An internet-based video intelligent production method and system

Abstract

This invention relates to the field of video production technology, specifically disclosing an intelligent video production method and system based on the Internet, comprising the following steps: S1: Obtaining user-inputted keywords, and calculating the matching degree based on keyword matching with material tags; S2: Obtaining the number of views and editing uses of the material, calculating the editing value, and filtering out materials that meet the requirements as reserve materials; S3: Dividing the reserve materials into material segments of preset duration, obtaining the number of bullet comments in the material segments and filtering out high-quality material segments, and calculating the popularity value of high-quality segments; S4: Calculating the propagation speed of high-quality segments, calculating the comprehensive editing value based on propagation speed and popularity value, sorting the reserve materials, and filtering out recommended materials. This invention improves the speed and accuracy of material retrieval through an intelligent method for filtering video materials, thereby increasing the video production rate and simplifying the video production process.

Description

Technical Field

[0001] This invention relates to the field of video production technology, specifically to an internet-based intelligent video production method and system. Background Technology

[0002] In today's digital age, the demand for video content is growing rapidly. However, traditional video content production typically requires a significant amount of time. Video production is the process of combining, editing, and processing a series of elements such as images, audio, and text to ultimately create an audiovisual work with a specific theme and expressive power. It can be used for various purposes, such as conveying information, telling stories, recording events, and entertaining audiences.

[0003] In video production, material retrieval is a crucial step, directly impacting video quality and production efficiency. First, the required material types and content must be determined based on the video's theme. Ensure the materials closely align with the theme and accurately convey the video's core message. Then, based on the video's stylistic characteristics, select materials that fit the theme.

[0004] In current technology, video production is a tedious process. People typically spend a significant amount of time searching, browsing, and evaluating vast amounts of footage to find suitable material for their video content. This tedious and repetitive manual work wastes time and resources. Therefore, a method for intelligently filtering video footage is needed to improve the speed and accuracy of footage retrieval, thereby increasing the video production rate and simplifying the process. Summary of the Invention

[0005] The purpose of this invention is to provide an internet-based intelligent video production method and system to solve the above-mentioned technical problems.

[0006] The objective of this invention can be achieved through the following technical solutions: A method for intelligent video production based on the Internet includes the following steps: S1: Obtain the keywords input by the user, and calculate the matching degree P based on the keywords and material tags. The material tags refer to the text that classifies the material according to content, scene and purpose. S2: Obtain the number of views V and the number of times the clip was used S, and calculate the clip value K using a formula, which is as follows: ; Among them, T v The average duration of time a content is viewed, T s The average duration of footage used in editing is represented by λ, which represents the preset magnification factor, where λ > 1. Editing value K > K aveThe material is recorded as spare material, K ave The mean of the editing value K; S3: Divide the spare material into material segments of preset duration, obtain the number of bullet comments D in the material segments, and ensure that the number of bullet comments D is greater than the preset standard number D. sta Footage clips are marked as high-quality footage clips, and the popularity value of high-quality clips is calculated. Where γ represents the preset heat coefficient, γ > 0; S4: Calculate the propagation speed of high-quality segments , where Q now Q represents the current date. end The release date of the backup material is represented by Sh, and the number of times the high-quality clips in the backup material have been shared is represented by Sh. Calculate the overall editing value of the backup footage Where δ represents the preset editing adjustment coefficient, and N all N represents the total number of source clips. R R represents the number of high-quality segments. i C represents the popularity value of the i-th high-quality segment. i This represents the propagation speed of the i-th high-quality segment; The backup materials were sorted from highest to lowest based on their overall value (ZK), and the top 10% of the backup materials were selected as recommended materials for video production.

[0007] As a further aspect of the present invention: in step S1, the method for calculating the matching degree P based on keywords and material tags includes: Let the words in the keywords be labeled as keywords A1, A2, ..., An, and the words in the material tags be labeled as tags B1, B2, ..., Bm. Let tag B1 be labeled as the sentinel word. Start matching keyword A1 from sentinel word B1 and proceed backward. If a tag Bs that is the same as keyword A1 is found, the match is successful. Update the sentinel word to tag Bs, where s is a positive integer and 1≤s≤m. Repeat the above process until all keywords are matched successfully. Calculate matching degree N g N represents the number of characters in the keywords. s N represents the number of characters in the material tag. same This represents the number of times the Sentinel word has been updated.

[0008] As a further aspect of the present invention: in step S2, the browsing duration t of the material is obtained; if the browsing duration t is less than a preset browsing duration threshold t... min If the data is not viewed, it will not be counted in the content's view count (V).

[0009] As a further aspect of the present invention: in step S2, the editing rate JK=S / V of the material is calculated. If the editing rate JK is less than or equal to 20%, it is not recommended material for video production.

[0010] As a further aspect of the present invention: in step S3, if the number of high-quality material fragments in the spare materials is equal to 0, the corresponding spare materials are removed and do not participate in subsequent steps.

[0011] As a further aspect of the present invention: In step S3, when dividing the spare material into material segments of a preset duration, material segments that are not equal to the preset duration are recorded as incomplete material segments, and the number D of bullet comments in the incomplete material segments is obtained. s ,like Retain incomplete footage fragments, if Delete incomplete footage clips, t s t represents the duration of an incomplete clip. sta This represents the preset duration.

[0012] As a further aspect of the present invention: in step S4, if there are spare materials with equal comprehensive editing value ZK, the spare materials with a larger number of high-quality clips are placed at the front of the sorting.

[0013] An internet-based intelligent video production system includes: Input module: Obtains keywords input by the user, and calculates the matching degree P based on the keywords and material tags. The material tags refer to the text that classifies the material according to content, scene and purpose. Calculation module: Obtains the number of views V and the number of times the clip is used S, and calculates the clip value K using a formula, the specific formula of which is: ; Among them, T v The average duration of time a content is viewed, T s The average duration of footage used in editing is represented by λ, which represents the preset magnification factor, where λ > 1. Editing value K > K ave The material is recorded as spare material, K ave The mean of the editing value K; Processing module: Divides the spare material into material segments of preset duration, obtains the number D of bullet comments in the material segments, and sets the number of bullet comments D to be greater than the preset standard number D. sta Footage clips are marked as high-quality footage clips, and the popularity value of high-quality clips is calculated. Where γ represents the preset heat coefficient, γ > 0; Recommended module: Calculate the propagation speed of high-quality segments. , where Q now Q represents the current date. end The release date of the backup material is represented by Sh, and the number of times the high-quality clips in the backup material have been shared is represented by Sh. Calculate the overall editing value of the backup footage Where δ represents the preset editing adjustment coefficient, and N all N represents the total number of source clips. R R represents the number of high-quality segments. i C represents the popularity value of the i-th high-quality segment. i This represents the propagation speed of the i-th high-quality segment; The backup materials were sorted from highest to lowest based on their overall value (ZK), and the top 10% of the backup materials were selected as recommended materials for video production.

[0014] The beneficial effects of this invention are as follows: First, by matching the keywords entered by the user, materials that match the user's theme are searched. Then, the matching degree between the keywords and material tags is calculated. This method can determine the degree of fit between the materials and the user's theme. Then, the editing value is calculated by the number of views and the number of times the material is used in the edit. As can be seen from the formula, the higher the number of views, the more popular the material is. However, too many times the material is used in the edit indicates that the material has been widely used, so using it again will reduce its novelty, thus leading to a decrease in the editing value. By selecting backup materials, the workload can be reduced and some materials with little value can be eliminated.

[0015] Dividing the footage into pre-set duration segments provides a more intuitive view of its usage. In video production, one or more segments are typically edited, rarely used as a whole. Therefore, calculating the popularity of a segment reflects its quality, followed by calculating the spread speed of high-quality segments. Spread speed reflects audience acceptance of high-quality segments. The formula for calculating the overall editing value of footage shows that the popularity of a high-quality segment is directly proportional to its quality, while spread speed represents audience willingness to share it; a higher spread speed indicates a greater willingness among viewers to share the video. Since videos are frequently shared with others, from the above two points, when making videos, it is necessary to meet both quality requirements and the desire for others to share them. Only by satisfying both of these conditions can a video be considered excellent. Therefore, the popularity value and the speed of dissemination are multiplied to obtain the comprehensive editing value of the backup materials. To obtain a high comprehensive editing value, both the popularity value and the speed of dissemination need to be at a high level. Then, the best recommended materials are selected based on the comprehensive editing value. In summary, this invention improves the speed and accuracy of material retrieval by intelligently filtering video materials, which not only optimizes the quality of video production but also increases the video production speed and simplifies the video production process. Attached Figure Description

[0016] The invention will now be further described with reference to the accompanying drawings.

[0017] Figure 1 This is a flowchart illustrating an internet-based intelligent video production method according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 As shown, this invention is an internet-based intelligent video production method, comprising the following steps: S1: Obtain the keywords input by the user, and calculate the matching degree P based on the keywords and material tags. The material tags refer to the text that classifies the material according to content, scene and purpose. S2: Obtain the number of views V and the number of times the clip was used S, and calculate the clip value K using a formula, which is as follows: ; Among them, T vThe average duration of time a content is viewed, T s The average duration of footage used in editing is represented by λ, which represents the preset magnification factor, where λ > 1. Editing value K > K ave The material is recorded as spare material, K ave The mean of the editing value K; S3: Divide the spare material into material segments of preset duration, obtain the number of bullet comments D in the material segments, and ensure that the number of bullet comments D is greater than the preset standard number D. sta Footage clips are marked as high-quality footage clips, and the popularity value of high-quality clips is calculated. Where γ represents the preset heat coefficient, γ > 0; S4: Calculate the propagation speed of high-quality segments , where Q now Q represents the current date. end The release date of the backup material is represented by Sh, and the number of times the high-quality clips in the backup material have been shared is represented by Sh. Calculate the overall editing value of the backup footage Where δ represents the preset editing adjustment coefficient, and N all N represents the total number of source clips. R R represents the number of high-quality segments. i C represents the popularity value of the i-th high-quality segment. i This represents the propagation speed of the i-th high-quality segment; The backup materials were sorted from highest to lowest based on their overall value (ZK), and the top 10% of the backup materials were selected as recommended materials for video production.

[0020] It's important to note that, firstly, after a user enters keywords, those keywords are matched with relevant material tags. This mechanism analyzes the connotation and potential needs of the user's entered keywords through simple literal matching. A comprehensive search is then conducted in the material library to find materials closely related to the user's theme. These materials may come from different sources and have different styles, but all contain potential value related to the user's theme.

[0021] After finding materials that initially match the user's theme, the next crucial step is to calculate the match between keywords and material tags. Material tags are like the material's "identity card," recording important information such as its core content, style characteristics, and applicable scenarios. Keywords, on the other hand, are the "password" for the user to express their needs. By comparing keywords with material tags across various dimensions, from semantic similarity to logical relevance, and analyzing these aspects, this in-depth matching calculation can accurately determine the degree of fit between the material and the user's theme, essentially building a precise bridge between the material and the user's needs. However, simply determining the fit between the material and the theme is far from sufficient. In the field of content creation, the editing value of the material is also an important criterion for measuring its quality. To accurately evaluate the editing value of the material, multiple factors need to be considered, among which the material's views and the number of times it is used in editing are two key indicators. Views are like the material's "popularity," directly reflecting its popularity in the online world. Generally speaking, the higher the views, the more the material attracts the audience's attention, and its content is likely to have higher entertainment value, practicality, or visual appeal.

[0022] One point to note is that, regarding the method for calculating the value of a clip, T... v The average duration of time a content is viewed, T s The average duration of footage being edited is represented by two units, both in seconds. The number of views (V) and the number of times the footage is edited (S) are quantifiers without dimensions, so the result of multiplying them is still in seconds.

[0023] However, excessive use of clips can lead to problems. When a piece of footage is widely used, its novelty in the eyes of the audience gradually diminishes. Imagine if a particular special effect or sound effect appeared repeatedly in numerous videos; viewers might experience aesthetic fatigue and even develop a negative attitude towards works that use that footage. Therefore, while high usage frequency reflects the versatility and practicality of the footage to some extent, overuse can also reduce its editing value.

[0024] Based on the above analysis, a formula can be used to calculate the editing value of footage. This formula considers the balance between views and the number of times the clips are used, ensuring that footage with higher views and a moderate number of uses receives a higher editing value score. Conversely, footage with high views but excessive use, or low views but very few uses, will have a lower editing value. This method allows for the selection of backup footage, significantly reducing the workload of subsequent screening and effectively eliminating low-value footage, thus improving the efficiency and accuracy of the selection process.

[0025] After selecting the backup footage, further analysis and processing are needed to gain a more comprehensive understanding of its usage. One crucial step is dividing the footage into segments of preset duration. This is because in actual video production, creators rarely use entire clips directly; instead, they typically carefully edit and combine one or more segments according to creative needs. Therefore, dividing the footage into segments of preset duration provides a more direct view of its usage. By analyzing each segment, it's possible to identify which parts received particular attention from viewers and which were relatively less popular.

[0026] Next, we need to calculate the spread speed of high-quality clips. Spread speed is a crucial metric, reflecting the reach and impact of a high-quality clip within its audience. A high spread speed indicates high audience approval and a willingness to actively share it. This sharing behavior acts as a powerful force, propelling the clip's rapid dissemination online, allowing more people to see and use it. If a clip resonates with its audience, they will readily share it on their social media platforms or with friends and family, resulting in extremely high exposure for the clip in a short period.

[0027] At the same time, it's also important to consider the overall editing value of the footage. Calculating this value involves multiple aspects, with the popularity and dissemination speed of high-quality clips being two core elements. Popularity represents the quality of a clip and is directly proportional to it. In other words, the higher the popularity of a clip, the more popular it is among viewers, and the more its quality is recognized. Dissemination speed, on the other hand, represents the exposure rate of the clip to viewers. A higher dissemination speed indicates that viewers are more willing to share the video with others, which also reflects the clip's appeal and influence.

[0028] From these two factors, excellent video footage must meet both quality requirements and high popularity, resonating with and appealing to viewers. Simultaneously, it must have a high rate of dissemination, encouraging others to share it and expand its influence and reach. Only by simultaneously satisfying both criteria can it be considered truly excellent video footage.

[0029] Therefore, multiplying the popularity value (R) by the dissemination speed (C) yields the overall editing value of the backup materials. This overall editing value acts like a comprehensive scoring system, objectively and accurately reflecting the overall quality and value of the backup materials. Only materials that excel in both popularity and dissemination speed can achieve a high overall editing value, thus standing out from the numerous backup materials and being selected as the best recommended materials. Through this rigorous and scientific selection method, users can be provided with the highest quality and most suitable materials, helping them achieve better results in content creation.

[0030] In another preferred embodiment of the present invention, the method for calculating the matching degree P based on keywords and material tags includes: Let the words in the keywords be labeled as keywords A1, A2, ..., An, and the words in the material tags be labeled as tags B1, B2, ..., Bm. Let tag B1 be labeled as the sentinel word. Start matching keyword A1 from sentinel word B1 and proceed backward. If a tag Bs that is the same as keyword A1 is found, the match is successful. Update the sentinel word to tag Bs, where s is a positive integer and 1≤s≤m. Repeat the above process until all keywords are matched successfully. Calculate matching degree N g N represents the number of characters in the keywords. s N represents the number of characters in the material tag. same This represents the number of times the Sentinel word has been updated.

[0031] It's worth noting that when a user enters keywords, the keywords are matched with relevant material tags. This mechanism analyzes the connotation and potential needs of the user's input keywords through simple literal matching. A comprehensive search is conducted in the material library to explore materials closely related to the user's theme. These materials may come from different sources and have different styles, but they all contain potential value related to the user's theme. By comparing keywords and material tags one by one across various dimensions, from semantic similarity to logical relevance, the system analyzes the results. Through this deep matching calculation, the system can accurately determine the degree of fit between the materials and the user's theme, essentially building a precise bridge between the materials and the user's needs.

[0032] In another preferred embodiment of the present invention, the browsing duration t of the material is obtained; if the browsing duration t is less than a preset browsing duration threshold t min If the data is not viewed, it will not be counted in the content's view count (V).

[0033] It is understandable that in practice, some users may make misjudgments when searching for video materials. For materials with short browsing durations, users may accidentally click on them while searching. If these are used directly without processing, it will lead to a large error in the final calculation, which will affect the recommended materials for subsequent video production, resulting in poor material recommendations and a decrease in video quality.

[0034] In another preferred embodiment of the present invention, the cut rate JK of the material is calculated as S / V. If the cut rate JK is less than or equal to 20%, it is not recommended as material for video production.

[0035] It is important to note that for some footage with a cut rate of less than 20%, it should be removed directly to simplify the process. The cut rate indicates that if the cut rate of a piece of footage is extremely low, it means that its quality is low, so there is no need to spend much effort on it. It should be deleted directly and not used as recommended footage for video production.

[0036] In another preferred embodiment of the present invention, if the number of high-quality material fragments in the spare material is equal to 0, the corresponding spare material is removed and does not participate in subsequent steps.

[0037] It should be noted that if the number of high-quality clips in the backup materials is zero, the corresponding backup materials will be removed. This means that there are no particularly outstanding clips in these backup materials. Therefore, in order to further simplify the process and reduce the computing power spent on finding video materials, these backup materials without high-quality clips will be removed, thereby further speeding up the video material screening process.

[0038] In another preferred embodiment of the present invention, when dividing the spare material into material segments of a preset duration, material segments that are not equal to the preset duration are recorded as incomplete material segments, and the number D of bullet comments in the incomplete material segments is obtained. s ,like Retain incomplete footage fragments, if Delete incomplete footage clips, t s t represents the duration of an incomplete clip. sta This represents the preset duration.

[0039] Understandably, when dividing the spare material into segments of preset duration, it is almost impossible to divide all the spare material into complete segments. In most cases, there will be an incomplete segment. This incomplete segment cannot be simply deleted. Instead, it should be judged based on the number of comments in the incomplete segment. The number of comments should be calculated based on the ratio of the incomplete segment's duration to the preset duration. If the number of comments in the incomplete segment is greater than the number calculated based on the ratio, then the incomplete segment should be retained.

[0040] In another preferred embodiment of the present invention, if there are spare materials with equal comprehensive editing value ZK, the spare materials with a larger number of high-quality clips are placed at the front of the sorting.

[0041] It's worth noting that if backup footage with equal overall editing value exists, the backup footage with more high-quality clips is ranked higher in the sorting. This is done to provide more and better options when creating the video. The overall editing value is calculated considering two factors, so the two backup footage can be considered to be of roughly the same quality. However, providing the creator with more choices during video production is beneficial for their editing activities. Therefore, ranking the backup footage with more high-quality clips higher in the sorting also enhances the invention's resilience to certain extreme situations.

[0042] An internet-based intelligent video production system includes: Input module: Obtains keywords input by the user, and calculates the matching degree P based on the keywords and material tags. The material tags refer to the text that classifies the material according to content, scene and purpose. Calculation module: Obtains the number of views V and the number of times the clip is used S, and calculates the clip value K using a formula, the specific formula of which is: ; Among them, T v The average duration of time a content is viewed, T s The average duration of footage used in editing is represented by λ, which represents the preset magnification factor, where λ > 1. Editing value K > K ave The material is recorded as spare material, K ave The mean of the editing value K; Processing module: Divides the spare material into material segments of preset duration, obtains the number D of bullet comments in the material segments, and sets the number of bullet comments D to be greater than the preset standard number D. sta Footage clips are marked as high-quality footage clips, and the popularity value of high-quality clips is calculated. Where γ represents the preset heat coefficient, γ > 0; Recommended module: Calculate the propagation speed of high-quality segments. , where Q now Q represents the current date. end The release date of the backup material is represented by Sh, and the number of times the high-quality clips in the backup material have been shared is represented by Sh. Calculate the overall editing value of the backup footage Where δ represents the preset editing adjustment coefficient, and N all N represents the total number of source clips. R R represents the number of high-quality segments. i C represents the popularity value of the i-th high-quality segment. i This represents the propagation speed of the i-th high-quality segment; The backup materials were sorted from highest to lowest based on their overall value (ZK), and the top 10% of the backup materials were selected as recommended materials for video production.

[0043] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A video intelligent production method based on the Internet, characterized in that, Includes the following steps: S1: Obtain the keywords input by the user, and calculate the matching degree P based on the keywords and material tags. The material tags refer to the text that classifies the material according to content, scene and purpose. S2: Obtain the number of views V and the number of times the clip was used S, and calculate the clip value K using a formula, which is as follows: ; Wherein, T v represents the average value of the material browsing time, T s represents the average value of the material clipping usage time, λ represents a preset amplification coefficient, λ>1; Editing value K > K ave The material is recorded as spare material, K ave The mean of the editing value K; S3: Divide the spare material into material segments of preset duration, obtain the number of bullet comments D in the material segments, and ensure that the number of bullet comments D is greater than the preset standard number D. sta Footage clips are marked as high-quality footage clips, and the popularity value of high-quality clips is calculated. Where γ represents the preset heat coefficient, γ > 0; S4: Calculate the propagation speed of high-quality segments , where Q now Q represents the current date. end The release date of the backup material is represented by Sh, and the number of times the high-quality clips in the backup material have been shared is represented by Sh. Calculate the overall editing value of the backup footage Where δ represents the preset editing adjustment coefficient, and N all N represents the total number of source clips. R R represents the number of high-quality segments. i C represents the popularity value of the i-th high-quality segment. i This represents the propagation speed of the i-th high-quality segment; The backup materials were sorted from highest to lowest based on their overall value (ZK), and the top 10% of the backup materials were selected as recommended materials for video production.

2. The Internet-based intelligent video production method according to claim 1, characterized in that, In step S1, the method for calculating the matching degree P based on keywords and material tags includes: Let the words in the keywords be labeled as keywords A1, A2, ..., An, and the words in the material tags be labeled as tags B1, B2, ..., Bm. Let tag B1 be labeled as the sentinel word. Start matching keyword A1 from sentinel word B1 and proceed backward. If a tag Bs that is the same as keyword A1 is found, the match is successful. Update the sentinel word to tag Bs, where s is a positive integer and 1≤s≤m. Repeat the above process until all keywords are matched successfully. Calculate matching degree N g N represents the number of characters in the keywords. s N represents the number of characters in the material tag. same This represents the number of times the Sentinel word has been updated.

3. The Internet-based intelligent video production method according to claim 1, characterized in that, In step S2, the viewing duration t of the material is obtained. If the viewing duration t is less than a preset viewing duration threshold t... min If the data is not viewed, it will not be counted in the content's view count (V).

4. The Internet-based intelligent video production method according to claim 1, characterized in that, In step S2, the cut rate JK = S / V of the material is calculated. If the cut rate JK is less than or equal to 20%, it will not be used as recommended material for video production.

5. The Internet-based intelligent video production method according to claim 1, characterized in that, In step S3, if the number of high-quality material fragments in the backup materials is equal to 0, the corresponding backup materials are removed and do not participate in subsequent steps.

6. The Internet-based intelligent video production method according to claim 1, characterized in that, In step S3, when dividing the spare material into material segments of a preset duration, material segments that are not equal to the preset duration are recorded as incomplete material segments, and the number of bullet comments D in the incomplete material segments is obtained. s ,like Retain incomplete footage fragments, if Delete incomplete footage clips, t s t represents the duration of an incomplete clip. sta This represents the preset duration.

7. The Internet-based intelligent video production method according to claim 1, characterized in that, In step S4, if there are spare clips with equal overall editing value ZK, the spare clips with a larger number of high-quality clips are placed at the top of the sorting.

8. An internet-based intelligent video production system, characterized in that, include: Input module: Obtains keywords input by the user, and calculates the matching degree P based on the keywords and material tags. The material tags refer to the text that classifies the material according to content, scene and purpose. Calculation module: Obtains the number of views V and the number of times the clip is used S, and calculates the clip value K using a formula, the specific formula of which is: Among them, T v The average duration of time a content is viewed, T s The average duration of footage used in editing is represented by λ, which represents the preset magnification factor, where λ > 1. Editing value K > K ave The material is recorded as spare material, K ave The mean of the editing value K; Processing module: Divides the spare material into material segments of preset duration, obtains the number D of bullet comments in the material segments, and sets the number of bullet comments D to be greater than the preset standard number D. sta Footage clips are marked as high-quality footage clips, and the popularity value of high-quality clips is calculated. Where γ represents the preset heat coefficient, γ > 0; Recommended module: Calculate the propagation speed of high-quality segments. , where Q now Q represents the current date. end Sh represents the release date of the backup footage, and Sh represents the number of times high-quality clips from the backup footage have been shared; the overall editing value of the backup footage is calculated. Where δ represents the preset editing adjustment coefficient, and N all N represents the total number of source clips. R R represents the number of high-quality segments. i C represents the popularity value of the i-th high-quality segment. i This represents the propagation speed of the i-th high-quality segment; The backup materials were sorted from highest to lowest based on their overall value (ZK), and the top 10% of the backup materials were selected as recommended materials for video production.

Images