Pat type motion video collection generation method and device, equipment and storage medium

Abstract

The invention discloses a method, a device and equipment for generating a photo-type sports video collection and a storage medium, and belongs to the technical field of photo-type sports media. According to the invention, real-time monitoring is carried out on shooting sports activities to obtain corresponding shooting sports videos; based on a preset collection judgment strategy, identifying and screening the shooting type motion videos to obtain a collection mark set; performing video editing on the basis of each collection mark in the collection mark set to obtain the shooting sports video collection; and uploading the shooting exercise video collection to a cloud end and displaying the shooting exercise video collection through a video terminal. According to the invention, the method comprises the steps: firstly obtaining the shooting type sports video, then carrying out the collection mark generation of the shooting type sports video, finally carrying out the collection editing according to the collection mark, obtaining the corresponding collection video, uploading the collection video to a cloud end, and carrying out the display, thereby achieving the autonomous generation of the shooting type sports video collection. And rapid and convenient high-quality game watching review experience is provided for audiences.

Description

Technical Field

[0001] The present invention relates to the field of sports media technology, and in particular to a method, device, equipment and storage medium for generating sports video highlights. Background Art

[0002] In today's multimedia era, the production of video highlights of sports events has become an important means of communication. Highlights of sports events can be used to review and share the wonderful moments of the game.

[0003] However, traditional methods of producing highlights of sports events mostly rely on manual editing, but manual editing methods have the following defects: 1. Editors are required to view the video frame by frame, select the wonderful moments, and perform editing and splicing. This process is time-consuming and labor-intensive, especially when processing long sports game videos. The efficiency problem is more prominent, which not only affects the update speed of the highlights, but also limits the number of videos that can be processed; 2. Affected by the personal preferences and judgments of the editors, different editors may have different definitions of "wonderful", resulting in large differences in the content of the highlights, lack of objectivity and consistency, and this subjectivity may affect the audience's perception of the real situation of the game; 3. It depends on the professional skills and experience of the editors, so the quality of the highlights may fluctuate from editor to editor. If the editor is not skilled enough or works too hard, it may result in missing the real wonderful moments in the highlights, or the edited clips may not be accurate and smooth enough, thus affecting the audience's viewing experience.

[0004] In view of this, there is an urgent need for a sports video generation method that can generate wonderful game highlights by itself, so as to solve the technical problem of game highlights production.

[0005] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0006] The main purpose of the present invention is to provide a method, device, equipment and storage medium for generating a collection of sports videos, aiming to solve the technical problem of how to generate a collection video based on sports videos in the prior art.

[0007] To achieve the above object, the present invention provides a method for generating a sports video collection, the method comprising the following steps:

[0008] Real-time monitoring of clapping sports activities to obtain corresponding clapping sports videos;

[0009] Based on a preset highlight judgment strategy, the sports video is identified and screened to obtain a set of highlight marks;

[0010] Perform video editing based on each highlight mark in the highlight mark set to obtain the sports video highlights;

[0011] The sports video collection is uploaded to the cloud and displayed through a video terminal.

[0012] Optionally, the sports video is identified and screened based on a preset highlight judgment strategy to obtain a set of highlight marks, specifically including:

[0013] Performing video preprocessing on the clapping motion video to obtain a processed clapping motion video;

[0014] Extracting data from the processed slap-type motion video to obtain a corresponding image data set;

[0015] The image data set is identified and screened to obtain a collection mark set.

[0016] Optionally, data extraction is performed on the processed slap-type motion video to obtain a corresponding image data set, specifically including:

[0017] Extracting image data from the processed slap-type motion video to obtain an initial image data set;

[0018] Performing data cleaning on the initial image data set to obtain a cleaned image data set;

[0019] The cleaned image data set is sorted based on an interpolation processing strategy to obtain a corresponding image data set.

[0020] Optionally, identifying and screening the image data set to obtain a collection mark set specifically includes:

[0021] Traversing and filtering the image data set to obtain a collection start frame set;

[0022] Based on the collection start frame set, the image data set is traversed and filtered to obtain a collection end frame set;

[0023] The highlight mark set is obtained by integrating and screening based on the highlight start frame set and the highlight end frame set.

[0024] Optionally, the image data set is traversed and filtered to obtain a collection start frame set, specifically including:

[0025] Performing a traversal check on the image data set based on a time sequence rule to obtain a corresponding image data information set;

[0026] Traversing and filtering the image data set based on the image data information set to obtain a pre-selected highlight start frame set;

[0027] Data backtracking is performed based on the pre-selected highlight start frame set to obtain the highlight start frame set.

[0028] Optionally, traversing and filtering the image data set based on the highlight start frame set to obtain a highlight end frame set specifically includes:

[0029] Based on the collection start frame set, the image data set is traversed and filtered to obtain a pre-selected collection end frame set;

[0030] Data deduction is performed based on the pre-selected highlight end frame set to obtain the highlight end frame set.

[0031] Optionally, integrating and screening the highlight start frame set and the highlight end frame set to obtain the highlight mark set specifically includes:

[0032] Based on the integration of the highlight start frame set and the highlight end frame set, a pre-selected highlight mark set is obtained;

[0033] The pre-selected highlights corresponding to each pre-selected highlight mark in the pre-selected highlight mark set are screened based on a preset highlight judgment rule to obtain the highlight mark set.

[0034] In addition, to achieve the above-mentioned purpose, the present invention further proposes a device for generating a collection of sports video highlights, the device comprising:

[0035] Motion monitoring module: real-time monitoring of sports activities and obtaining corresponding sports videos;

[0036] Highlight screening module: based on a preset highlight judgment strategy, the sports video is identified and screened to obtain a set of highlight marks;

[0037] Highlight generation module: based on each highlight mark in the highlight mark set, video editing is performed to obtain the sports video highlights;

[0038] Highlights display module: upload the sports video highlights to the cloud and display them through the video terminal.

[0039] In addition, to achieve the above-mentioned purpose, the present invention also proposes a device for generating a video highlight of a beat-type sports, and the video highlight generating device of the beat-type sports comprises: a memory, a processor, and a beat-type sports video highlight generating program stored in the memory and executable on the processor, and the beat-type sports video highlight generating program is configured to implement the steps of the method for generating a video highlight of a beat-type sports as described above.

[0040] In addition, to achieve the above-mentioned purpose, the present invention also proposes a computer-readable storage medium storing a computer program, wherein a program for generating a video highlight of a beat-type sports is stored on the storage medium, and when the program for generating a video highlight of a beat-type sports is executed by a processor, the steps of the method for generating a video highlight of a beat-type sports are implemented as described above.

[0041] The present invention monitors the slapping sports activities in real time to obtain corresponding slapping sports videos; identifies and screens the slapping sports videos based on a preset highlight judgment strategy to obtain a highlight mark set; performs video editing based on each highlight mark in the highlight mark set to obtain the slapping sports video highlights; uploads the slapping sports video highlights to the cloud and displays them through a video terminal. The present invention first obtains slapping sports videos, then generates highlight marks for the slapping sports videos, and finally performs highlight editing based on the highlight marks to obtain corresponding highlight videos for uploading to the cloud and displaying them, thereby realizing the autonomous generation of slapping sports video highlights and providing viewers with a fast and convenient high-quality viewing and reviewing experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a structural schematic diagram of a device for generating a collection of sports video in a hardware operating environment according to an embodiment of the present invention;

[0043] Figure 2 It is a flowchart of a first embodiment of a method for generating a collection of sports video clips according to the present invention;

[0044] Figure 3 It is a flowchart of a second embodiment of the method for generating a collection of sports video clips according to the present invention;

[0045] Figure 4 It is a flowchart of a third embodiment of the method for generating a collection of sports video clips according to the present invention;

[0046] Figure 5 It is a flowchart of a fourth embodiment of the method for generating a collection of sports video clips according to the present invention;

[0047] Figure 6 It is a structural block diagram of the first embodiment of the device for generating a collection of sports video clips according to the present invention.

[0048] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0049] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0050] Reference Figure 1 , Figure 1 The present invention is a schematic diagram of the structure of a device for generating a collection of sports video in a hardware operating environment according to an embodiment of the present invention.

[0051] like Figure 1 As shown, the device for generating highlights of sports video may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0052] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the device for generating a video highlight of sports shots, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0053] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a shooting sports video highlights generation program.

[0054] exist Figure 1 In the beat-type sports video highlight generation device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the beat-type sports video highlight generation device of the present invention can be set in the beat-type sports video highlight generation device, and the beat-type sports video highlight generation device calls the beat-type sports video highlight generation program stored in the memory 1005 through the processor 1001, and executes the beat-type sports video highlight generation method provided in the embodiment of the present invention.

[0055] The embodiment of the present invention provides a method for generating a video collection of sports shots, referring to Figure 2 , Figure 2 The figure is a flow chart of a first embodiment of a method for generating a collection of sports video clips according to the present invention.

[0056] In this embodiment, the method for generating a sports video collection includes the following steps:

[0057] Step S10: real-time monitoring of the clapping sports activities to obtain corresponding clapping sports videos;

[0058] It should be noted that, in a specific implementation, the technical solution of this embodiment can monitor racket sports (a sport that requires the use of a racket, such as table tennis, badminton, and tennis, etc.) in real time. The specific real-time monitoring method can be to monitor racket sports in real time by installing monitoring equipment on the racket sports field according to monitoring requirements, so as to obtain racket sports videos with the best viewing angle and the largest observation range. For example, a high-definition camera can be set above the central axis of one side of the badminton court to monitor badminton activities in real time.

[0059] It should also be noted that, in a specific implementation, the equipment for monitoring the filming motion process may be specifically composed of multiple camera devices. For example, multiple monitoring cameras may be arranged on a filming sports field to obtain multi-angle filming sports videos, thereby meeting the monitoring needs of regular shots and close-up shots in a filming sports video collection.

[0060] Step S20: identifying and screening the sports videos based on a preset highlight judgment strategy to obtain a set of highlight marks;

[0061] It should be noted that, in the specific implementation, the preset highlight judgment strategy is essentially a preset highlight editing judgment standard for racket sports videos, which can specifically select part or all of the racket sports data from the racket sports data such as the length of the racket sports rounds, the number of swings of the racket athletes, the speed of the ball during the racket sports, the movement of the racket athletes, the changes in the amplitude of the racket athletes' postures, and the jumping height of the racket athletes as the judgment standard. Only when the racket sports data in a certain unit time of the video clip in the racket sports video meets the preset highlight editing judgment standard, the video of the unit time will be judged as a racket sports video highlight, and the starting frame image and the ending frame image of the racket sports video can be regarded as the highlight mark of the highlight.

[0062] It should also be noted that, in the specific implementation, the preset highlight judgment strategy exists in the artificial intelligence model, and its recognition and screening process of shot-type sports videos is essentially an autonomous decision-making process of the artificial intelligence model.

[0063] Step S30: performing video editing based on each highlight mark in the highlight mark set to obtain a sports video highlight;

[0064] It can be understood that, in a specific implementation, the highlight mark is essentially a clip mark of a shot-type motion video clip that is determined to be a shot-type motion highlight, specifically including a highlight start frame and a highlight end frame.

[0065] Step S40: Upload the sports video highlights to the cloud and display them through the video terminal.

[0066] It should be noted that, in the specific implementation, the final shooting sports video highlights will be uploaded to the cloud and the display screen background of the shooting sports venue through communication equipment, so that the shooting sports video highlights can be updated and played in real time on the big screen in the shooting sports venue.

[0067] It should also be noted that, in a specific implementation, if there is multi-perspective monitoring, the final generated sports video highlights can be composed of normal perspective video highlights and close-up perspective video highlights, and when they are finally displayed on the video terminal, the normal perspective video highlights and close-up perspective video highlights can be played in combination (for example, the close-up video highlights can be played after the normal perspective video highlights are played; or after a certain normal perspective video is played, the close-up perspective video corresponding to the time period can be played).

[0068] This embodiment monitors the slapping sports activities in real time to obtain corresponding slapping sports videos; identifies and screens the slapping sports videos based on a preset highlight judgment strategy to obtain a set of highlight marks; performs video editing based on each highlight mark in the set of highlight marks to obtain the slapping sports video highlights; uploads the slapping sports video highlights to the cloud and displays them through a video terminal. This embodiment first obtains slapping sports videos, then generates highlight marks for the slapping sports videos, and finally performs highlight editing based on the highlight marks to obtain corresponding highlight videos for uploading to the cloud and displaying them, thereby realizing the autonomous generation of slapping sports video highlights and providing viewers with a fast and convenient high-quality viewing and review experience.

[0069] refer to Figure 3 , Figure 3 The figure is a flow chart of a second embodiment of a method for generating a collection of sports video clips according to the present invention.

[0070] Based on the above first embodiment, in this embodiment, step S20 specifically includes:

[0071] Step S21: performing video preprocessing on the slapping motion video to obtain a processed slapping motion video;

[0072] It can be understood that, in a specific implementation, video preprocessing of the beat-type motion video is essentially performing video processing operations such as denoising, dedistortion, contrast adjustment, color correction, and audio overlay on the beat-type motion video frames, thereby improving the accuracy and completeness of subsequent video processing.

[0073] Step S22: extracting data from the processed sports video to obtain a corresponding image data set;

[0074] It should be noted that, in a specific implementation, extracting data from the processed beat-type motion video essentially means extracting video frames from the processed beat-type motion video, and then using the extracted video frames as an image data set. For example, video frames may be extracted from the processed beat-type motion video at an extraction interval of 1 second, thereby obtaining a corresponding image data set.

[0075] It should also be noted that, in the specific implementation, the processing process of the artificial intelligence model on the racket sports video is essentially to obtain the features of the racket sports video, that is, to obtain the corresponding information through the artificial intelligence model. Specifically, the information that can be obtained is the position information of the person, the position information of the badminton, and the racket motion trajectory information of the person and the ball. The artificial intelligence model will also determine the position and specific point of the venue in the video, and the position of the net, so as to realize the image data processing of the racket sports video.

[0076] It can be understood that in a specific implementation, the processed close-up motion video also includes close-up motion video monitored by a pre-processed close-up monitoring camera. Therefore, data extraction of the processed close-up motion video also includes extracting video frames from the processed close-up motion video, and then taking the extracted close-up video frames and other extracted ordinary video frames together as an image data set.

[0077] Step S23: Identify and screen the image data set to obtain a collection mark set.

[0078] It should be noted that, in the specific implementation, the identification and screening process is essentially to identify and judge all the extracted video frames (image data sets), so as to screen out the video frames that meet the highlight mark standards, and integrate all the screened video frames into a highlight mark set.

[0079] This embodiment first performs video preprocessing on the beat-type sports video, then extracts data from the processed beat-type sports video, and finally identifies and screens the image data set to obtain a highlight mark set, which provides a reference for the subsequent editing of the beat-type sports video highlights.

[0080] refer to Figure 4 , Figure 4The figure is a flow chart of a third embodiment of a method for generating a collection of sports video clips according to the present invention.

[0081] Based on the above second embodiment, in this embodiment, step S22 specifically includes:

[0082] Step S221: extracting image data from the processed sports video to obtain an initial image data set;

[0083] It should be noted that, in a specific implementation, extracting image data from the processed beat-type motion video essentially means extracting video frames in an orderly manner from the processed beat-type motion video according to a preset video frame extraction rule, and then using the extracted video frames as the initial image data set. For example, video frames may be extracted starting from the first frame of the processed beat-type motion video at an extraction interval of 0.1 seconds, and then all extracted video frames may be used as the initial image data set.

[0084] Step S222: performing data cleaning on the initial image data set to obtain a cleaned image data set;

[0085] It should be noted that, in a specific implementation, data cleaning of the initial image data set is essentially to delete abnormal ball data and duplicate ball data in the extracted video frame (initial image data set), thereby obtaining a cleaned image data set.

[0086] It should also be noted that, in the specific process, the specific process of deleting duplicate ball data is to screen each video frame in the initial image data set, so as to screen out the balls with the same position coordinates and delete other video frames with the balls with the same coordinates after retaining one video frame; the specific process of deleting abnormal ball data is to screen each video frame in the initial image data set, if a ball appears in a certain video frame but does not appear in a specified time period before and after the video frame, then the ball is considered to be an abnormal ball that appears suddenly, and the video frame corresponding to the abnormal ball needs to be deleted.

[0087] Step S223: sorting the cleaned image data set based on the interpolation processing strategy to obtain a corresponding image data set.

[0088] It should be noted that in the specific implementation, if the speed of the ball's racket motion is too fast in a certain round of playing, the frame set of the racket motion trajectory of the ball in that round may not be able to fully and clearly reflect the racket motion of the ball. Therefore, it is necessary to use an interpolation processing strategy to supplement the extraction of the video frames of the ball, so as to fully reflect the racket motion trajectory of the ball.

[0089] It should also be noted that in the specific implementation, the interpolation processing strategy specifically includes the first interpolation processing strategy and the second interpolation processing strategy; wherein, the first interpolation processing strategy means that if a ball-racing motion trajectory in the video frame coordinate system (the direction of the video frame parallel to the ground is the X axis, and the direction perpendicular to the ground is the Y axis) reflects a downward trend of the ball-racing motion in several consecutive video frames (moving toward the ground, the Y coordinate continuously decreases), it is necessary to start from the actual position of the ball recorded last (actual video frame) and predict the position of the ball forward until the next actual existing ball position is encountered. The specific implementation process includes: 1. Record the last valid position: before the ball disappears, record the last known position of the ball (X, Y) and the corresponding frame number; 2. Determine the interpolation starting point: the starting point is the last valid position. 3. Calculate the velocity vector: Based on the position change of the ball in the last few frames, calculate the velocity vector of the ball (this can be achieved by calculating the position difference (ΔX, ΔY) of the ball between two adjacent frames); 4. Perform interpolation: Starting from the starting point, predict the position of the ball frame by frame according to the calculated velocity vector until the next actual recorded position of the ball is reached or the maximum interpolation frame limit (for example, 100 frames) is reached; 5. Adjust the interpolation end point: If the next actual position is not found within the fixed frame, the interpolation process terminates at the position of a certain frame; if the next actual position is found in less than the fixed frame, the interpolation process terminates at the actual position (actual video frame).

[0090] The second interpolation processing strategy is that if the trajectory of a ball-racing motion is upward (moving toward the ground, the Y coordinate increases continuously) in a video frame coordinate system (the direction of the video frame parallel to the ground is the X axis, and the direction perpendicular to the ground is the Y axis) reflected by several consecutive video frames, the position of the ball needs to be predicted backwards from the actual position of the ball recorded last (actual video frame) until the last actual existing ball position is encountered. The specific implementation process steps are basically the same as the first interpolation processing strategy, but in the opposite direction.

[0091] It is understandable that, in a specific implementation, the implementation process of the interpolation processing strategy also includes restrictions such as linear interpolation, quadratic interpolation, smoothing, and boundary conditions. Linear interpolation refers to predicting the intermediate position by connecting two known points with a straight line; quadratic interpolation needs to consider changes in acceleration to more accurately predict the position; smoothing is specifically to prevent the jumping feeling caused by interpolation, that is, smoothing algorithms such as moving averages or filters can be used to make the interpolation processing trajectory more natural; boundary conditions specifically refer to ensuring that when processing interpolation, the interpolation result does not exceed a reasonable range, for example, the position of the ball should not exceed the screen boundary.

[0092] This embodiment first extracts image data from the processed sports video, then cleans the obtained initial image data set, and finally organizes the cleaned image data set based on the interpolation processing strategy to obtain the corresponding image data set, which provides a basis for the subsequent generation of highlight marks.

[0093] refer to Figure 5 , Figure 5 The figure is a flow chart of a third embodiment of a method for generating a collection of sports video clips according to the present invention.

[0094] Based on the above second embodiment, in this embodiment, step S23 specifically includes:

[0095] Step S231: traverse and filter the image data set to obtain a collection start frame set;

[0096] It should be noted that, in a specific implementation, highlights of exciting rounds usually refer to those shot sports video clips that include high-speed hitting, multi-shot duels and high-level competitive performances, among which the highlights start frame refers to the first video frame at the beginning of such video clips; the highlights end frame refers to the last video frame at the end of such video clips.

[0097] It should also be noted that, in a specific implementation, the process of traversing and filtering the image data set to obtain the collection start frame set specifically refers to traversing from the first frame of the image data set in time sequence (the generation time sequence corresponding to the video frame), performing a collection start frame judgment on each video frame, and then integrating the video frames that meet the requirements to obtain the collection start frame set. Among them, the collection start frame judgment specifically refers to checking whether there is a sphere in the video frames within a certain range (for example, the next 100 frames or 200 frames) thereafter for each frame. If the number of times the sphere exists in the range of frames exceeds the judgment threshold number (for example, 10 times or 20 times), then mark the current frame as a "potential start frame"; then trace back a preset number of frames (for example, 100 frames or 200 frames) from the "potential start frame", and determine this traced frame as the real collection start frame of the beat type motion round; traverse and process all other video frames in the image data set in turn, and summarize all the obtained collection start frames into the collection start frame set.

[0098] It is understandable that in a specific implementation, if it is in a multi-view monitoring situation, it is necessary to consider the data of multiple viewpoints when confirming the start frame and the end frame. For example, if a potential start frame is found in a certain viewpoint, it is necessary to confirm whether the frame also meets the conditions in other viewpoints, that is, for each frame of each viewpoint video, check whether there is a sphere in the video frames within a certain range (for example, the next 100 frames or 200 frames) thereafter. If the number of times the sphere exists in the range of frames exceeds the judgment threshold number (for example, 10 times or 20 times), then mark the current frame as a "potential start frame"; then trace back a preset number of frames (for example, 100 frames or 200 frames) from the "potential start frame", and determine this traced frame as the real highlight start frame of the motion round of this type; traverse and process all other video frames in the image data set (including close-up view videos and ordinary view videos) in turn, and summarize all the obtained highlight start frames into a highlight start frame set.

[0099] Step S232: traverse and filter the image data set based on the highlight start frame set to obtain the highlight end frame set;

[0100] It should be noted that, in a specific implementation, the process of traversing and filtering the image data set based on the highlight start frame set to obtain the highlight end frame set specifically refers to traversing the image data set starting from the first highlight start frame confirmed in the highlight start frame set, checking and judging whether there is a sphere in each frame and video frames within a certain range (for example, the next 100 frames or 200 frames); if the number of times the sphere exists in the frames in the range is less than the judgment threshold number (for example, 10 times or 20 times), the current frame is marked as a "potential end frame"; then, the preset number of frames (for example, 100 frames or 200 frames) is advanced backward from the "potential end frame", and the advanced video frame is determined as the true highlight end frame of the beat type motion round; all other video frames in the image data set are traversed and processed in turn, and all the obtained highlight end frames are summarized as the highlight end frame set.

[0101] Step S233: integrating and filtering based on the highlight start frame set and the highlight end frame set to obtain a highlight mark set.

[0102] It should be noted that, in a specific implementation, the process of integration based on the highlight start frame set and the highlight end frame set specifically refers to matching the highlight start frame set and the highlight end frame one by one according to the timing rules, and each highlight start frame corresponds to the corresponding highlight end frame, and the two together constitute a highlight mark.

[0103] It should also be noted that in the specific implementation, all the integrated and generated highlight marks need to be further screened to obtain the highlight mark set. The screening process is essentially a comprehensive comparative evaluation. Specifically, the comprehensive comparative evaluation is to compare and evaluate the highlight segment content data corresponding to each highlight mark based on factors such as the length of the round of racket sports, the number of rackets, the ball speed, the movement of the athlete, and the change in the amplitude of the athlete's posture (such as the jump height), and then select the highlight marks corresponding to the highlight segments that meet all the evaluation factors or are closer to meeting the evaluation factors as the final highlight mark set.

[0104] It can be understood that, in a specific implementation, the highlight mark set also includes a close-up highlight mark generated based on a close-up perspective video, wherein any close-up highlight mark is composed of a corresponding close-up highlight start frame and a corresponding close-up highlight end frame.

[0105] This embodiment first traverses and filters the image data set, and then traverses and filters the image data set based on the obtained highlight start frame set to obtain a highlight end frame set, and finally integrates and filters the highlight start frame set and the highlight end frame set to obtain a highlight mark set, which provides a basis for subsequent video highlight editing.

[0106] Furthermore, the image data set is traversed and screened to obtain a highlight start frame set, specifically including: traversing and checking the image data set based on a timing rule to obtain a corresponding image data information set; traversing and screening the image data set based on the image data information set to obtain a pre-selected highlight start frame set; and performing data backtracking based on the pre-selected highlight start frame set to obtain a highlight start frame set.

[0107] It can be understood that, in a specific implementation, the timing rule is specifically the order in which each video frame in the image data set is generated (ie, the time order in a motion video).

[0108] Furthermore, the image data set is traversed and filtered based on the highlight start frame set to obtain a highlight end frame set, specifically including: traversing and filtering the image data set based on the highlight start frame set to obtain a pre-selected highlight end frame set; and performing data deduction based on the pre-selected highlight end frame set to obtain a highlight end frame set.

[0109] Furthermore, the highlight mark set is obtained by integrating and screening based on the highlight start frame set and the highlight end frame set, specifically including: obtaining a pre-selected highlight mark set by integrating based on the highlight start frame set and the highlight end frame set; and obtaining a highlight mark set by screening the pre-selected highlights corresponding to each pre-selected highlight mark in the pre-selected highlight mark set based on a preset highlight judgment rule.

[0110] In addition, an embodiment of the present invention also proposes a computer-readable storage medium storing a computer program, wherein a program for generating a video highlight of a beat-type sports is stored on the storage medium, and when the program for generating a video highlight of a beat-type sports is executed by a processor, the steps of the method for generating a video highlight of a beat-type sports are implemented as described above.

[0111] Since the storage medium adopts all the technical solutions of all the above embodiments, it has at least all the beneficial effects brought by the technical solutions of the above embodiments, which will not be described one by one here.

[0112] Reference Figure 6 , Figure 6 It is a structural block diagram of the first embodiment of the device for generating a collection of sports video clips according to the present invention.

[0113] like Figure 6 As shown, the apparatus for generating a collection of sports video highlights provided by an embodiment of the present invention includes:

[0114] Motion monitoring module 10: monitors the sports activities of the beat type in real time and obtains the corresponding sports videos of the beat type;

[0115] Highlight screening module 20: identifying and screening the sports videos based on a preset highlight judgment strategy to obtain a set of highlight marks;

[0116] Highlight generation module 30: based on each highlight mark in the highlight mark set, video editing is performed to obtain a sports video highlight;

[0117] Highlights display module 40: Upload sports video highlights to the cloud and display them through the video terminal.

[0118] This embodiment monitors the slapping sports activities in real time to obtain corresponding slapping sports videos; identifies and screens the slapping sports videos based on a preset highlight judgment strategy to obtain a set of highlight marks; performs video editing based on each highlight mark in the set of highlight marks to obtain the slapping sports video highlights; uploads the slapping sports video highlights to the cloud and displays them through a video terminal. This embodiment first obtains slapping sports videos, then generates highlight marks for the slapping sports videos, and finally performs highlight editing based on the highlight marks to obtain corresponding highlight videos for uploading to the cloud and displaying them, thereby realizing the autonomous generation of slapping sports video highlights and providing viewers with a fast and convenient high-quality viewing and review experience.

[0119] It should be understood that the above is only an example and does not constitute any limitation on the technical solution of the present invention. In specific applications, technicians in this field can make settings as needed, and the present invention does not limit this.

[0120] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of them according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0121] In addition, for technical details not fully described in this embodiment, reference can be made to the method for generating a sports video highlight provided in any embodiment of the present invention, and will not be repeated here.

[0122] In addition, it should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0123] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0124] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory (ROM) / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0125] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for generating a collection of sports video highlights, characterized in that: include: Real-time monitoring of clapping sports activities to obtain corresponding clapping sports videos; Based on a preset highlight judgment strategy, the sports video is identified and screened to obtain a set of highlight marks; Perform video editing based on each highlight mark in the highlight mark set to obtain the sports video highlights; The sports video collection is uploaded to the cloud and displayed through a video terminal.

2. The method for generating a collection of sports video highlights according to claim 1, characterized in that: Based on the preset highlight judgment strategy, the sports video of the shooting type is identified and screened to obtain a set of highlight marks, which specifically includes: Performing video preprocessing on the clapping motion video to obtain a processed clapping motion video; Extracting data from the processed slap-type motion video to obtain a corresponding image data set; The image data set is identified and screened to obtain a collection mark set.

3. The method for generating a collection of sports video highlights according to claim 2, characterized in that: Data extraction is performed on the processed slap-type motion video to obtain a corresponding image data set, specifically including: Extracting image data from the processed slap-type motion video to obtain an initial image data set; Performing data cleaning on the initial image data set to obtain a cleaned image data set; The cleaned image data set is sorted based on an interpolation processing strategy to obtain a corresponding image data set.

4. The method for generating a collection of sports video highlights according to claim 2, characterized in that: The image data set is identified and screened to obtain a collection mark set, which specifically includes: Traversing and filtering the image data set to obtain a collection start frame set; Based on the collection start frame set, the image data set is traversed and filtered to obtain a collection end frame set; The highlight mark set is obtained by integrating and screening based on the highlight start frame set and the highlight end frame set.

5. The method for generating a collection of sports video highlights according to claim 4, characterized in that: The image data set is traversed and filtered to obtain a collection start frame set, which specifically includes: Performing a traversal check on the image data set based on a time sequence rule to obtain a corresponding image data information set; Traversing and filtering the image data set based on the image data information set to obtain a pre-selected highlight start frame set; Data backtracking is performed based on the pre-selected highlight start frame set to obtain the highlight start frame set.

6. The method for generating a collection of sports video highlights according to claim 4, characterized in that: The image data set is traversed and filtered based on the highlight start frame set to obtain a highlight end frame set, specifically including: Based on the collection start frame set, the image data set is traversed and filtered to obtain a pre-selected collection end frame set; Data deduction is performed based on the pre-selected highlight end frame set to obtain the highlight end frame set.

7. The method for generating a collection of sports video highlights according to claim 4, characterized in that: Based on the collection start frame set and the collection end frame set, integration and screening are performed to obtain the collection mark set, which specifically includes: Based on the integration of the highlight start frame set and the highlight end frame set, a pre-selected highlight mark set is obtained; The pre-selected highlights corresponding to each pre-selected highlight mark in the pre-selected highlight mark set are screened based on a preset highlight judgment rule to obtain the highlight mark set.

8. A device for generating a collection of sports video clips, characterized in that: The device for generating a collection of sports video clips comprises: Motion monitoring module: real-time monitoring of sports activities and obtaining corresponding sports videos; Highlight screening module: based on a preset highlight judgment strategy, the sports video is identified and screened to obtain a set of highlight marks; Highlight generation module: based on each highlight mark in the highlight mark set, video editing is performed to obtain the sports video highlights; Highlights display module: upload the sports video highlights to the cloud and display them through the video terminal.

9. A device for generating a collection of sports video clips, characterized in that: The device for generating a video highlight of a beat-type motion comprises: a memory, a processor, and a video highlight generation program of a beat-type motion stored in the memory and executable on the processor, wherein the video highlight generation program of a beat-type motion is configured to implement a video highlight generation method of a beat-type motion according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps in the method for generating a video highlight of a sports shooting type as described in any one of claims 1 to 7 can be implemented.

Images