wherein The playing based on the first target point of interest and recording the playing path includes: Determine a playback sequence corresponding to the first target point of interest, play based on the playback sequence, and record a playback path.

5. The method according to claim 2, characterized in that The first target interest point is an interest point determined from the at least one candidate interest point or an interest point determined from other non-candidate interest points in the current video frame.

6. The method according to claim 1, characterized in that The method further comprises: Determining a key video frame in a subsequent video frame of the current video frame, and determining a second target interest point in the key video frame; The playing based on the first target point of interest and recording the playing path includes: Play the video based on the first target point of interest and the second target point of interest, and record the playback path.

7. The method according to claim 6, characterized in that The playing based on the first target point of interest and the second target point of interest and recording the playing path includes: A viewing angle of an intermediate video frame between the current video frame and the key video frame is determined, where the viewing angle is used to gradually transition from the first target interest point to the second target interest point.

8. The method according to claim 3, characterized in that The tracking and playing based on the first target point of interest and recording the playback path include: determining a viewing angle of each video frame according to the first target point of interest, wherein the viewing angle is used to gradually transition the first target point of interest to the center of a viewing screen and maintain the first target point of interest at the center of the viewing screen; Play each video frame based on the determined viewing angle and record the playback path.

9. The method according to claim 2, characterized in that The method further comprises: A forward path identifier corresponding to each candidate interest point is displayed on the current video frame.

10. The method according to claim 9, characterized in that The forward path identifier is used to indicate the moving direction corresponding to the point of interest at the next moment.

11. The method according to claim 9, characterized in that The forward path identifier is used to indicate the subsequent moving direction of the point of interest.

12. The method according to claim 9, characterized in that The displaying of the forward path identifier corresponding to each candidate point of interest includes: Based on the interest point tracking model, identifying the motion trajectory of each candidate interest point; Based on the motion trajectory, displaying a forward path identifier corresponding to each candidate point of interest; Wherein, the interest point tracking model uses a target tracking algorithm.

13. The method according to claim 3, characterized in that The tracking and playing based on the first target point of interest and recording the playback path include: Tracking playback is performed based on the moving direction indicated by the forward path identifier corresponding to the first target point of interest, and the playback path is recorded.

14. The method according to claim 3, characterized in that The tracking and playing based on the first target point of interest and recording the playback path include: During the tracking playback based on the first target interest point, upon receiving a pause or rewind instruction from the user, pausing or rewinding to a video frame selected by the user, using the paused or rewound video frame as a new current video frame, and displaying the candidate interest point on the current video frame; In response to the user selecting a second target point of interest from the candidate points of interest for tracking playback, a playback path of each video frame is recorded.

15. The method according to claim 3, characterized in that The tracking and playing based on the first target point of interest and recording the playback path include: During the tracking playback based on the first target interest point, upon receiving a pause or rewind instruction from the user, the video is paused or rewound to the video frame selected by the user, and the paused or rewound video frame is used as the new current video frame. Displaying candidate interest points and corresponding forward path identifiers on the current video frame; In response to a second target point of interest selected by the user from the candidate points of interest, tracking playback is performed based on the moving direction indicated by the corresponding forward path identifier, and the playback path of each video frame is recorded.

16. The method according to claim 2, characterized in that The method further comprises: In response to a user's preview instruction for a candidate point of interest, a preview playback screen of the candidate point of interest is displayed, and the preview playback screen can be used to preview a motion trajectory screen of the candidate point of interest within a preset time period.

17. The method according to claim 2, characterized in that The method further comprises: In response to a frame skipping instruction from the user, the video stops at a new video frame selected by the user, takes the new video frame as the current video frame, and displays candidate interest points in the current video frame.

18. The method according to claim 1, wherein The method further comprises: The camera's forward direction indicator is displayed on the current video frame, where the forward direction indicator is used to indicate the camera's moving direction.

19. The method according to claim 1, wherein The method further comprises: A forward viewing angle mark is displayed on the current video frame.

20. The method according to claim 19, characterized in that The method further comprises: Receive an instruction to play based on a forward perspective, play based on the forward perspective, and record the playback path.

21. The method according to claim 3, characterized in that The tracking and playing based on the first target point of interest and recording the playback path include: When tracking playback is performed based on the first target point of interest, playback is paused when it is detected that the first target point of interest is lost, and the video frame where the paused playback stops is used as the new current video frame. Candidate points of interest are displayed in the new current video frame and the user is prompted to select a new point of interest.

22. The method according to claim 2, characterized in that The displaying of at least one candidate interest point in the current video frame includes: Determining a viewing picture currently displayed on a display, where the viewing picture is a partial area in the current video frame; At least one of the candidate points of interest is displayed in the viewing screen.

23. The method according to claim 2, characterized in that The displaying of at least one candidate interest point on the current video frame includes: identifying at least one candidate interest point on the current video frame based on an interest point detection model; At least one candidate interest point on the current video frame is displayed.

24. The method according to claim 23, wherein The interest point detection model uses at least one of the following algorithms: target detection algorithm, saliency detection algorithm, and video highlight detection algorithm.

25. The method according to claim 1, wherein The video editing based on the playback path includes: The target video frames involved in the editing are determined, and the editing is performed according to the recorded playback path of each of the target video frames to generate an edited video, where the edited video is a planar video.

26. The method according to any one of claims 1 to 25, characterized in that: The current video frame is a video frame of a panoramic video.

27. A video editing device, characterized in that: include: a display module for displaying the current video frame and determining a first target point of interest in response to a user's selection; A processing module, configured to play the video based on the first target point of interest and record the playback path; The processing module is further configured to edit the video based on the playback path.

28. An electronic device comprising a memory, a processor and a display, wherein the memory stores a computer program, wherein: When the computer program instructions are executed by the processor, the display is configured to present a current video frame, and the processor is configured to determine a first target point of interest in response to a user selection; Playing based on the first target point of interest, and recording the playback path; The video is edited based on the playback path.

29. The electronic device according to claim 28, wherein The display is configured to present at least one candidate point of interest of a current video frame, and the processor is configured to determine a first target point of interest in response to a user's selection.

30. The electronic device according to claim 28, wherein The processor is configured to perform tracking playback based on the first target point of interest and record the playback path.

31. The electronic device according to claim 28, wherein The processor is configured to determine a playback sequence corresponding to the first target point of interest, play based on the playback sequence, and record a playback path.

32. The electronic device according to claim 29, wherein The first target interest point is an interest point determined from the at least one candidate interest point or an interest point determined from other non-candidate interest points in the current video frame.

33. The electronic device according to claim 28, wherein The processor is configured to determine a key video frame in a subsequent video frame of the current video frame, determine a second target point of interest in the key video frame; play based on the first target point of interest and the second target point of interest, and record a playback path.

34. The electronic device according to claim 33, wherein: The processor is configured to determine a viewing angle of an intermediate video frame between the current video frame and the key video frame, where the viewing angle is used to gradually transition from the first target point of interest to the second target point of interest.

35. The electronic device according to claim 30, wherein The processor is configured to determine a viewing angle of each video frame based on the first target point of interest, and the viewing angle is used to gradually transition the first target point of interest to the center of the viewing screen and maintain it in the center of the viewing screen; play based on the determined viewing angle of each video frame and record the playback path.

36. The electronic device according to claim 29, wherein The display is configured to display a forward path indicator corresponding to each candidate point of interest on the current video frame.

37. The electronic device according to claim 36, wherein: The forward path identifier is used to indicate the moving direction corresponding to the point of interest at the next moment.

38. The electronic device according to claim 36, wherein: The forward path identifier is used to indicate the subsequent moving direction of the point of interest.

39. The electronic device according to claim 36, wherein: The processor is configured to identify the motion trajectory of each candidate point of interest based on a point of interest tracking model; the processor is configured to display a forward path identifier corresponding to each candidate point of interest based on the motion trajectory; wherein the point of interest tracking model uses a target tracking algorithm.

40. The electronic device according to claim 30, wherein: The processor is configured to perform tracking playback based on the moving direction indicated by the forward path identifier corresponding to the first target point of interest, and record the playback path.

41. The electronic device according to claim 30, wherein The processor is configured to, when receiving a pause or rewind instruction from the user during tracking playback based on the first target point of interest, pause or rewind to the video frame selected by the user, use the paused or rewinded video frame as the new current video frame, and display the candidate points of interest on the current video frame; in response to the user selecting a second target point of interest from the candidate points of interest for tracking playback, record the playback path of each video frame.

42. The electronic device according to claim 30, wherein: The processor is configured to, when receiving a pause or rewind instruction from the user during tracking playback based on the first target point of interest, pause or rewind to the video frame selected by the user, use the paused or rewinded video frame as the new current video frame, and display the candidate points of interest and the corresponding forward path identifier on the current video frame; in response to the user selecting a second target point of interest from the candidate points of interest, track playback based on the moving direction indicated by the corresponding forward path identifier, and record the playback path of each video frame.

43. The electronic device according to claim 29, wherein The display is configured to display a preview playback screen of the candidate point of interest in response to a user's preview instruction for the candidate point of interest, and the preview playback screen can be used to preview a motion trajectory screen of the candidate point of interest within a preset time period.

44. The electronic device according to claim 29, wherein The processor is configured to, in response to a frame skip instruction from a user, stop at a new video frame selected by the user and use the new video frame as a current video frame. The display is configured to display candidate interest points in the current video frame.

45. The electronic device according to claim 28, wherein The display is configured to display a forward direction mark of the camera on the current video frame, where the forward direction mark is used to indicate a moving direction of the camera.

46. ​​The electronic device according to claim 28, wherein The display is configured to display a forward view mark on the current video frame.

47. The electronic device according to claim 46, characterized in that The processor is configured to receive an instruction to play based on a forward perspective, play based on the forward perspective, and record a playback path.

48. The electronic device according to claim 30, wherein: The processor is configured to pause playback when detecting that the first target point of interest is lost during tracking playback based on the first target point of interest, use the video frame where the paused playback stops as a new current video frame, display candidate points of interest in the new current video frame, and prompt the user to select a new point of interest.

49. The electronic device according to claim 29, wherein The processor is configured to determine a viewing screen currently displayed by the display, where the viewing screen is a partial area in the current video frame; and the display is configured to display at least one candidate point of interest in the viewing screen.

50. The electronic device according to claim 29, wherein The processor is configured to identify at least one candidate interest point on the current video frame based on an interest point detection model; and the display is configured to display the at least one candidate interest point on the current video frame.

51. The electronic device according to claim 50, characterized in that The interest point detection model uses at least one of the following algorithms: target detection algorithm, saliency detection algorithm, and video highlight detection algorithm.

52. The electronic device according to claim 28, wherein The processor is configured to determine target video frames involved in editing, perform editing according to the recorded playback path of each target video frame, and generate an edited video, where the edited video is a planar video.

53. The electronic device according to any one of claims 28 to 52, characterized in that: The current video frame is a video frame of a panoramic video.

54. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 26 are implemented.

Images