KR102632867B1 - Youtube content creation system - Google Patents

Abstract

The present invention relates to a YouTube content production system, and more specifically, to a YouTube content production system that supports more efficient production of YouTube content. For this purpose, the YouTube content production system includes a task instruction unit that instructs the producer terminal to produce a video according to the scenario transmitted from the writer terminal through the content production service, and YouTube streaming of the filmed video uploaded from the producer terminal through the content production service. A video management unit that converts the target content into target content, a thumbnail production unit that creates target thumbnails by editing the thumbnail keywords and thumbnail images input as the target content is shared on employee terminals through the content creation service through a preset design template, and After uploading the target content through the YouTube web page, it includes an integrated management unit that adds video information confirmed from the scenario, the target content, and the target thumbnail to the YouTube web page.

Description

YouTube Content Creation System{YOUTUBE CONTENT CREATION SYSTEM}

The present invention relates to a YouTube content production system, and more specifically, to a YouTube content production system that supports more efficient production of YouTube content.

Recently, UCC services such as YouTube have been provided that allow multiple users to enjoy video content through the Internet or mobile services.

Numerous videos are produced, published, and shared through these YouTube channels. As the number of subscribers to a YouTube channel increases, profits are generated for those who produce videos, so many creators are active on YouTube channels.

However, in order to earn satisfactory profits from the YouTube network, a large number of content must be uploaded to YouTube and at the same time, a distinctive thumbnail must be created to encourage other users to appreciate it, resulting in low profits compared to the invested effort. It can be loaded.

Accordingly, the present invention seeks to provide a YouTube content production system that can sequentially perform the tasks of writers, producers, and employees and support automation tasks so that YouTube content can be produced more efficiently.

The present invention is intended to solve the above problems, and the purpose of the present invention is to produce YouTube content that can support automation tasks while sequentially performing the tasks of writers, producers, and employees so that YouTube content can be produced more efficiently. This is to provide a system.

The above and other objects and advantages of the present invention will become apparent from the following description of preferred embodiments.

In order to achieve the above object, the YouTube content production system according to an embodiment of the present invention includes a task instruction unit that instructs the producer terminal to produce a video according to a scenario transmitted from the writer terminal through a content production service, and a task instruction unit that instructs the producer terminal to produce a video. A video management unit that converts the captured video uploaded through the content production service into target content for YouTube streaming, and a preset design of thumbnail keywords and thumbnail images input as the target content is shared to employee terminals through the content production service. The thumbnail production department creates the target thumbnail by editing it through a template, uploads the target content through the YouTube web page, and then adds the scenario, the target content, and video information confirmed from the target thumbnail to the YouTube web page. It includes an integrated management department.

In an embodiment, the preset template design means design information for one preset thumbnail among thumbnails with a certain number of views or more, and the design information includes the number of main objects, main object location, font position, font size, and Includes font color.

In an embodiment, the video information includes main information including video title, video description described below the video, target thumbnail, viewership, age limit, paid promotions, automatic chapters, tags, subtitles, recording date, license, and shorts. Contains sub-information including video, category, and comment evaluation.

In an embodiment, the integrated management unit includes a search unit that searches a plurality of search thumbnails using the video title identified from the video information, and a collection unit that collects each learning image by individually capturing the plurality of search thumbnails. , a labeling unit that labels each learning image with a popularity level according to the daily average number of views of the corresponding thumbnail, classifies each learning image by popularity level and learns it through machine learning, modeling an artificial intelligence-based popularity diagnosis algorithm. It includes a learning unit that determines whether the target thumbnail is eligible for replacement based on an output value derived by applying the target thumbnail to the popularity diagnosis algorithm.

In an embodiment, when the target thumbnail is determined to be a replacement target, the determination unit extracts a scene frame in which a preset event keyword is recognized by voice as the target content is played, and extracts the scene frame and the event keyword. A relay module that receives production thumbnail feedback as it is relayed to the employee terminal; based on the number of preferred comments that are fed back as the target thumbnail and the production thumbnail are uploaded through the content production service, which of the target thumbnail and the production thumbnail A decision module that determines one preferred thumbnail, if the preferred thumbnail is the produced thumbnail, a replacement module that replaces the target thumbnail uploaded through the YouTube web page with the produced thumbnail, and if the preferred thumbnail is the target thumbnail. , and a correction module that synthesizes the emoticon corresponding to the thumbnail keyword among the already collected emoticons for each keyword into the target thumbnail.

In an embodiment, when the target thumbnail is not determined to be a replacement target, the correction module corrects the thumbnail keyword and the thumbnail image detected from the target thumbnail through a preset correction filter.

In an embodiment, the video management unit includes a voice extraction unit that extracts voice data from the target content, a subtitle conversion unit that recognizes the voice output as the voice data is played and converts it into subtitle data, and the subtitles in the target content. As data is set and played, an image recognition unit recognizes the subtitle image recognized frame by frame through an OCR (Optical Character Reader) program, and based on whether there is a difference in playback time between the subtitle image and the subtitle data, the subtitle data and a sync adjustment unit that adjusts the frame sync to synchronize the subtitle image of the target content.

In an embodiment, the image recognition unit generates a shorts video by extracting frames before and after a certain time from the event action based on the event action of the object of interest recognized from the target content.

According to an embodiment of the present invention, an environment in which YouTube content can be produced more efficiently can be provided by sequentially performing the tasks of writers, producers, and employees while simultaneously supporting automated tasks.

Figure 1 is a diagram schematically showing a YouTube content production system 100 according to an embodiment of the present invention.
Figure 2 shows examples of target thumbnails.
FIG. 3 is a block diagram showing an embodiment of the integrated management unit 140 of FIG. 1.
FIG. 4 is a block diagram specifically showing the determination unit 145 of FIG. 3.
FIG. 5 is a block diagram specifically showing the video management unit 120 of FIG. 1.

Hereinafter, the present invention will be described in detail with reference to embodiments of the present invention and drawings. These examples are merely presented as examples to explain the present invention in more detail, and it will be apparent to those skilled in the art that the scope of the present invention is not limited by these examples. .

Additionally, unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as commonly understood by a person skilled in the art to which the present invention pertains, and in case of conflict, this specification including definitions The description will take precedence.

In order to clearly explain the proposed invention in the drawings, parts unrelated to the description have been omitted, and similar reference numerals have been assigned to similar parts throughout the specification. And, when it is said that a part "includes" a certain component, this means that it does not exclude other components, but may further include other components, unless specifically stated to the contrary. Additionally, “unit” as used in the specification refers to a unit or block that performs a specific function.

Collection codes (1st, 2nd, etc.) for each step are used for convenience of explanation. The collection codes do not explain the order of each step, and each step does not clearly state a specific order in context. It may be carried out differently from the order specified above. That is, each step may be performed in the same order as specified, may be performed substantially simultaneously, or may be performed in the opposite order.

Figure 1 is a diagram schematically showing a YouTube content production system 100 according to an embodiment of the present invention, and Figure 2 shows examples of target thumbnails.

When described with reference to FIGS. 1 and 2 , the YouTube content production system 100 may include a production instruction unit 110, a video conversion unit 200, a thumbnail management unit 130, and an integrated management unit 140.

First, the production instruction unit 110 may instruct the producer terminal 12 to produce a video according to a scenario transmitted from the writer terminal 11 through the content production service.

Here, the content production service is an app or program provided and installed on each of the writer terminal 11, producer terminal 12, and employee terminal 13 through a web server, and registers the scenario, target content, and target thumbnail from each terminal. It can be implemented as a graphical user interface for

At this time, the scenario refers to the script of the production video and serves as the basis for turning the story expressed in writing into a video story. In other words, it is called screen play or script.

Next, the video conversion unit 200 can receive the production video produced according to the scenario from the producer terminal 12 through the content production service and convert the production video into target content for YouTube streaming.

For example, the produced video is a codec file of one of MP4, M4V, 3GP, 3G2, ASF, AVI, FLV, MKV, and WEBM, and the target content is WMV, WMA, H.264/MPEG-, which can be streamed on the web. 4 It can be any one codec file among AVC, Divx, or MPEG-2-TS.

Next, the thumbnail management unit 130 shares the target content with the employee terminal 13 through the content creation service, and edits the thumbnail keyword and thumbnail image input from the employee terminal 13 through a preset template design to create the target content. You can create thumbnails.

Here, the preset template design may mean design information about one preset thumbnail among thumbnails that have a certain number of views or more.

At this time, the design information may include the number of main objects, main object location, font location, font size, and font color.

Next, the integrated management unit 140 can upload the target content through the YouTube webpage and then add the video information confirmed from the scenario, target content, and target thumbnail to the YouTube webpage.

Here, video information includes video title, video description below the video, main information including target thumbnail, viewership, age limit, paid promotion, automatic chapter, tag, subtitle, recording date, license, short video, category, Sub-information including comment evaluation may be included.

According to one embodiment, the integrated management unit 140 may copy a frame section of a certain period of time containing a thumbnail image that is confirmed as the target content is played and output, and cut-edit it to be played at the beginning of the target content.

According to another embodiment, the integrated management unit 140 may enlarge the facial area to a certain size through pixel fluidization correction for the facial area of the object of interest based on the preset facial expression state of the object of interest detected from the target thumbnail. .

According to another embodiment, when the integrated management unit 140 receives a scenario from the writer terminal 11 through a content production service, the writer terminal 11 detects the writer through a radar sensor (not shown). Based on breathing and heart rate signals, the writer's psychological state can be judged as either stable or unstable.

Here, if the writer's psychological state is determined to be stable, the integrated management unit 140 provides the target content to the writer terminal 11 through the content production service, receives the writer's editing section for the target content and sends it to the producer terminal ( 12) It can be relayed.

At this time, the author's editing section may include a cutting section where the target content is removed and a speed adjustment section where the playback speed is adjusted in the target content.

According to another embodiment, the integrated management unit 140 may search for sales sites that handle product objects based on product objects identified when playing and outputting target content and integrate them into the video information.

Hereinafter, the configuration of the present invention and its effects will be described in more detail through specific examples and comparative examples. However, these examples are for illustrating the present invention in more detail, and the scope of the present invention is not limited to these examples.

FIG. 3 is a block diagram showing an embodiment of the integrated management unit 140 of FIG. 1.

Referring to Figures 1 to 3, the integrated management unit 140 may include a search unit 141, a collection unit 142, a labeling unit 143, a learning unit 144, and a determination unit 145.

First, the search unit 141 can search a plurality of search thumbnails through YouTube 302 using the video title confirmed from the video information.

Next, the collection unit 142 may individually capture a plurality of search thumbnails and collect each learning image in the storage DB 301.

For example, when the video title is 'Destruction of Humanity,' the collection unit 142 sorts a plurality of search thumbnails searched through YouTube 302 in order of number of views, and captures the sorted plurality of search thumbnails in units of a certain number. You can collect a list of popular thumbnails by ringing them.

Next, the labeling unit 143 may label each learning image collected through the collection unit 142 with a popularity level according to the average number of daily video views of the corresponding thumbnail.

For example, if the number of video views confirmed for the thumbnail for one learning image is 200,000, and the video posting period is 20 days ago, the average number of video views per day for that thumbnail is 10,000, and any one learning image It can be labeled as level 1, which has a high popularity level. In addition, if the number of video views confirmed for the thumbnail of another learning image is 50,000, and the video posting period is 50 days ago, the average number of video views per day for the thumbnail is 1,000, and the other learning image has a popularity rating. This could be labeled as a low grade 5.

Next, the learning unit 144 can model an artificial intelligence-based popularity diagnosis algorithm by classifying each learning image by popularity level and learning it through machine learning.

Here, the artificial intelligence-based popularity diagnosis algorithm is an algorithm that receives target thumbnails and outputs output values corresponding to popularity ratings, including artificial neural network, SVM (Support Vector Machine), and decision tree. and Random Forest.

For example, artificial neural networks are mainly used in deep learning, are statistical learning algorithms inspired by machine learning and neural networks in biology, and can be convolutional neural networks that include feature extraction neural networks and classification neural networks. At this time, the convolutional neural network is a type of deep, feed-forward artificial neural network used to analyze visual images. It can be divided into the process of extracting image features and classifying classes, extracting and extracting features of a specific image, and Images can be recognized based on the features provided.

Next, the determination unit 145 determines whether the target thumbnail is eligible for replacement based on the popularity rating corresponding to the output value derived by applying the target thumbnail produced through the thumbnail management unit 130 to the popularity diagnosis algorithm. You can.

FIG. 4 is a block diagram specifically showing the determination unit 145 of FIG. 3.

Referring to FIGS. 1 to 4 , the determination unit 145 may include an extraction module 145_1, a relay module 145_2, a decision module 145_3, a replacement module 145_4, and a correction module 145_5.

First, when the target thumbnail is determined to be a replacement target through the determination unit 145, the extraction module 145_1 can extract a scene frame in which a preset event keyword is voice recognized as the target content is played.

Here, the preset event keywords may include keywords corresponding to aggro, such as shock, surprise, and goosebumps.

Next, the relay module 145_2 can receive production thumbnail feedback by relaying the scene frame and event keyword to the employee terminal 13.

Here, the production thumbnail may refer to a thumbnail image directly edited and produced by an employee holding the employee terminal 13.

Next, the decision module 145_3 uploads the target thumbnail and production thumbnail through the content production service and shares them on each terminal (e.g., 11 to 13) and the community web page 303 based on the number of preferred comments received as feedback. , you can decide either the target thumbnail or the production thumbnail as the preferred thumbnail.

Next, if the preferred thumbnail is determined to be the production thumbnail, the replacement module 145_4 can replace the target thumbnail uploaded through the YouTube web page with the production thumbnail.

Next, if the preferred thumbnail is the target thumbnail, the correction module 145_5 may composite the emoticon corresponding to the thumbnail keyword among the already collected emoticons for each keyword into the target thumbnail.

If the target thumbnail is not determined to be a replacement target, the correction module 145_5 according to the embodiment may correct the thumbnail keyword and thumbnail image detected from the target thumbnail through a preset correction filter.

Here, the preset correction filter may mean correction data that reflects the correction value of any one of color, brightness, saturation, and contrast so that the thumbnail keyword and thumbnail image are displayed more clearly than the target content.

FIG. 5 is a block diagram specifically showing the video management unit 120 of FIG. 1.

Referring to Figures 1 and 5, the video management unit 120 may include a voice extraction unit 121, a subtitle conversion unit 122, an image recognition unit 123, and a sync adjustment unit 124.

First, the voice extraction unit 121 can extract voice data from target content stored in the storage DB 301.

Next, the subtitle conversion unit 122 can recognize the voice output as voice data is played and convert it into caption data.

Next, the image recognition unit 123 sets subtitle data in the target content, and as it plays it, the subtitle image recognized through an OCR (Optical Character Reader) program can be recognized frame by frame.

Next, the sync adjustment unit 124 can synchronize the subtitle image of the target content by adjusting the frame sync for the subtitle data based on whether there is a difference in playback time between the subtitle image and the subtitle data.

Depending on the embodiment, the image recognition unit 123 generates a shorts video by extracting frames before and after a certain time from the event action based on the event action of the object of interest recognized from the target content, and sends it to the employee terminal 13. You can receive thumbnails for shorts videos as provided.

In this specification, only a few examples of various embodiments implemented by the present inventors are described, but the technical idea of the present invention is not limited or limited thereto, and of course, it can be modified and implemented in various ways by those skilled in the art.

11: Author terminal
12: Producer terminal
13: Employee terminal
100: YouTube content production system
110: Production instructions
120: Video conversion unit
130: Thumbnail management department
140: Integrated Management Department

Claims (8)

A task instruction unit that instructs the producer terminal to produce a video according to the scenario transmitted from the writer terminal through the content production service;
a video management unit that converts the captured video uploaded from the producer terminal through the content production service into target content for YouTube streaming;
A thumbnail production unit that produces target thumbnails by editing thumbnail keywords and thumbnail images input as the target content is shared to employee terminals through the content creation service through a preset design template; and
An integrated management unit that uploads the target content through the YouTube webpage and then adds video information confirmed from the scenario, the target content, and the target thumbnail to the YouTube webpage,
The above video information includes video title, video description below the video, main information including target thumbnail, audience base, age limit, paid promotions, automatic chapters, tags, subtitles, recording date, license, short video, category, and comment. Contains sub-information including evaluation,
The integrated management unit includes a search unit that searches a plurality of search thumbnails using the video title confirmed from the video information;
a collection unit that collects each learning image by individually capturing the plurality of search thumbnails;
a labeling unit that labels each learning image with a popularity level according to the average number of daily views of the corresponding thumbnail;
A learning unit that models an artificial intelligence-based popularity diagnosis algorithm by classifying each learning image by popularity level and learning it through machine learning; and
A YouTube content production system comprising a determination unit that determines whether the target thumbnail is eligible for replacement based on an output value derived by applying the target thumbnail to the popularity diagnosis algorithm. According to paragraph 1,
The preset template design refers to design information about one preset thumbnail among thumbnails that have a certain number of views or more,
The design information includes the number of main objects, main object location, font location, font size, and font color. delete delete According to paragraph 1,
The determination unit includes an extraction module that, when the target thumbnail is determined to be a replacement target, extracts a scene frame in which a preset event keyword is voice recognized as the target content is played;
A relay module that receives production thumbnail feedback by relaying the scene frame and the event keyword to the employee terminal;
a decision module that determines one of the target thumbnail and the production thumbnail as a preferred thumbnail based on the number of preference comments fed back as the target thumbnail and the production thumbnail are uploaded through the content creation service;
When the preferred thumbnail is the production thumbnail, a replacement module that replaces the target thumbnail uploaded through the YouTube web page with the production thumbnail; and
When the preferred thumbnail is the target thumbnail, a YouTube content production system comprising a correction module for synthesizing an emoticon corresponding to the thumbnail keyword among emoticons for each keyword already collected into the target thumbnail. According to clause 5,
The correction module is a YouTube content production system that corrects the thumbnail keyword and the thumbnail image detected from the target thumbnail through a preset correction filter when the target thumbnail is not determined to be a replacement target. According to paragraph 1,
The video management unit includes a voice extraction unit that extracts voice data from the target content;
a subtitle conversion unit that recognizes the audio output as the audio data is played and converts it into subtitle data;
An image recognition unit that sets the subtitle data to the target content and recognizes the subtitle image recognized frame by frame through an OCR (Optical Character Reader) program as it is played back; and
A YouTube content production system comprising a sync adjustment unit that synchronizes the subtitle image of the target content by adjusting frame synchronization for the subtitle data based on whether there is a difference in playback time between the subtitle image and the subtitle data. In clause 7,
The video recognition unit generates a short video by extracting frames before and after a certain time from the event action based on the event action of the object of interest recognized from the target content.