CN114339407A - Automatic short video generation method and system - Google Patents

Automatic short video generation method and system Download PDF

Info

Publication number
CN114339407A
CN114339407A CN202210234535.XA CN202210234535A CN114339407A CN 114339407 A CN114339407 A CN 114339407A CN 202210234535 A CN202210234535 A CN 202210234535A CN 114339407 A CN114339407 A CN 114339407A
Authority
CN
China
Prior art keywords
video
user
scene
database
template
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202210234535.XA
Other languages
Chinese (zh)
Other versions
CN114339407B (en
Inventor
张莹
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Entertainment Interactive Technology Beijing Co ltd
Original Assignee
Entertainment Interactive Technology Beijing Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Entertainment Interactive Technology Beijing Co ltd filed Critical Entertainment Interactive Technology Beijing Co ltd
Priority to CN202210234535.XA priority Critical patent/CN114339407B/en
Publication of CN114339407A publication Critical patent/CN114339407A/en
Application granted granted Critical
Publication of CN114339407B publication Critical patent/CN114339407B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)

Abstract

The invention provides a method and a system for automatically generating a short video. The method comprises the following steps: a user registers user information and inputs keywords belonging to the field of video content for making short videos in the process of registering the user information; judging whether a user imports a video file after system login in real time, and generating a video template according to the field keywords of the video content imported by the user for making the short video and the video file imported by the user; automatically calling a plurality of video models matched with the making elements needing to make the short video input by the user according to the making elements needing to make the short video and the text information input by the user, and automatically generating the short video by combining the making elements according to a video template selected by the user; and automatically carrying out adaptive template pattern adjustment on the generated video template according to the short video file which is manufactured by the user and the use condition of the video template. The system comprises modules corresponding to the method steps.

Description

Automatic short video generation method and system
Technical Field
The invention provides a method and a system for automatically generating a short video, belonging to the technical field of video production.
Background
The short video is a short video, is an internet content transmission mode, is generally a video which is transmitted on a new internet medium within 5 minutes, integrates the topics of skill sharing, humorous work, fashion trend, social hotspots, street interviews, public education, advertising creativity, commercial customization and the like, and is suitable for being watched in a mobile state and a short-time leisure state.
Along with the popularization of mobile terminals and the acceleration of networks, short and fast large-flow propagation content gradually obtains the favor of each large platform and capital, people also prefer to record and share life through short videos, most of the existing methods or systems for making short videos are too complex, the making process is complicated, users need to have higher video processing knowledge and spend a large amount of time for video processing when making short videos, and therefore most of platforms and systems for automatically generating short videos are promoted.
Disclosure of Invention
The invention provides a method and a system for automatically generating short videos, which are used for solving the problems of poor diversity and low video quality of automatically generated short videos in the prior art, and adopt the following technical scheme:
a short video automatic generation method comprises the following steps:
a user registers user information and inputs keywords belonging to the field of video content for making short videos in the process of registering the user information; wherein the domain keywords comprise life, history, cultural and educational, entertainment, pets and the like;
judging whether a user imports a video file after system login in real time, and generating a video template according to the field keywords of the video content imported by the user for making the short video and the video file imported by the user;
automatically calling a plurality of video templates matched with the making elements needing to make the short video input by the user according to the making elements needing to make the short video and the text information input by the user; selecting a corresponding video template according to the preference of the user, and automatically generating a short video according to the video template selected by the user and the manufacturing element; the production elements comprise materials such as pictures, videos, music, animations and characters required by short video production; the user can adjust and change the automatically generated short video on line to form a final short video file.
And automatically carrying out adaptive template pattern adjustment on the generated video template according to the short video file which is manufactured by the user and the use condition of the video template.
Further, the method for generating the video template according to the domain keywords of the video content for making the short video imported by the user and the video file imported by the user comprises the following steps of:
automatically generating a first type of video template according to the field keywords of the video content of the short video imported by the user and by combining the video elements in the scene video database and the video element database;
and judging whether the user imports the video file after the system logs in real time, and automatically generating a second type of video template by using the video file imported by the user and combining the video elements in the scene video database and the video element database after detecting that the user imports the video file.
The scene video database comprises a plurality of types of scene videos, and the types of the scene videos correspond to the domain types of the domain keywords one to one; the video element database comprises a plurality of video elements, wherein the video elements comprise video elements such as text adding areas, decoration icon inserting areas, shapes and positions of special effect decoration loading areas, and correspondingly loaded text fonts, decoration icons, special effect decoration effects and the like in videos.
Further, automatically generating a first type of video template according to the field keywords of the video content imported by the user for making the short video and by combining the video elements in the scene video database and the video element database, and the method comprises the following steps:
extracting a plurality of scene videos corresponding to the domain keywords in the scene video database according to the domain keywords input by a user to serve as element scenes;
obtaining a network high-quality video through network search, splitting the network high-quality video, and obtaining a video clip and a script corresponding to the network high-quality video; extracting video segments and scripts in the element scene, and replacing the video segments and the scripts with video segments and scripts corresponding to the network high-quality video to generate an initial template;
and adding a text adding area, a decoration icon inserting area, a special effect decoration loading area, and corresponding fonts, decoration icons and special effect decorations on a video interface corresponding to each scene video on the initial template to generate a final first-class video template.
The text adding area, the decoration icon inserting area, the special effect decoration loading area, the corresponding fonts, the decoration icons and the special effect decorations in the first type of video template are in position and content adjustable states, and a user can adjust the position or the content of the first type of video template according to the self requirement.
Further, automatically generating a second type of video template by using the video file imported by the user and combining the video elements in the scene video database and the video element database, wherein the second type of video template comprises the following steps:
when the fact that a user imports a formed video is detected, carrying out video element and video scene splitting on the formed video imported by the user to obtain background audio, subtitle information, pictures and decorative elements in picture frames in the formed video;
sending the background audio, the subtitle information, the pictures and the decoration elements in the picture frames to a database to form a user video element sub-database;
respectively extracting scene videos and video elements in the scene video database and the video element database and scene videos and video elements in the user video element sub-database according to domain keywords input during user registration;
and randomly combining and fusing the video element scene videos and the video elements extracted from the scene video database and the video element database and the scene videos and the video elements extracted from the user video element sub-database to form a second type of video template.
The text adding area, the decoration icon inserting area, the special effect decoration loading area, the corresponding fonts, the decoration icons and the special effect decorations in the second type of video template are in position and content adjustable states, and a user can adjust the position or the content of the second type of video template according to the self requirement.
Further, automatically performing adaptive template pattern adjustment on the generated video template according to the use condition of the short video file which is manufactured by the user and the video template, wherein the adaptive template pattern adjustment comprises the following steps:
monitoring the automatic generation process of the short video of the user and the manual adjustment process of the user aiming at the automatically generated short video in real time;
after a user finishes the current short video works and uploads or downloads the short video works, extracting a video template adopted by the finished short videos;
extracting the scene video and the video elements which are finally adopted in the video template and come from the scene video database, the video element database and the user video element sub-database;
marking the use times and the use time of the finally adopted scene video and video elements in the scene video database, the video element database and the user video element sub-database, wherein the use time is the finish time of the short video work; generating a use index numerical value of each scene video and each video element by using the use times and the use time marks of the scene videos and the video elements in the scene video database, the video element database and the user video element sub-database;
randomly extracting partial scene videos and video elements from the scene videos and the video elements in the scene video database, the video element database and the user video element sub-database which are finally adopted in the current finished video template according to an element extraction rule, and taking the extracted partial scene videos and video elements as elements to be replaced;
extracting the scene videos and the video elements which are different from the elements to be replaced and correspond to the number and the attributes of the elements to be replaced from the scene videos and the video elements of the scene video database, the video element database and the user video element sub-database, and replacing the elements to be replaced by the scene videos and the video elements which meet the preset use index numerical condition to generate a video template after self-adaptive adjustment;
wherein the element extraction rules are as follows:
the ratio ranges between the scene videos and the video elements extracted from the first type of video template and the second type of video template and the total amount of all the scene videos and the video elements adopted in the video template are respectively as follows: the scene video accounts for 20% -33%; the video element proportion range is 15% -25%;
aiming at the same video template, the proportion of extracted scene videos in the adaptive adjustment process of the template after every two adjacent video templates are used is different, and the difference range of the proportion of extracted scene videos every two adjacent video templates is 3% -5%;
aiming at the same video template, the proportion of extracted video elements in the adaptive adjustment process of the template after every two adjacent video templates are used is different, and the difference range of the proportion of extracted video elements in every two adjacent video templates is 4-8%;
the use index value of each scene video and each video element is obtained through the following formula;
Figure 597164DEST_PATH_IMAGE001
wherein the content of the first and second substances,H 1representing a usage index value of each scene video;H 2representing a usage index value for the video element;C 1representing the total use times of the same scene video in all the completed short videos;C z representing the total number of completed short videos;N i is shown asiThe total number of all scene videos adopted in each completed short video;T 1the time interval between the completed short videos used by the same scene video and the short videos completed by using the scene video last time is represented;mrepresenting the number of videos which finish the short video by using the scene elements;C i is shown asiThe total number of the same scene element contained in each completed short video;T ni is shown asiGrouping the use time interval between two adjacent short videos which are manufactured by using the same scene element; m represents the same asAnd the total number of short videos finished by one scene element.
At the same time, whenH 1When the use index exceeds a preset use index threshold value of the scene video, the scene video corresponding to the use index value is the scene video meeting the preset use index value condition; when in useH 2And when the video element usage index is lower than the preset video element usage index threshold value, the video element corresponding to the usage index value is the scene video meeting the preset usage index value condition.
An automatic short video generation system, comprising:
the registration module is used for registering user information by a user and inputting keywords belonging to the field of video content for making short videos in the process of registering the user information; wherein the domain keywords comprise life, history, cultural and educational, entertainment, pets and the like;
the template generation module is used for judging whether a user imports a video file after system login in real time and generating a video template according to the field keywords of the video content imported by the user for making the short video and the video file imported by the user;
the video generation module is used for automatically calling a plurality of video templates matched with the making elements needing to make the short video, which are input by the user, according to the making elements needing to make the short video and the text information input by the user; selecting a corresponding video template according to the preference of the user, and automatically generating a short video according to the video template selected by the user and the manufacturing element; the production elements comprise materials such as pictures, videos, music, animations and characters required by short video production; the user can adjust and change the automatically generated short video on line to form a final short video file.
And the self-adaptive adjusting module is used for automatically carrying out self-adaptive template pattern adjustment on the generated video template according to the use condition of the short video file which is manufactured by the user and the video template.
Further, the template generation module includes:
the first-class template generating module is used for automatically generating a first-class video template according to the field keywords of the video content imported by the user for making the short video and by combining the video elements in the scene video database and the video element database;
and the second type template generating module is used for judging whether a user imports the video file after the system login in real time, and automatically generating a second type video template by using the video file imported by the user and combining the video elements in the scene video database and the video element database after detecting that the user imports the video file.
The scene video database comprises a plurality of types of scene videos, and the types of the scene videos correspond to the domain types of the domain keywords one to one; the video element database comprises a plurality of video elements, wherein the video elements comprise video elements such as text adding areas, decoration icon inserting areas, shapes and positions of special effect decoration loading areas, and correspondingly loaded text fonts, decoration icons, special effect decoration effects and the like in videos.
Further, the first type template generating module comprises:
the first extraction module is used for extracting a plurality of scene videos corresponding to the domain keywords in the scene video database according to the domain keywords input by a user and using the scene videos as element scenes;
the initial template generation module is used for obtaining a network high-quality video through network search, splitting the network high-quality video and obtaining a video clip and a script corresponding to the network high-quality video; extracting video segments and scripts in the element scene, and replacing the video segments and the scripts with video segments and scripts corresponding to the network high-quality video to generate an initial template; the network high-quality video is screened through a preset number threshold of commentary and praise conversion, and when any two of the number of commentary, praise and praise of one network video exceeds the number threshold of commentary and praise conversion at the same time, the network high-quality video is determined to be the network high-quality video.
And the first-class video template generating module is used for adding a text adding area, a decoration icon inserting area, a special effect decoration loading area and corresponding fonts, decoration icons and special effect decorations on a video interface corresponding to each scene video on the initial template to generate a final first-class video template. The special effect decoration comprises a picture-in-picture, a bullet screen, a transition effect and the like.
The text adding area, the decoration icon inserting area, the special effect decoration loading area, the corresponding fonts, the decoration icons and the special effect decorations in the first type of video template are in position and content adjustable states, and a user can adjust the position or the content of the first type of video template according to the self requirement.
Further, the second type template generating module comprises:
the video splitting module is used for splitting video elements and video scenes of the formed video imported by the user when the user is detected to import the formed video, so as to obtain background audio, subtitle information, pictures and decorative elements in picture frames in the formed video;
the database forming module is used for sending the background audio, the subtitle information, the pictures and the decoration elements in the picture frames to a database to form a user video element sub-database;
the second extraction module is used for respectively extracting the scene videos and the video elements in the scene video database and the video element database and the scene videos and the video elements in the user video element sub-database according to the field keywords input during user registration;
and the second type video template generating module is used for randomly combining and fusing the video element scene videos and the video elements extracted from the scene video database and the video element database and the scene videos and the video elements extracted from the user video element sub-database to form a second type video template.
The text adding area, the decoration icon inserting area, the special effect decoration loading area, the corresponding fonts, the decoration icons and the special effect decorations in the second type of video template are in position and content adjustable states, and a user can adjust the position or the content of the second type of video template according to the self requirement.
Further, the adaptive adjustment module comprises:
the real-time monitoring module is used for monitoring the automatic generation process of the short video of the user and the manual adjustment process of the user aiming at the automatically generated short video in real time;
the video template extraction module is used for extracting a video template adopted by the manufactured short video after the user finishes the current short video work and uploads or downloads the short video work;
the video element extraction module is used for extracting the finally adopted scene videos and video elements in the scene video database, the video element database and the user video element sub-database in the video template;
the marking module is used for marking the use times and the use time of the finally adopted scene videos and video elements in the scene video database, the video element database and the user video element sub-database, wherein the use time is the finish time of the short video work; generating a use index numerical value of each scene video and each video element by using the use times and the use time marks of the scene videos and the video elements in the scene video database, the video element database and the user video element sub-database;
a replacement module, configured to randomly extract, according to an element extraction rule, part of scene videos and video elements from scene videos and video elements in the scene video database, the video element database, and the user video element sub-database that are finally adopted in the current completed video template, and use the extracted part of scene videos and video elements as elements to be replaced;
the automatic adjustment module is used for extracting the number and attributes of elements to be replaced from scene videos and video elements which are different from the elements to be replaced and in the scene video database, the video element database and the user video element sub-database, and replacing the elements to be replaced by the scene videos and the video elements which meet preset use index numerical conditions to generate a video template after self-adaptive adjustment;
wherein the element extraction rules are as follows:
the ratio ranges between the scene videos and the video elements extracted from the first type of video template and the second type of video template and the total amount of all the scene videos and the video elements adopted in the video template are respectively as follows: the scene video accounts for 20% -33%; the video element proportion range is 15% -25%;
aiming at the same video template, the proportion of extracted scene videos in the adaptive adjustment process of the template after every two adjacent video templates are used is different, and the difference range of the proportion of extracted scene videos every two adjacent video templates is 3% -5%;
aiming at the same video template, the proportion of extracted video elements in the adaptive adjustment process of the template after every two adjacent video templates are used is different, and the difference range of the proportion of extracted video elements in every two adjacent video templates is 4-8%;
the use index value of each scene video and each video element is obtained through the following formula;
Figure 563240DEST_PATH_IMAGE002
wherein the content of the first and second substances,H 1representing a usage index value of each scene video;H 2representing a usage index value for the video element;C 1representing the total use times of the same scene video in all the completed short videos;C z representing the total number of completed short videos;N i is shown asiThe total number of all scene videos adopted in each completed short video;T 1the time interval between the completed short videos used by the same scene video and the short videos completed by using the scene video last time is represented;mrepresenting the number of videos which finish the short video by using the scene elements;C i is shown asiThe total number of the same scene element contained in each completed short video;T ni is shown asiIn-use between two adjacent short videos made using the same scene elementSpacing; and m represents the total number of short videos finished by using the same scene element.
At the same time, whenH 1When the use index exceeds a preset use index threshold value of the scene video, the scene video corresponding to the use index value is the scene video meeting the preset use index value condition; when in useH 2And when the video element usage index is lower than the preset video element usage index threshold value, the video element corresponding to the usage index value is the scene video meeting the preset usage index value condition.
The invention has the beneficial effects that:
according to the automatic short video generation method and system, the video template with high matching degree with the user application is automatically generated during user registration according to the mode that the user frequently makes keywords in the field to which the video belongs, the formed video template is provided for later-stage video production of the user, the finished product efficiency of the video produced by the user can be effectively improved, and the video production time is effectively reduced. On the other hand, by establishing the user video element sub-database and the video template mode self-adaptive adjustment mode, the diversity of the video template is effectively improved, the diversity of user video production is further improved, and the problems that the diversity of short video works is low and the video quality is gradually reduced along with the low diversity caused by the fixed video template mode are effectively prevented. Meanwhile, the diversity of the video templates is improved through the user video element sub-database, and the problems that under the condition of fixing the video templates, the production amount of videos generated by manually adjusting a large number of video elements when a user needs to increase the diversity is gradually increased along with the operation duration of the user, and the production efficiency is reduced are completely solved. Furthermore, the method and the system for automatically generating the short video can always keep high efficiency and low time consumption of the user for making the video, and can avoid the problem that the video making efficiency is reduced along with the increase of the system service time because the user gradually needs to manually adjust the video time due to poor video diversity.
Drawings
FIG. 1 is a first flow chart of the method of the present invention;
FIG. 2 is a second flow chart of the method of the present invention;
fig. 3 is a system block diagram of the system of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
The embodiment of the invention provides an automatic short video generation method, as shown in fig. 1, the automatic short video generation method includes:
s1, the user registers the user information and inputs keywords belonging to the field of the video content for making the short video in the process of registering the user information; wherein the domain keywords comprise life, history, cultural and educational, entertainment, pets and the like;
s2, judging whether a user imports a video file after system login in real time, and generating a video template according to the field keywords of the video content imported by the user for making short videos and the video file imported by the user;
s3, automatically calling a plurality of video templates matched with the making elements needing to make the short video, which are input by the user, according to the making elements needing to make the short video and the text information input by the user; selecting a corresponding video template according to the preference of the user, and automatically generating a short video according to the video template selected by the user and the manufacturing element; the production elements comprise materials such as pictures, videos, music, animations and characters required by short video production; the user can adjust and change the automatically generated short video on line to form a final short video file.
And S4, automatically carrying out adaptive template pattern adjustment on the generated video template according to the short video file which is manufactured by the user and the use condition of the video template.
The working principle of the technical scheme is as follows: firstly, a user registers user information, and inputs keywords belonging to the field of video content for making short videos in the process of registering the user information; wherein the domain keywords comprise life, history, cultural and educational, entertainment, pets and the like; then, judging whether a user imports a video file after system login in real time, and generating a video template according to the field keywords of the video content imported by the user for making the short video and the video file imported by the user; then, automatically calling a plurality of video templates matched with the making elements needing to make the short video, which are input by the user, according to the making elements needing to make the short video and the text information input by the user; selecting a corresponding video template according to the preference of the user, and automatically generating a short video according to the video template selected by the user and the manufacturing element; the production elements comprise materials such as pictures, videos, music, animations and characters required by short video production; the user can adjust and change the automatically generated short video on line to form a final short video file. And finally, automatically carrying out adaptive template pattern adjustment on the generated video template according to the short video file which is manufactured by the user and the use condition of the video template.
The effect of the above technical scheme is as follows: according to the automatic short video generation method provided by the embodiment, the video template with higher matching degree with the user application is automatically generated during user registration according to the way that the user frequently makes keywords in the field to which the video belongs, the formed video template is provided for the later-stage video production of the user, the finished product efficiency of producing the video by the user can be effectively improved, and the video production time is effectively reduced. On the other hand, by establishing the user video element sub-database and the video template mode self-adaptive adjustment mode, the diversity of the video template is effectively improved, the diversity of user video production is further improved, and the problems that the diversity of short video works is low and the video quality is gradually reduced along with the low diversity caused by the fixed video template mode are effectively prevented. Meanwhile, the diversity of the video templates is improved through the user video element sub-database, and the problems that under the condition of fixing the video templates, the production amount of videos generated by manually adjusting a large number of video elements when a user needs to increase the diversity is gradually increased along with the operation duration of the user, and the production efficiency is reduced are completely solved. Furthermore, by the automatic short video generation method provided by the embodiment, high efficiency and low time consumption of a user for making a video can be always maintained, and the problem that the video making efficiency is reduced along with the increase of the system service time because the user gradually needs to manually adjust the video time due to poor video diversity is avoided.
An embodiment of the present invention, as shown in fig. 2, determines in real time whether a user imports a video file after system login, and generates a video template according to the domain keyword of the video content imported by the user to make a short video and the video file imported by the user, including:
s201, automatically generating a first type of video template according to the field keywords of the video content of the short video imported by the user and by combining the video elements in the scene video database and the video element database;
s202, judging whether a user imports a video file after system login in real time, and automatically generating a second type of video template by using the video file imported by the user and combining video elements in a scene video database and a video element database after detecting that the user imports the video file.
The scene video database comprises a plurality of types of scene videos, and the types of the scene videos correspond to the domain types of the domain keywords one to one; the video element database comprises a plurality of video elements, wherein the video elements comprise video elements such as text adding areas, decoration icon inserting areas, shapes and positions of special effect decoration loading areas, and correspondingly loaded text fonts, decoration icons, special effect decoration effects and the like in videos.
The method for automatically generating the first type of video template by combining the field keywords with the video elements in the scene video database and the video element database according to the video content of the short video imported by the user comprises the following steps:
s2011, extracting a plurality of scene videos corresponding to the domain keywords in the scene video database according to the domain keywords input by a user to serve as element scenes;
s2012, obtaining a network high-quality video through network search, splitting the network high-quality video, and obtaining video clips and scripts corresponding to the network high-quality video; extracting video segments and scripts in the element scene, and replacing the video segments and the scripts with video segments and scripts corresponding to the network high-quality video to generate an initial template;
and S2013, adding a text adding area, a decoration icon inserting area, a special effect decoration loading area and corresponding fonts, decoration icons and special effect decorations to a video interface corresponding to each scene video on the initial template to generate a final first-class video template.
The text adding area, the decoration icon inserting area, the special effect decoration loading area, the corresponding fonts, the decoration icons and the special effect decorations in the first type of video template are in position and content adjustable states, and a user can adjust the position or the content of the first type of video template according to the self requirement.
On the other hand, the method for automatically generating the second type of video template by using the video file imported by the user and combining the video elements in the scene video database and the video element database comprises the following steps:
s2021, when the fact that the user introduces the formed video is detected, splitting video elements and video scenes of the formed video introduced by the user to obtain background audio, subtitle information, pictures and decoration elements in picture frames in the formed video;
s2022, sending the background audio, the subtitle information, the picture and the decoration elements in the picture frame to a database to form a user video element sub-database;
s2023, respectively extracting scene videos and video elements in the scene video database and the video element database and scene videos and video elements in the user video element sub-database according to domain keywords input during user registration;
s2024, randomly combining and fusing the video element scene videos and the video elements extracted from the scene video database and the video element database and the scene videos and the video elements extracted from the user video element sub-database to form a second type of video template.
The text adding area, the decoration icon inserting area, the special effect decoration loading area, the corresponding fonts, the decoration icons and the special effect decorations in the second type of video template are in position and content adjustable states, and a user can adjust the position or the content of the second type of video template according to the self requirement.
The working principle of the technical scheme is as follows: firstly, automatically generating a first type of video template according to the field keywords of the video content for making the short video imported by a user and by combining the video elements in a scene video database and a video element database; and then, judging whether the user imports the video file after the system logs in real time, and automatically generating a second type of video template by using the video file imported by the user and combining the video elements in the scene video database and the video element database after detecting that the user imports the video file.
The effect of the above technical scheme is as follows: by establishing the user video element sub-database mode, the diversity of the video template is effectively improved, the diversity of user video production is further improved, and the problems that the diversity of short video works is low and the video quality is gradually reduced along with the low diversity caused by the fixed style of the video template are effectively prevented. Meanwhile, the diversity of the video templates is improved through the user video element sub-database, and the problems that under the condition of fixing the video templates, the production amount of videos generated by manually adjusting a large number of video elements when a user needs to increase the diversity is gradually increased along with the operation duration of the user, and the production efficiency is reduced are completely solved. Furthermore, the above-mentioned method provided by this embodiment can always maintain high efficiency and low time consumption for the user to produce the video, and the problem that the user gradually needs to manually adjust the video time to increase due to poor video diversity and the video production efficiency decreases along with the increase of the system service time does not occur.
According to the embodiment of the invention, the self-adaptive template pattern adjustment is automatically carried out on the generated video template according to the use condition of the short video file which is finished by the user and is combined with the video template, and the self-adaptive template pattern adjustment comprises the following steps:
s401, monitoring an automatic short video generation process of a user and a manual adjustment process of the user aiming at the automatically generated short video in real time;
s402, after the user finishes the current short video works and uploads or downloads the short video works, extracting a video template adopted by the finished short videos;
s403, extracting the finally adopted scene videos and video elements in the scene video database, the video element database and the user video element sub-database in the video template;
s404, marking the use times and the use time of the finally adopted scene videos and video elements in the scene video database, the video element database and the user video element sub-database, wherein the use time is the finish time of the short video work; generating a use index numerical value of each scene video and each video element by using the use times and the use time marks of the scene videos and the video elements in the scene video database, the video element database and the user video element sub-database;
s405, randomly extracting partial scene videos and video elements from the scene videos and the video elements in the scene video database, the video element database and the user video element sub-database which are finally adopted in the current finished video template according to an element extraction rule, and taking the extracted partial scene videos and video elements as elements to be replaced;
s406, extracting the number and attributes of the elements to be replaced from the scene videos and the video elements different from the elements to be replaced in the scene video database, the video element database and the user video element sub-database, and replacing the elements to be replaced by the scene videos and the video elements meeting the preset use index numerical conditions to generate a video template after self-adaptive adjustment;
wherein the element extraction rules are as follows:
the ratio ranges between the scene videos and the video elements extracted from the first type of video template and the second type of video template and the total amount of all the scene videos and the video elements adopted in the video template are respectively as follows: the scene video accounts for 20% -33%; the video element proportion range is 15% -25%;
aiming at the same video template, the proportion of extracted scene videos in the adaptive adjustment process of the template after every two adjacent video templates are used is different, and the difference range of the proportion of extracted scene videos every two adjacent video templates is 3% -5%;
aiming at the same video template, the proportion of extracted video elements in the adaptive adjustment process of the template after every two adjacent video templates are used is different, and the difference range of the proportion of extracted video elements in every two adjacent video templates is 4-8%;
the use index value of each scene video and each video element is obtained through the following formula;
Figure 139715DEST_PATH_IMAGE003
wherein the content of the first and second substances,H 1representing a usage index value of each scene video;H 2representing a usage index value for the video element;C 1representing the total use times of the same scene video in all the completed short videos;C z representing the total number of completed short videos;N i is shown asiThe total number of all scene videos adopted in each completed short video;T 1the time interval between the completed short videos used by the same scene video and the short videos completed by using the scene video last time is represented;mrepresenting the number of videos which finish the short video by using the scene elements;C i is shown asiThe total number of the same scene element contained in each completed short video;T ni is shown asiGrouping the use time interval between two adjacent short videos which are manufactured by using the same scene element; and m represents the total number of short videos finished by using the same scene element. Wherein, theH 1AndH 2the expressed use index value is a constant value and has no unit and dimension, and all the calculation parameters are calculated in the use index calculation processIs a numerical value without units and dimensions.
At the same time, whenH 1When the use index exceeds a preset use index threshold value of the scene video, the scene video corresponding to the use index value is the scene video meeting the preset use index value condition; when in useH 2And when the video element usage index is lower than the preset video element usage index threshold value, the video element corresponding to the usage index value is the scene video meeting the preset usage index value condition.
The working principle of the technical scheme is as follows: firstly, monitoring the automatic generation process of the short video of a user and the manual adjustment process of the user aiming at the automatically generated short video in real time; then, after the user finishes the current short video works and uploads or downloads the short video works, extracting a video template adopted by the finished short videos; extracting the scene video and the video elements which are finally adopted in the video template and come from the scene video database, the video element database and the user video element sub-database; marking the use times and the use time of the finally adopted scene video and video elements in the scene video database, the video element database and the user video element sub-database, wherein the use time is the finish time of the short video work; generating a use index numerical value of each scene video and each video element by using the use times and the use time marks of the scene videos and the video elements in the scene video database, the video element database and the user video element sub-database; then, randomly extracting partial scene videos and video elements according to element extraction rules from the scene videos and video elements in the scene video database, the video element database and the user video element sub-database which are finally adopted in the current finished video template, and taking the extracted partial scene videos and video elements as elements to be replaced; and finally, extracting the scene videos and the video elements which are different from the elements to be replaced and correspond to the number and the attributes of the elements to be replaced from the scene video database, the video element database and the user video element sub-database, and replacing the elements to be replaced by the scene videos and the video elements which meet the preset use index numerical conditions to generate the video template after self-adaptive adjustment.
The effect of the above technical scheme is as follows: the above-mentioned mode of this embodiment passes through video template mode self-adaptation adjustment mode, effectively improves the variety of video template, and then improves the variety of user's video preparation, and the problem that effectively prevents to be lower because of the short video works variety that the video template style leads to is fixed, video quality along with the variety is lower and reduces gradually takes place. Meanwhile, the diversity of the video templates is improved through the user video element sub-database, and the problems that under the condition of fixing the video templates, the production amount of videos generated by manually adjusting a large number of video elements when a user needs to increase the diversity is gradually increased along with the operation duration of the user, and the production efficiency is reduced are completely solved. Furthermore, by the automatic short video generation method provided by the embodiment, high efficiency and low time consumption of a user for making a video can be always maintained, and the problem that the video making efficiency is reduced along with the increase of the system service time because the user gradually needs to manually adjust the video time due to poor video diversity is avoided.
Meanwhile, the diversity of the video template can be improved to the maximum extent by carrying out the self-adaptive adjustment of the video template through the element extraction rule. And under the condition that the quantity and the diversity of the video elements and the scene videos cannot be increased infinitely all the time, the video template generated by the element extraction rule can effectively limit the replacement and the use frequency of different video production elements while ensuring that the diversity is strong enough, effectively prevent the problem that the use frequency of the video elements and the scene videos is too high due to too many replacement parts, further under the condition that the quantity and the diversity of the video elements and the scene videos are limited, the problem that the video elements and the scene videos meet the replacement condition and gradually decrease along with the increase of the subsequent video production time is caused, and further effectively prevent the problem that the diversity of the subsequent videos is gradually decreased along with the increase of the video production working time due to the gradual decrease of the video elements and the scene videos. Therefore, the adaptive adjustment of the video template through the extraction rule can keep the duration of the video diversity to the maximum extent, and the problem of excessive use of various scene videos and video elements in the adaptive adjustment process of the video template occurs in an effective mode.
On the other hand, the use index values of the video elements and the scene videos obtained by the formula are combined in a mode that the use actual conditions of the video elements in the video making process of the user are combined with the watching conditions of the user uploading the videos, so that the use indexes and the viewing indexes of the scene videos and the video elements in the video making process and the video browsing process are effectively reflected, an effective diversity adjustment basis can be provided for the self-adaptive adjustment of the subsequent video template to the utmost extent, the diversity degree of the short video making and the duration of the diversity of the self-adaptive adjustment of the template are improved, the video quality of the automatic generation of the videos is improved to the utmost extent, and the making duration of the manual participation of the user is reduced.
An embodiment of the present invention provides an automatic short video generation system, and as shown in fig. 3, the automatic short video generation system includes:
the registration module is used for registering user information by a user and inputting keywords belonging to the field of video content for making short videos in the process of registering the user information; wherein the domain keywords comprise life, history, cultural and educational, entertainment, pets and the like;
the template generation module is used for judging whether a user imports a video file after system login in real time and generating a video template according to the field keywords of the video content imported by the user for making the short video and the video file imported by the user;
the video generation module is used for automatically calling a plurality of video templates matched with the making elements needing to make the short video, which are input by the user, according to the making elements needing to make the short video and the text information input by the user; selecting a corresponding video template according to the preference of the user, and automatically generating a short video according to the video template selected by the user and the manufacturing element; the production elements comprise materials such as pictures, videos, music, animations and characters required by short video production; the user can adjust and change the automatically generated short video on line to form a final short video file.
And the self-adaptive adjusting module is used for automatically carrying out self-adaptive template pattern adjustment on the generated video template according to the use condition of the short video file which is manufactured by the user and the video template.
The working principle of the technical scheme is as follows: firstly, a user registers user information through a registration module, and keywords belonging to the field of video content for making short videos are input in the user information registration process; wherein the domain keywords comprise life, history, cultural and educational, entertainment, pets and the like; then, judging whether a user imports a video file after system login in real time through a template generation module, and generating a video template according to the field keywords of the video content imported by the user for making the short video and the video file imported by the user; then, automatically calling a plurality of video templates matched with the making elements needing to make the short video, which are input by the user, by utilizing a video generating module according to the making elements needing to make the short video and the text information, which are input by the user; selecting a corresponding video template according to the preference of the user, and automatically generating a short video according to the video template selected by the user and the manufacturing element; the production elements comprise materials such as pictures, videos, music, animations and characters required by short video production; the user can adjust and change the automatically generated short video on line to form a final short video file. And finally, automatically carrying out adaptive template pattern adjustment on the generated video template through an adaptive adjustment module according to the short video file which is manufactured by the user and the use condition of the video template.
The effect of the above technical scheme is as follows: according to the automatic short video generation system provided by the embodiment, the video template with higher matching degree with the user application is automatically generated during user registration according to the way that the user frequently makes keywords in the field to which the video belongs, the formed video template is provided for the later-stage video production of the user, the finished product efficiency of producing the video by the user can be effectively improved, and the video production time is effectively reduced. On the other hand, by establishing the user video element sub-database and the video template mode self-adaptive adjustment mode, the diversity of the video template is effectively improved, the diversity of user video production is further improved, and the problems that the diversity of short video works is low and the video quality is gradually reduced along with the low diversity caused by the fixed video template mode are effectively prevented. Meanwhile, the diversity of the video templates is improved through the user video element sub-database, and the problems that under the condition of fixing the video templates, the production amount of videos generated by manually adjusting a large number of video elements when a user needs to increase the diversity is gradually increased along with the operation duration of the user, and the production efficiency is reduced are completely solved. Furthermore, the automatic short video generation system provided by the embodiment can always maintain high efficiency and low time consumption for the user to produce the video, and the problem that the video production efficiency is reduced along with the increase of the system service time because the user gradually needs to manually adjust the increase of the video time due to poor video diversity does not occur.
In an embodiment of the present invention, the template generating module includes:
the first-class template generating module is used for automatically generating a first-class video template according to the field keywords of the video content imported by the user for making the short video and by combining the video elements in the scene video database and the video element database;
and the second type template generating module is used for judging whether a user imports the video file after the system login in real time, and automatically generating a second type video template by using the video file imported by the user and combining the video elements in the scene video database and the video element database after detecting that the user imports the video file.
The scene video database comprises a plurality of types of scene videos, and the types of the scene videos correspond to the domain types of the domain keywords one to one; the video element database comprises a plurality of video elements, wherein the video elements comprise video elements such as text adding areas, decoration icon inserting areas, shapes and positions of special effect decoration loading areas, and correspondingly loaded text fonts, decoration icons, special effect decoration effects and the like in videos.
Wherein the first type template generating module comprises:
the first extraction module is used for extracting a plurality of scene videos corresponding to the domain keywords in the scene video database according to the domain keywords input by a user and using the scene videos as element scenes;
the initial template generation module is used for obtaining a network high-quality video through network search, splitting the network high-quality video and obtaining a video clip and a script corresponding to the network high-quality video; extracting video segments and scripts in the element scene, and replacing the video segments and the scripts with video segments and scripts corresponding to the network high-quality video to generate an initial template;
and the first-class video template generating module is used for adding a text adding area, a decoration icon inserting area, a special effect decoration loading area and corresponding fonts, decoration icons and special effect decorations on a video interface corresponding to each scene video on the initial template to generate a final first-class video template.
The text adding area, the decoration icon inserting area, the special effect decoration loading area, the corresponding fonts, the decoration icons and the special effect decorations in the first type of video template are in position and content adjustable states, and a user can adjust the position or the content of the first type of video template according to the self requirement.
Specifically, the second-class template generating module includes:
the video splitting module is used for splitting video elements and video scenes of the formed video imported by the user when the user is detected to import the formed video, so as to obtain background audio, subtitle information, pictures and decorative elements in picture frames in the formed video;
the database forming module is used for sending the background audio, the subtitle information, the pictures and the decoration elements in the picture frames to a database to form a user video element sub-database;
the second extraction module is used for respectively extracting the scene videos and the video elements in the scene video database and the video element database and the scene videos and the video elements in the user video element sub-database according to the field keywords input during user registration;
and the second type video template generating module is used for randomly combining and fusing the video element scene videos and the video elements extracted from the scene video database and the video element database and the scene videos and the video elements extracted from the user video element sub-database to form a second type video template.
The text adding area, the decoration icon inserting area, the special effect decoration loading area, the corresponding fonts, the decoration icons and the special effect decorations in the second type of video template are in position and content adjustable states, and a user can adjust the position or the content of the second type of video template according to the self requirement.
The working principle of the technical scheme is as follows: the operation process of the template generation module comprises the following steps:
firstly, automatically generating a first-class video template by a first-class template generation module according to the field keywords of the video content for making short videos imported by a user and by combining video elements in a scene video database and a video element database;
and then, judging whether the user imports the video file after the system login by using a second type template generation module in real time, and automatically generating a second type video template by using the video file imported by the user in combination with the video elements in the scene video database and the video element database after detecting that the user imports the video file.
The scene video database comprises a plurality of types of scene videos, and the types of the scene videos correspond to the domain types of the domain keywords one to one; the video element database comprises a plurality of video elements, wherein the video elements comprise video elements such as text adding areas, decoration icon inserting areas, shapes and positions of special effect decoration loading areas, and correspondingly loaded text fonts, decoration icons, special effect decoration effects and the like in videos.
The first type template generation module operation process comprises the following steps:
firstly, extracting a plurality of scene videos corresponding to domain keywords in a scene video database as element scenes by an extraction module I according to the domain keywords input by a user;
then, an initial template generation module searches through a network to obtain a network high-quality video, splits the network high-quality video and obtains a video clip and a script corresponding to the network high-quality video; extracting video segments and scripts in the element scene, and replacing the video segments and the scripts with video segments and scripts corresponding to the network high-quality video to generate an initial template;
and finally, adding a text adding area, a decoration icon inserting area, a special effect decoration loading area, and corresponding fonts, decoration icons and special effect decorations on a video interface corresponding to each scene video on the initial template by adopting a first-class video template generating module to generate a final first-class video template.
The text adding area, the decoration icon inserting area, the special effect decoration loading area, the corresponding fonts, the decoration icons and the special effect decorations in the first type of video template are in position and content adjustable states, and a user can adjust the position or the content of the first type of video template according to the self requirement.
Specifically, the operation process of the second type template generation module includes:
firstly, when a video splitting module detects that a user introduces a formed video, splitting video elements and video scenes of the formed video introduced by the user to obtain background audio, subtitle information, pictures and decorative elements in picture frames in the formed video;
then, a database forming module is used for sending the background audio, the subtitle information, the pictures and the decorative elements in the picture frames to a database to form a user video element sub-database;
then, respectively extracting scene videos and video elements in the scene video database and the video element database and scene videos and video elements in the user video element sub-database by adopting two extraction modules according to domain keywords input by a user during registration;
and finally, randomly combining and fusing the video element scene videos and the video elements extracted from the scene video database and the video element database and the scene videos and the video elements extracted from the user video element sub-database through a second type video template generating module to form a second type video template.
The text adding area, the decoration icon inserting area, the special effect decoration loading area, the corresponding fonts, the decoration icons and the special effect decorations in the second type of video template are in position and content adjustable states, and a user can adjust the position or the content of the second type of video template according to the self requirement.
The effect of the above technical scheme is as follows: by establishing the user video element sub-database mode, the diversity of the video template is effectively improved, the diversity of user video production is further improved, and the problems that the diversity of short video works is low and the video quality is gradually reduced along with the low diversity caused by the fixed style of the video template are effectively prevented. Meanwhile, the diversity of the video templates is improved through the user video element sub-database, and the problems that under the condition of fixing the video templates, the production amount of videos generated by manually adjusting a large number of video elements when a user needs to increase the diversity is gradually increased along with the operation duration of the user, and the production efficiency is reduced are completely solved. Furthermore, the above-mentioned method provided by this embodiment can always maintain high efficiency and low time consumption for the user to produce the video, and the problem that the user gradually needs to manually adjust the video time to increase due to poor video diversity and the video production efficiency decreases along with the increase of the system service time does not occur.
In an embodiment of the present invention, the adaptive adjustment module includes:
the real-time monitoring module is used for monitoring the automatic generation process of the short video of the user and the manual adjustment process of the user aiming at the automatically generated short video in real time;
the video template extraction module is used for extracting a video template adopted by the manufactured short video after the user finishes the current short video work and uploads or downloads the short video work;
the video element extraction module is used for extracting the finally adopted scene videos and video elements in the scene video database, the video element database and the user video element sub-database in the video template;
the marking module is used for marking the use times and the use time of the finally adopted scene videos and video elements in the scene video database, the video element database and the user video element sub-database, wherein the use time is the finish time of the short video work; generating a use index numerical value of each scene video and each video element by using the use times and the use time marks of the scene videos and the video elements in the scene video database, the video element database and the user video element sub-database;
a replacement module, configured to randomly extract, according to an element extraction rule, part of scene videos and video elements from scene videos and video elements in the scene video database, the video element database, and the user video element sub-database that are finally adopted in the current completed video template, and use the extracted part of scene videos and video elements as elements to be replaced;
the automatic adjustment module is used for extracting the number and attributes of elements to be replaced from scene videos and video elements which are different from the elements to be replaced and in the scene video database, the video element database and the user video element sub-database, and replacing the elements to be replaced by the scene videos and the video elements which meet preset use index numerical conditions to generate a video template after self-adaptive adjustment;
wherein the element extraction rules are as follows:
the ratio ranges between the scene videos and the video elements extracted from the first type of video template and the second type of video template and the total amount of all the scene videos and the video elements adopted in the video template are respectively as follows: the scene video accounts for 20% -33%; the video element proportion range is 15% -25%;
aiming at the same video template, the proportion of extracted scene videos in the adaptive adjustment process of the template after every two adjacent video templates are used is different, and the difference range of the proportion of extracted scene videos every two adjacent video templates is 3% -5%;
aiming at the same video template, the proportion of extracted video elements in the adaptive adjustment process of the template after every two adjacent video templates are used is different, and the difference range of the proportion of extracted video elements in every two adjacent video templates is 4-8%;
the use index value of each scene video and each video element is obtained through the following formula;
Figure 331661DEST_PATH_IMAGE004
wherein the content of the first and second substances,H 1representing a usage index value of each scene video;H 2representing a usage index value for the video element;C 1representing the total use times of the same scene video in all the completed short videos;C z representing the total number of completed short videos;N i is shown asiThe total number of all scene videos adopted in each completed short video;T 1the time interval between the completed short videos used by the same scene video and the short videos completed by using the scene video last time is represented;mrepresenting the number of videos which finish the short video by using the scene elements;C i is shown asiThe total number of the same scene element contained in each completed short video;T ni is shown asiGrouping the use time interval between two adjacent short videos which are manufactured by using the same scene element; and m represents the total number of short videos finished by using the same scene element. Wherein, theH 1AndH 2the expressed use index value is a constant value and has no unit and dimension, and all the calculation parameters are numerical values without the unit and the dimension in the use index calculation process.
At the same time, whenH 1Exceeding preset scene video usageWhen the index threshold value is reached, the scene video corresponding to the use index value is the scene video meeting the preset use index value condition; when in useH 2And when the video element usage index is lower than the preset video element usage index threshold value, the video element corresponding to the usage index value is the scene video meeting the preset usage index value condition.
The operation principle of the technical scheme is as follows: firstly, monitoring an automatic short video generation process of a user and a manual adjustment process of the user aiming at the automatically generated short video in real time through a real-time monitoring module; then, after the user finishes the current short video works and uploads or downloads the short video works, a video template adopted by the finished short videos is extracted by a video template extraction module; then, a video element extraction module is used for extracting the finally adopted scene video and video elements in the video template, wherein the scene video and the video elements are from the scene video database, the video element database and the user video element sub-database; then, a marking module is adopted to mark the use times and the use time of the finally adopted scene videos and video elements in the scene video database, the video element database and the user video element sub-database, wherein the use time is the finish time of the short video work; generating a use index numerical value of each scene video and each video element by using the use times and the use time marks of the scene videos and the video elements in the scene video database, the video element database and the user video element sub-database; then, a replacement module is utilized to randomly extract partial scene videos and video elements from the scene videos and the video elements in the scene video database, the video element database and the user video element sub-database which are finally adopted in the current completed video template according to an element extraction rule, and the extracted partial scene videos and video elements are taken as elements to be replaced; finally, extracting the number and attributes of the elements to be replaced corresponding to the elements to be replaced from the scene videos and the video elements different from the elements to be replaced in the scene video database, the video element database and the user video element sub-database through an automatic adjusting module, and replacing the elements to be replaced by the scene videos and the video elements meeting preset use index numerical conditions to generate a video template after self-adaptive adjustment;
the effect of the above technical scheme is as follows: the above-mentioned mode of this embodiment passes through video template mode self-adaptation adjustment mode, effectively improves the variety of video template, and then improves the variety of user's video preparation, and the problem that effectively prevents to be lower because of the short video works variety that the video template style leads to is fixed, video quality along with the variety is lower and reduces gradually takes place. Meanwhile, the diversity of the video templates is improved through the user video element sub-database, and the problems that under the condition of fixing the video templates, the production amount of videos generated by manually adjusting a large number of video elements when a user needs to increase the diversity is gradually increased along with the operation duration of the user, and the production efficiency is reduced are completely solved. Furthermore, by the automatic short video generation method provided by the embodiment, high efficiency and low time consumption of a user for making a video can be always maintained, and the problem that the video making efficiency is reduced along with the increase of the system service time because the user gradually needs to manually adjust the video time due to poor video diversity is avoided.
Meanwhile, the diversity of the video template can be improved to the maximum extent by carrying out the self-adaptive adjustment of the video template through the element extraction rule. And under the condition that the quantity and the diversity of the video elements and the scene videos cannot be increased infinitely all the time, the video template generated by the element extraction rule can effectively limit the replacement and the use frequency of different video production elements while ensuring that the diversity is strong enough, effectively prevent the problem that the use frequency of the video elements and the scene videos is too high due to too many replacement parts, further under the condition that the quantity and the diversity of the video elements and the scene videos are limited, the problem that the video elements and the scene videos meet the replacement condition and gradually decrease along with the increase of the subsequent video production time is caused, and further effectively prevent the problem that the diversity of the subsequent videos is gradually decreased along with the increase of the video production working time due to the gradual decrease of the video elements and the scene videos. Therefore, the adaptive adjustment of the video template through the extraction rule can keep the duration of the video diversity to the maximum extent, and the problem of excessive use of various scene videos and video elements in the adaptive adjustment process of the video template occurs in an effective mode.
On the other hand, the use index values of the video elements and the scene videos obtained by the formula are combined in a mode that the use actual conditions of the video elements in the video making process of the user are combined with the watching conditions of the user uploading the videos, so that the use indexes and the viewing indexes of the scene videos and the video elements in the video making process and the video browsing process are effectively reflected, an effective diversity adjustment basis can be provided for the self-adaptive adjustment of the subsequent video template to the utmost extent, the diversity degree of the short video making and the duration of the diversity of the self-adaptive adjustment of the template are improved, the video quality of the automatic generation of the videos is improved to the utmost extent, and the making duration of the manual participation of the user is reduced.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (10)

1. An automatic short video generation method, comprising:
a user registers user information and inputs keywords belonging to the field of video content for making short videos in the process of registering the user information;
judging whether a user imports a video file after system login in real time, and generating a video template according to the field keywords of the video content imported by the user for making the short video and the video file imported by the user;
automatically calling a plurality of video templates matched with the making elements needing to make the short video input by the user according to the making elements needing to make the short video and the text information input by the user, and automatically generating the short video according to the video template selected by the user and the making elements;
and automatically carrying out adaptive template pattern adjustment on the generated video template according to the short video file which is manufactured by the user and the use condition of the video template.
2. The method for automatically generating short videos according to claim 1, wherein the step of judging whether a user imports a video file after system login in real time and generating a video template according to the domain keyword of the video content imported by the user for making short videos and the video file imported by the user comprises the steps of:
automatically generating a first type of video template according to the field keywords of the video content of the short video imported by the user and by combining the video elements in the scene video database and the video element database;
judging whether a user imports a video file after system login in real time, and automatically generating a second type of video template by using the video file imported by the user and combining video elements in a scene video database and a video element database after the video file imported by the user is detected;
the scene video database comprises a plurality of types of scene videos, and the types of the scene videos correspond to the domain types of the domain keywords one to one; the video element database comprises a plurality of video elements, wherein the video elements comprise shapes and positions of a text adding area, a decoration icon inserting area and a special effect decoration loading area in a video, and correspondingly loaded character fonts, decoration icons and special effect decoration effects.
3. The method for automatically generating short videos according to claim 2, wherein automatically generating a first type of video template according to the domain keyword of the video content imported by the user for making short videos and in combination with the video elements in the scene video database and the video element database comprises:
extracting a plurality of scene videos corresponding to the domain keywords in the scene video database according to the domain keywords input by a user to serve as element scenes;
obtaining a network high-quality video through network search, splitting the network high-quality video, and obtaining a video clip and a script corresponding to the network high-quality video; extracting video segments and scripts in the element scene, and replacing the video segments and the scripts with video segments and scripts corresponding to the network high-quality video to generate an initial template;
and adding a text adding area, a decoration icon inserting area, a special effect decoration loading area, and corresponding fonts, decoration icons and special effect decorations on a video interface corresponding to each scene video on the initial template to generate a final first-class video template.
4. The method for automatically generating a short video according to claim 2, wherein the automatically generating a second type of video template by using a video file imported by a user in combination with video elements in a scene video database and a video element database comprises:
when the fact that a user imports a formed video is detected, carrying out video element and video scene splitting on the formed video imported by the user to obtain background audio, subtitle information, pictures and decorative elements in picture frames in the formed video;
sending the background audio, the subtitle information, the pictures and the decoration elements in the picture frames to a database to form a user video element sub-database;
respectively extracting scene videos and video elements in the scene video database and the video element database and scene videos and video elements in the user video element sub-database according to domain keywords input during user registration;
and randomly combining and fusing the video element scene videos and the video elements extracted from the scene video database and the video element database and the scene videos and the video elements extracted from the user video element sub-database to form a second type of video template.
5. The method for automatically generating a short video according to claim 1, wherein automatically performing adaptive template pattern adjustment on the generated video template according to the use condition of the short video file which is completely made by the user and the video template comprises:
monitoring the automatic generation process of the short video of the user and the manual adjustment process of the user aiming at the automatically generated short video in real time;
after a user finishes the current short video works and uploads or downloads the short video works, extracting a video template adopted by the finished short videos;
extracting the scene video and the video elements which are finally adopted in the video template and come from the scene video database, the video element database and the user video element sub-database;
marking the use times and the use time of the finally adopted scene video and video elements in the scene video database, the video element database and the user video element sub-database, wherein the use time is the finish time of the short video work; generating a use index numerical value of each scene video and each video element by using the use times and the use time marks of the scene videos and the video elements in the scene video database, the video element database and the user video element sub-database;
randomly extracting partial scene videos and video elements from the scene videos and the video elements in the scene video database, the video element database and the user video element sub-database which are finally adopted in the current finished video template according to an element extraction rule, and taking the extracted partial scene videos and video elements as elements to be replaced;
extracting the scene videos and the video elements which are different from the elements to be replaced and correspond to the number and the attributes of the elements to be replaced from the scene videos and the video elements of the scene video database, the video element database and the user video element sub-database, and replacing the elements to be replaced by the scene videos and the video elements which meet the preset use index numerical condition to generate a video template after self-adaptive adjustment;
wherein the element extraction rules are as follows:
the ratio ranges between the scene videos and the video elements extracted from the first type of video template and the second type of video template and the total amount of all the scene videos and the video elements adopted in the video template are respectively as follows: the scene video accounts for 20% -33%; the video element proportion range is 15% -25%;
aiming at the same video template, the proportion of extracted scene videos in the adaptive adjustment process of the template after every two adjacent video templates are used is different, and the difference range of the proportion of extracted scene videos every two adjacent video templates is 3% -5%;
aiming at the same video template, the proportion of extracted video elements in the adaptive adjustment process of the template after every two adjacent video templates are used is different, and the difference range of the proportion of the extracted video elements in every two adjacent video templates is 4-8%.
6. An automatic short video generation system, comprising:
the registration module is used for registering user information by a user and inputting keywords belonging to the field of video content for making short videos in the process of registering the user information;
the template generation module is used for judging whether a user imports a video file after system login in real time and generating a video template according to the field keywords of the video content imported by the user for making the short video and the video file imported by the user;
the video generation module is used for automatically calling a plurality of video templates matched with the making elements needing to make the short video input by the user according to the making elements needing to make the short video and the text information input by the user, and automatically generating the short video according to the video template selected by the user and the making elements;
and the self-adaptive adjusting module is used for automatically carrying out self-adaptive template pattern adjustment on the generated video template according to the use condition of the short video file which is manufactured by the user and the video template.
7. The system for automatically generating short video according to claim 6, wherein the template generating module comprises:
the first-class template generating module is used for automatically generating a first-class video template according to the field keywords of the video content imported by the user for making the short video and by combining the video elements in the scene video database and the video element database;
the second-class template generation module is used for judging whether a user imports a video file after system login in real time, and automatically generating a second-class video template by using the video file imported by the user and combining video elements in the scene video database and the video element database after the video file imported by the user is detected;
the scene video database comprises a plurality of types of scene videos, and the types of the scene videos correspond to the domain types of the domain keywords one to one; the video element database comprises a plurality of video elements, wherein the video elements comprise shapes and positions of a text adding area, a decoration icon inserting area and a special effect decoration loading area in a video, and correspondingly loaded character fonts, decoration icons and special effect decoration effects.
8. The system for automatically generating short videos according to claim 7, wherein the first template generation module comprises:
the first extraction module is used for extracting a plurality of scene videos corresponding to the domain keywords in the scene video database according to the domain keywords input by a user and using the scene videos as element scenes;
the initial template generation module is used for obtaining a network high-quality video through network search, splitting the network high-quality video and obtaining a video clip and a script corresponding to the network high-quality video; extracting video segments and scripts in the element scene, and replacing the video segments and the scripts with video segments and scripts corresponding to the network high-quality video to generate an initial template;
and the first-class video template generating module is used for adding a text adding area, a decoration icon inserting area, a special effect decoration loading area and corresponding fonts, decoration icons and special effect decorations on a video interface corresponding to each scene video on the initial template to generate a final first-class video template.
9. The system for automatically generating short videos according to claim 7, wherein the second type template generating module comprises:
the video splitting module is used for splitting video elements and video scenes of the formed video imported by the user when the user is detected to import the formed video, so as to obtain background audio, subtitle information, pictures and decorative elements in picture frames in the formed video;
the database forming module is used for sending the background audio, the subtitle information, the pictures and the decoration elements in the picture frames to a database to form a user video element sub-database;
the second extraction module is used for respectively extracting the scene videos and the video elements in the scene video database and the video element database and the scene videos and the video elements in the user video element sub-database according to the field keywords input during user registration;
and the second type video template generating module is used for randomly combining and fusing the video element scene videos and the video elements extracted from the scene video database and the video element database and the scene videos and the video elements extracted from the user video element sub-database to form a second type video template.
10. The system for automatically generating short video according to claim 6, wherein the adaptive adjusting module comprises:
the real-time monitoring module is used for monitoring the automatic generation process of the short video of the user and the manual adjustment process of the user aiming at the automatically generated short video in real time;
the video template extraction module is used for extracting a video template adopted by the manufactured short video after the user finishes the current short video work and uploads or downloads the short video work;
the video element extraction module is used for extracting the finally adopted scene videos and video elements in the scene video database, the video element database and the user video element sub-database in the video template;
the marking module is used for marking the use times and the use time of the finally adopted scene videos and video elements in the scene video database, the video element database and the user video element sub-database, wherein the use time is the finish time of the short video work; generating a use index numerical value of each scene video and each video element by using the use times and the use time marks of the scene videos and the video elements in the scene video database, the video element database and the user video element sub-database;
a replacement module, configured to randomly extract, according to an element extraction rule, part of scene videos and video elements from scene videos and video elements in the scene video database, the video element database, and the user video element sub-database that are finally adopted in the current completed video template, and use the extracted part of scene videos and video elements as elements to be replaced;
the automatic adjustment module is used for extracting the number and attributes of elements to be replaced from scene videos and video elements which are different from the elements to be replaced and in the scene video database, the video element database and the user video element sub-database, and replacing the elements to be replaced by the scene videos and the video elements which meet preset use index numerical conditions to generate a video template after self-adaptive adjustment;
wherein the element extraction rules are as follows:
the ratio ranges between the scene videos and the video elements extracted from the first type of video template and the second type of video template and the total amount of all the scene videos and the video elements adopted in the video template are respectively as follows: the scene video accounts for 20% -33%; the video element proportion range is 15% -25%;
aiming at the same video template, the proportion of extracted scene videos in the adaptive adjustment process of the template after every two adjacent video templates are used is different, and the difference range of the proportion of extracted scene videos every two adjacent video templates is 3% -5%;
aiming at the same video template, the proportion of extracted video elements in the adaptive adjustment process of the template after every two adjacent video templates are used is different, and the difference range of the proportion of the extracted video elements in every two adjacent video templates is 4-8%.
CN202210234535.XA 2022-03-11 2022-03-11 Automatic short video generation method and system Active CN114339407B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210234535.XA CN114339407B (en) 2022-03-11 2022-03-11 Automatic short video generation method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210234535.XA CN114339407B (en) 2022-03-11 2022-03-11 Automatic short video generation method and system

Publications (2)

Publication Number Publication Date
CN114339407A true CN114339407A (en) 2022-04-12
CN114339407B CN114339407B (en) 2022-06-03

Family

ID=81033069

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210234535.XA Active CN114339407B (en) 2022-03-11 2022-03-11 Automatic short video generation method and system

Country Status (1)

Country Link
CN (1) CN114339407B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017051793A1 (en) * 2015-09-25 2017-03-30 株式会社グロウ・ムービージャパン System for automatically generating video content
CN108924441A (en) * 2018-08-07 2018-11-30 上海奇邑文化传播有限公司 Video intelligent makes terminal, system and method
CN110177219A (en) * 2019-07-01 2019-08-27 百度在线网络技术（北京）有限公司 The template recommended method and device of video
CN110266971A (en) * 2019-05-31 2019-09-20 上海萌鱼网络科技有限公司 A kind of short video creating method and system
CN112732977A (en) * 2021-01-21 2021-04-30 网娱互动科技(北京)股份有限公司 Method for quickly generating short video based on template
CN112911399A (en) * 2021-01-18 2021-06-04 网娱互动科技(北京)股份有限公司 Method for quickly generating short video

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017051793A1 (en) * 2015-09-25 2017-03-30 株式会社グロウ・ムービージャパン System for automatically generating video content
CN108924441A (en) * 2018-08-07 2018-11-30 上海奇邑文化传播有限公司 Video intelligent makes terminal, system and method
CN110266971A (en) * 2019-05-31 2019-09-20 上海萌鱼网络科技有限公司 A kind of short video creating method and system
CN110177219A (en) * 2019-07-01 2019-08-27 百度在线网络技术（北京）有限公司 The template recommended method and device of video
CN112911399A (en) * 2021-01-18 2021-06-04 网娱互动科技(北京)股份有限公司 Method for quickly generating short video
CN112732977A (en) * 2021-01-21 2021-04-30 网娱互动科技(北京)股份有限公司 Method for quickly generating short video based on template

Also Published As

Publication number Publication date
CN114339407B (en) 2022-06-03

Similar Documents

Publication Publication Date Title
CN110557659B (en) Video recommendation method and device, server and storage medium
US8284231B2 (en) Video selector
Rasmussen Panel discussion II: Culture and media technology. The internet and differentiation in the political public sphere
CN110019286B (en) Expression recommendation method and device based on user social relationship
CN101692269B (en) Method and device for processing video programs
DE102008044635A1 (en) Apparatus and method for providing a television sequence
CN110415023A (en) A kind of elevator card recommended method, device, equipment and storage medium
CN111583415A (en) Information processing method and device and electronic equipment
CN114339407B (en) Automatic short video generation method and system
CN110030704A (en) Control method and device of air conditioner, storage medium and air conditioner
CN111460094A (en) Method and device for optimizing audio splicing based on TTS (text to speech)
Silvia et al. Strategi content creator pada dakwah di media sosial
CN114024737A (en) Method and device for determining brushing amount of live broadcast room
CN111368131B (en) User relationship identification method and device, electronic equipment and storage medium
CN117290605A (en) Vehicle-mounted intelligent scene recommendation method, device, equipment and medium
CN113694533B (en) Game development system based on big data for game stage representation state
WO2021051943A1 (en) Information pushing method and related apparatus
Li Strategies Behind Tik Tok’s Global Rise
CN114996509A (en) Method and device for training video feature extraction model and video recommendation
CN113965700A (en) Automatic adjusting method and system for intelligent television scene
CN102547458A (en) Novel digital media playing system and method based on user behavior
O'Regan Knowing The Processes But Not The Outcomes: Australian Cinema Faces The Millenium
Liu et al. A fusion approach to video quality assessment based on temporal decomposition
CN112035703A (en) Method and system for searching actors in short video
Susanthi et al. Conversational Structure Used in Guiding Conversation on YouTube Channel

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant