Cloud video content management method and system and storage medium

Abstract

The invention provides a cloud video content management method and system and a storage medium. The method comprises the following steps: drawing a comment popularity curve by taking comment time as a horizontal coordinate and a comment number as a vertical coordinate; performing video editing on the cloud video according to the comment popularity curve to obtain a hot video; classifying the video frames according to the video scenes to obtain video scene frame sets, and obtaining user fixation point coordinates in the same video scene frame set; determining an interested object according to the user fixation point coordinate, determining an uninterested object according to an object position area of the interested object, and performing image blurring on the uninterested object to obtain a hotspot optimization video; performing semantic recognition on the comment information to obtain comment semantics, and determining a video type label according to the comment semantics; and performing content type marking on the hotspot optimized video according to the video type label. According to the invention, the hotspot video in the cloud video can be automatically edited, and the cloud video content processing efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of cloud video technology, and in particular to a cloud video content management method, system and storage medium. Background Art

[0002] Cloud video refers to a video network platform service based on the cloud computing business model. On the cloud platform, all video suppliers, agents, planning service providers, producers, industry associations, management agencies, industry media, legal structures, etc. are integrated into a resource pool. Each resource is displayed and interacted with each other, and communication is carried out on demand to reach an agreement, thereby reducing costs and improving efficiency.

[0003] In the existing cloud video content management process, cloud video content is generally processed manually, which leads to low efficiency in cloud video content processing. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide a cloud video content management method, system and storage medium to solve the problem of low efficiency in cloud video content processing in the prior art.

[0005] The embodiment of the present invention is implemented as follows: a cloud video content management method, the method comprising: Obtain comment information in the cloud video, and draw a comment heat curve with the comment time in the comment information as the horizontal axis and the number of comments at the comment time as the vertical axis; Performing video editing on the cloud video according to the comment heat curve to obtain a hot video, and obtaining a video scene of a video frame in the hot video; Classifying the video frames according to the video scene to obtain a video scene frame set, and obtaining the coordinates of the user's gaze point of the video frame in the same video scene frame set; Determine an object of interest according to the coordinates of the user's gaze point, and obtain an object position area of ​​the object of interest in the video frame; Determine the uninteresting objects in the video scene according to the object position area, and blur the image of the uninteresting objects in the video scene in the hotspot video to obtain a hotspot optimized video; Performing semantic recognition on the comment information to obtain comment semantics, and determining a video type label according to the comment semantics; The hotspot optimized video is marked with a content type according to the video type tag.

[0006] Preferably, determining the uninteresting object in the video scene according to the object position area includes: Determine a non-attention area in the video frame according to the object position area, and set the video object in the non-attention area as a candidate object; If the candidate object exists in all the video frames in the same video scene frame set, the candidate object is determined as the uninteresting object in the video scene corresponding to the video scene frame set.

[0007] Preferably, the cloud video is edited according to the comment heat curve to obtain a hot video, including: If the number of comments at any comment time in the comment heat curve is greater than the number threshold, the comment time is determined as a hot time; A hotspot duration range is generated according to a preset duration and the hotspot time, and the cloud video is edited according to the hotspot duration range to obtain the hotspot video.

[0008] Preferably, determining the object of interest according to the coordinates of the user's gaze point includes: Acquire the area range of the video object in the video frame, and determine the number of gazes of the video object according to the user's gaze point coordinates and the area range; If the number of gazes on any of the video objects is greater than a number threshold, the video object is determined as the object of interest.

[0009] Preferably, before obtaining the video scene of the video frame in the hot video, the method further includes: Obtain like information in the cloud video, and draw a like heat curve with the like time in the like information as the horizontal axis and the number of likes at the like time as the vertical axis; The cloud video is edited according to the likes heat curve to obtain the hot video.

[0010] Another object of an embodiment of the present invention is to provide a cloud video content management system, the system comprising: A heat analysis module is used to obtain comment information in the cloud video, and draw a comment heat curve with the comment time in the comment information as the horizontal axis and the number of comments at the comment time as the vertical axis; A video editing module, used to edit the cloud video according to the comment heat curve to obtain a hot video, and obtain a video scene of a video frame in the hot video; A video classification module, used to classify the video frames according to the video scene to obtain a video scene frame set, and obtain the coordinates of the user's gaze point of the video frame in the same video scene frame set; An object determination module, used to determine an object of interest according to the coordinates of the user's gaze point, and obtain an object position area of ​​the object of interest in the video frame; An image blur module, used for determining uninteresting objects in the video scene according to the object position area, and blurring the images of the uninteresting objects in the video scene in the hotspot video to obtain a hotspot optimized video; The content marking module is used to perform semantic recognition on the comment information, obtain comment semantics, and determine the video type label according to the comment semantics; and mark the content type of the hot spot optimized video according to the video type label.

[0011] Preferably, the image blur module is further used for: Determine a non-attention area in the video frame according to the object position area, and set the video object in the non-attention area as a candidate object; If the candidate object exists in all the video frames in the same video scene frame set, the candidate object is determined as the uninteresting object in the video scene corresponding to the video scene frame set.

[0012] Preferably, the video editing module is also used for: If the number of comments at any comment time in the comment heat curve is greater than the number threshold, the comment time is determined as a hot time; A hotspot duration range is generated according to a preset duration and the hotspot time, and the cloud video is edited according to the hotspot duration range to obtain the hotspot video.

[0013] Preferably, the object determination module is further used for: Acquire the area range of the video object in the video frame, and determine the number of gazes of the video object according to the user's gaze point coordinates and the area range; If the number of gazes on any of the video objects is greater than a number threshold, the video object is determined as the object of interest.

[0014] The embodiments of the present invention can effectively draw a comment heat curve based on the comment time and the number of comments, and can automatically edit hot videos in the cloud video based on the comment heat curve. By acquiring the video scene of the video frame, the video frame can be effectively classified to obtain a video scene frame set. By acquiring the coordinates of the user's gaze point, the object of interest can be automatically determined. Based on the object position area of ​​the object of interest in the video frame, the uninteresting objects in the video scene can be automatically determined. By blurring the image of the uninteresting object, the video optimization effect is effectively achieved for the hot video. By performing semantic recognition on the comment information, the comment semantics can be effectively obtained. Based on the comment semantics, the video type label can be effectively determined. Based on the video type label, the content type of the hot optimization video can be automatically marked. There is no need to manually edit the hot video and mark the content type, thereby improving the cloud video content processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a flow chart of a cloud video content management method provided by a first embodiment of the present invention; Figure 2 is a schematic diagram of the structure of a cloud video content management system provided by a second embodiment of the present invention; Figure 3 It is a schematic diagram of the structure of a terminal device provided in the third embodiment of the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0017] In order to illustrate the technical solution of the present invention, a specific embodiment is provided below for illustration.

[0018] Embodiment 1 See also Figure 1 , is a flow chart of a cloud video content management method provided by a first embodiment of the present invention. The cloud video content management method can be applied to any device or system. The cloud video content management method includes the steps of: Step S10, obtaining comment information in the cloud video, and drawing a comment heat curve with the comment time in the comment information as the horizontal axis and the number of comments at the comment time as the vertical axis; Among them, the comment information includes the user's comment text and the comment time of the user's comment text. The number of comments is the number corresponding to the same comment time. When the number of comments at the comment time is larger, the more users comment on the video corresponding to the comment time in the cloud video. When the more users comment on the video, the higher the popularity of the video corresponding to the comment time.

[0019] Step S20, performing video editing on the cloud video according to the comment heat curve to obtain a hot video, and obtaining a video scene of a video frame in the hot video; Among them, by editing the cloud video through the comment heat curve, the hot videos in the cloud video can be automatically edited, and there is no need to manually edit the hot videos in the cloud video, which improves the efficiency of cloud video processing. The video scene is used to represent the background scene in the video frame. For example, the video scene is a bank, a playground, a hospital, etc. In this step, the video scene is obtained by matching the background image of the video frame with the scene database, and the scene database stores the corresponding relationship between different background images and corresponding video scenes.

[0020] Optionally, the cloud video is edited according to the comment heat curve to obtain a hot video, including: If the number of comments at any comment time in the comment heat curve is greater than the number threshold, the comment time is determined as a hot time; wherein the number threshold can be set according to demand; A hotspot duration range is generated according to the preset duration and the hotspot time, and the cloud video is edited according to the hotspot duration range to obtain the hotspot video; wherein the preset duration can be set according to demand, and the video corresponding to the hotspot duration range in the cloud video is edited to obtain the hotspot video.

[0021] Furthermore, before obtaining the video scene of the video frame in the hot video, the method further includes: Obtain like information in the cloud video, and draw a like heat curve with the like time in the like information as the horizontal axis and the number of likes at the like time as the vertical axis; Edit the cloud video according to the likes heat curve to obtain the hot video; Among them, if the number of likes at any like time is greater than the preset number, the like time will be determined as the hot spot time, and the hot spot time range will be generated according to the preset time and the hot spot time. The cloud video will be edited according to the hot spot time range to obtain the hot spot video.

[0022] Step S30, classifying the video frames according to the video scene to obtain a video scene frame set, and obtaining the coordinates of the user's gaze point of the video frame in the same video scene frame set; Among them, video frames of the same video scene are classified into the same set to obtain a video scene frame set, and the user gaze point coordinates in the same video scene frame set are obtained respectively. The user gaze point coordinates are the coordinates of the user's gaze point when watching the corresponding video frame, that is, when the cloud video is played on any playback device, the user's gaze point is recognized at the same time to obtain the user's gaze point coordinates.

[0023] Optionally, in this step, the facial image, the left eye image, the right eye image and the position data of the eye and face in the video frame are obtained; Obtain facial feature vectors according to a preset convolutional neural network and facial images; Obtaining a position feature vector according to a preset first fully connected network and position data; Obtaining a binocular fusion feature vector according to a preset eye feature fusion network, a left eye image, and a right eye image; The coordinates of the user's gaze point are determined according to the preset second fully connected network, the facial feature vector, the position feature vector and the binocular fusion feature vector.

[0024] Step S40, determining an object of interest according to the coordinates of the user's gaze point, and obtaining an object position area of ​​the object of interest in the video frame; Optionally, determining the object of interest according to the coordinates of the user's gaze point includes: Obtaining the area range of the video object in the video frame, and determining the number of gazes of the video object according to the user gaze point coordinates and the area range; wherein, obtaining the coordinates of each user gaze point in the same video scene frame set, matching the coordinates of each user gaze point with the area range of each video object, and determining the number of user gaze point coordinates falling within the area range of the video object as the number of gazes of the corresponding video object; Obtaining the number of user gaze point coordinates in the object position area of ​​the video object to obtain the gaze count; If the number of gazes on any of the video objects is greater than the number threshold, the video object is determined as the object of interest; wherein the number threshold can be set according to demand or the number of times the cloud video is played. When the number of gazes on any video object is greater than the number threshold, it is determined that the user is more concerned about the video object, and therefore, the video object is determined as the object of interest.

[0025] Step S50, determining the uninteresting objects in the video scene according to the object position area, and blurring the images of the uninteresting objects in the video scene in the hotspot video to obtain a hotspot optimized video; Among them, the uninteresting objects are objects that are relatively ignored when the user watches the video corresponding to the video scene.

[0026] Optionally, determining the uninteresting object in the video scene according to the object position area includes: A non-attention area in the video frame is determined according to the object position area, and the video object in the non-attention area is set as a candidate object; wherein an object position frame is generated according to the object position area, the object position frame is a rectangular frame diagram, and is used to represent the position of the object of interest corresponding to the object position area, and the area is expanded with the boundary of the object position frame as the starting point and the preset distance as the interval to obtain an extended area, and the area within the non-extended area in the video frame is determined as the non-attention area. When there are multiple extended areas in the same video frame, the area vertices of different extended areas are connected to obtain an area connection line, and the area within the area connection line is determined as the attention connected area, and the area within the non-attention connected area in the video frame is determined as the non-attention area.

[0027] If the candidate object exists in all the video frames in the same video scene frame set, the candidate object is determined as the uninteresting object in the video scene corresponding to the video scene frame set.

[0028] Step S60, performing semantic recognition on the comment information to obtain comment semantics, and determining a video type label according to the comment semantics; Among them, semantic recognition of comment information can be performed based on a natural language processing (NLP) algorithm to obtain comment semantics, and the comment semantics can be matched with a type label query table to obtain a video type label. The type label query table stores the correspondence between different comment semantics and corresponding video type labels.

[0029] Step S70: marking the content type of the hotspot optimized video according to the video type tag.

[0030] This embodiment also includes the management of digital assets. The main process of digital asset management includes: Collection and creation: Video data is collected through professional video acquisition equipment, such as high-definition cameras or webcams. The collected video data needs to be processed initially, such as editing and format conversion, to form usable digital assets.

[0031] Storage and organization: Digital assets need to be stored on secure and reliable storage devices, such as local servers, cloud storage, etc. In order to facilitate search and use, digital assets need to be reasonably classified and organized, such as by product type, usage channel, etc.

[0032] Metadata management: Metadata is additional information about digital assets, such as file name, author, creation date, description, etc. Metadata is essential for the search, location and use of digital assets. By adding appropriate metadata to digital assets, their searchability and usability can be greatly improved.

[0033] Distribution and use: Digital assets need to be distributed and used both inside and outside the enterprise. Internal use may include employee training, brand promotion, etc.; external use may include advertising, customer interaction, etc. The security and compliance of digital assets need to be ensured during distribution and use to avoid leakage or abuse.

[0034] This embodiment also includes multi-dimensional reports to analyze video interaction data: User behavior dimensions include: Watch time: An important indicator to measure the user's interest in video content. By analyzing the user's viewing behavior in different time periods, you can understand which time periods have the best video effects.

[0035] Interaction: including user behaviors such as likes, comments, and shares. These behaviors reflect the user's recognition and participation in the content. A high interaction rate usually means that the video content can resonate with the audience.

[0036] User portrait: including user information such as gender, age, region, etc. This data helps to understand the composition of the audience and adjust the content direction.

[0037] Video content dimensions include: Topic analysis: Whether the topic of the video is popular and whether it meets the interests of the audience can be evaluated through data analysis.

[0038] Style evaluation: Videos of different styles, such as funny, touching, and educational, will attract different audience groups. Through data analysis, we can understand which style is more popular.

[0039] Length analysis: The length of the video is also a factor that affects the viewing experience. Videos that are too long may cause user loss, while videos that are too short may not convey the complete message. Data analysis can help you find the optimal length for your target audience.

[0040] Competitor dimensions include: Video content comparison: Analyze competitors’ popular videos and find out the reasons for their success, such as topic selection, style positioning, etc.

[0041] Release time analysis: Observe the effects of competitors’ videos released at different time periods to adjust your own release strategy.

[0042] User interaction comparison: By analyzing the comments and sharing of competitors’ videos, you can understand users’ feedback on their content and optimize your own content strategy.

[0043] This embodiment also includes a localized server storage design for video interaction data: data can be stored in a direct attached storage (DAS), network attached storage (NAS) or storage area network (SAN) manner. The storage architecture design can adopt a distributed storage architecture or a centralized storage architecture.

[0044] In the present embodiment, a comment heat curve can be effectively drawn based on the comment time and the number of comments, and hot videos in the cloud video can be automatically edited based on the comment heat curve. By acquiring the video scene of the video frame, the video frame can be effectively classified to obtain a video scene frame set. By acquiring the coordinates of the user's gaze point, the object of interest can be automatically determined. Based on the object position area of ​​the object of interest in the video frame, the uninteresting objects in the video scene can be automatically determined. By blurring the image of the uninteresting object, the hot video is effectively optimized. By performing semantic recognition on the comment information, the comment semantics can be effectively obtained. Based on the comment semantics, the video type label can be effectively determined. Based on the video type label, the content type of the hot optimized video can be automatically marked. There is no need to manually edit the hot video and mark the content type, thereby improving the cloud video content processing efficiency.

[0045] Embodiment 2 See also Figure 2 , is a schematic diagram of the structure of a cloud video content management system 100 provided in a second embodiment of the present invention, including: The heat analysis module 10 is used to obtain comment information in the cloud video, and draw a comment heat curve with the comment time in the comment information as the horizontal axis and the number of comments at the comment time as the vertical axis.

[0046] The video editing module 11 is used to perform video editing on the cloud video according to the comment heat curve to obtain a hot video and acquire a video scene of a video frame in the hot video.

[0047] Optionally, the video editing module 11 is further used for: if the number of comments at any comment time in the comment heat curve is greater than a number threshold, determining the comment time as a hot time; A hotspot duration range is generated according to a preset duration and the hotspot time, and the cloud video is edited according to the hotspot duration range to obtain the hotspot video.

[0048] Furthermore, the video editing module 11 is also used to: obtain the like information in the cloud video, and draw a like heat curve with the like time in the like information as the horizontal axis and the number of likes at the like time as the vertical axis; The cloud video is edited according to the likes heat curve to obtain the hot video.

[0049] The video classification module 12 is used to classify the video frames according to the video scene to obtain a video scene frame set, and obtain the coordinates of the user's gaze point of the video frame in the same video scene frame set.

[0050] The object determination module 13 is used to determine the object of interest according to the coordinates of the user's gaze point, and obtain the object position area of ​​the object of interest in the video frame.

[0051] Optionally, the object determination module 13 is further used to: obtain a region range of the video object in the video frame, and determine the gaze count of the video object according to the user gaze point coordinates and the region range; If the number of gazes on any of the video objects is greater than a number threshold, the video object is determined as the object of interest.

[0052] The image blur module 14 is used to determine the uninteresting objects in the video scene according to the object position area, and blur the images of the uninteresting objects in the video scene in the hotspot video to obtain a hotspot optimized video.

[0053] Optionally, the image blurring module 14 is further configured to: determine a non-attention area in the video frame according to the object position area, and set the video object in the non-attention area as a candidate object; If the candidate object exists in all the video frames in the same video scene frame set, the candidate object is determined as the uninteresting object in the video scene corresponding to the video scene frame set.

[0054] The content marking module 15 is used to perform semantic recognition on the comment information, obtain comment semantics, and determine the video type label according to the comment semantics; and perform content type labeling on the hotspot optimized video according to the video type label.

[0055] In the present embodiment, a comment heat curve can be effectively drawn based on the comment time and the number of comments, and hot videos in the cloud video can be automatically edited based on the comment heat curve. By acquiring the video scene of the video frame, the video frame can be effectively classified to obtain a video scene frame set. By acquiring the coordinates of the user's gaze point, the object of interest can be automatically determined. Based on the object position area of ​​the object of interest in the video frame, the uninteresting objects in the video scene can be automatically determined. By blurring the image of the uninteresting object, the hot video is effectively optimized. By performing semantic recognition on the comment information, the comment semantics can be effectively obtained. Based on the comment semantics, the video type label can be effectively determined. Based on the video type label, the content type of the hot optimized video can be automatically marked. There is no need to manually edit the hot video and mark the content type, thereby improving the cloud video content processing efficiency.

[0056] Embodiment 3 Figure 3 2 is a block diagram of a terminal device 2 provided in the third embodiment of the present application. Figure 3 As shown, the terminal device 2 of this embodiment includes: a processor 20, a memory 21, and a computer program 22 stored in the memory 21 and executable on the processor 20, such as a program of the cloud video content management method. When the processor 20 executes the computer program 22, the steps in each embodiment of the above-mentioned cloud video content management method are implemented.

[0057] Exemplarily, the computer program 22 may be divided into one or more modules, which are stored in the memory 21 and executed by the processor 20 to complete the present application. The one or more modules may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 22 in the terminal device 2. The terminal device may include, but is not limited to, a processor 20 and a memory 21.

[0058] The processor 20 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0059] The memory 21 may be an internal storage unit of the terminal device 2, such as a hard disk or memory of the terminal device 2. The memory 21 may also be an external storage device of the terminal device 2, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device 2. Further, the memory 21 may also include both an internal storage unit and an external storage device of the terminal device 2. The memory 21 is used to store the computer program and other programs and data required by the terminal device. The memory 21 may also be used to temporarily store data that has been output or is to be output.

[0060] In addition, each functional module in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional unit.

[0061] If the integrated module is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Among them, the computer-readable storage medium can be non-volatile or volatile. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in computer-readable storage media can be appropriately increased or decreased according to the requirements of legislation and patent practices in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practices, computer-readable storage media do not include electrical carrier signals and telecommunication signals.

[0062] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A cloud video content management method, characterized in that: The method comprises: Obtain comment information in the cloud video, and draw a comment heat curve with the comment time in the comment information as the horizontal axis and the number of comments at the comment time as the vertical axis; Performing video editing on the cloud video according to the comment heat curve to obtain a hot video, and obtaining a video scene of a video frame in the hot video; Classifying the video frames according to the video scene to obtain a video scene frame set, and obtaining the coordinates of the user's gaze point of the video frame in the same video scene frame set; Determine an object of interest according to the coordinates of the user's gaze point, and obtain an object position area of ​​the object of interest in the video frame; Determine the uninteresting objects in the video scene according to the object position area, and blur the image of the uninteresting objects in the video scene in the hotspot video to obtain a hotspot optimized video; Performing semantic recognition on the comment information to obtain comment semantics, and determining a video type label according to the comment semantics; The hotspot optimized video is marked with a content type according to the video type tag.

2. The cloud video content management method according to claim 1, characterized in that: Determining the uninteresting object in the video scene according to the object position area includes: Determine a non-attention area in the video frame according to the object position area, and set the video object in the non-attention area as a candidate object; If the candidate object exists in all the video frames in the same video scene frame set, the candidate object is determined as the uninteresting object in the video scene corresponding to the video scene frame set.

3. The cloud video content management method according to claim 1, characterized in that: The cloud video is edited according to the comment heat curve to obtain a hot video, including: If the number of comments at any comment time in the comment heat curve is greater than the number threshold, the comment time is determined as a hot time; A hotspot duration range is generated according to a preset duration and the hotspot time, and the cloud video is edited according to the hotspot duration range to obtain the hotspot video.

4. The cloud video content management method according to claim 1, characterized in that: Determining an object of interest according to the coordinates of the user's gaze point includes: Acquire the area range of the video object in the video frame, and determine the number of gazes of the video object according to the user's gaze point coordinates and the area range; If the number of gazes on any of the video objects is greater than a number threshold, the video object is determined as the object of interest.

5. The cloud video content management method according to claim 1, characterized in that: Before obtaining the video scene of the video frame in the hot video, the method further includes: Obtain like information in the cloud video, and draw a like heat curve with the like time in the like information as the horizontal axis and the number of likes at the like time as the vertical axis; The cloud video is edited according to the likes heat curve to obtain the hot video.

6. A cloud video content management system, characterized in that: The system comprises: A heat analysis module is used to obtain comment information in the cloud video, and draw a comment heat curve with the comment time in the comment information as the horizontal axis and the number of comments at the comment time as the vertical axis; A video editing module, used to edit the cloud video according to the comment heat curve to obtain a hot video, and obtain a video scene of a video frame in the hot video; A video classification module, used to classify the video frames according to the video scene to obtain a video scene frame set, and obtain the coordinates of the user's gaze point of the video frame in the same video scene frame set; An object determination module, used to determine an object of interest according to the coordinates of the user's gaze point, and obtain an object position area of ​​the object of interest in the video frame; An image blur module, used for determining uninteresting objects in the video scene according to the object position area, and blurring the images of the uninteresting objects in the video scene in the hotspot video to obtain a hotspot optimized video; The content marking module is used to perform semantic recognition on the comment information, obtain comment semantics, and determine the video type label according to the comment semantics; and mark the content type of the hot spot optimized video according to the video type label.

7. The cloud video content management system according to claim 6, characterized in that: The image blur module is also used for: Determine a non-attention area in the video frame according to the object position area, and set the video object in the non-attention area as a candidate object; If the candidate object exists in all the video frames in the same video scene frame set, the candidate object is determined as the uninteresting object in the video scene corresponding to the video scene frame set.

8. The cloud video content management system according to claim 6, characterized in that: The video clipping module is also used for: If the number of comments at any comment time in the comment heat curve is greater than the number threshold, the comment time is determined as a hot time; A hotspot duration range is generated according to a preset duration and the hotspot time, and the cloud video is edited according to the hotspot duration range to obtain the hotspot video.

9. The cloud video content management system according to claim 6, characterized in that: The object determination module is also used for: Acquire the area range of the video object in the video frame, and determine the number of gazes of the video object according to the user's gaze point coordinates and the area range; If the number of gazes on any of the video objects is greater than a number threshold, the video object is determined as the object of interest.

10. A computer-readable storage medium storing a computer program, 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 5 are implemented.

Images