CN114786032B - Training video management method and system - Google Patents
Training video management method and system Download PDFInfo
- Publication number
- CN114786032B CN114786032B CN202210688101.7A CN202210688101A CN114786032B CN 114786032 B CN114786032 B CN 114786032B CN 202210688101 A CN202210688101 A CN 202210688101A CN 114786032 B CN114786032 B CN 114786032B
- Authority
- CN
- China
- Prior art keywords
- video
- instructor
- picture
- training
- lecturer
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/21—Server components or server architectures
- H04N21/218—Source of audio or video content, e.g. local disk arrays
- H04N21/2187—Live feed
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B5/00—Electrically-operated educational appliances
- G09B5/02—Electrically-operated educational appliances with visual presentation of the material to be studied, e.g. using film strip
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/431—Generation of visual interfaces for content selection or interaction; Content or additional data rendering
- H04N21/4312—Generation of visual interfaces for content selection or interaction; Content or additional data rendering involving specific graphical features, e.g. screen layout, special fonts or colors, blinking icons, highlights or animations
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/431—Generation of visual interfaces for content selection or interaction; Content or additional data rendering
- H04N21/4318—Generation of visual interfaces for content selection or interaction; Content or additional data rendering by altering the content in the rendering process, e.g. blanking, blurring or masking an image region
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/433—Content storage operation, e.g. storage operation in response to a pause request, caching operations
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/433—Content storage operation, e.g. storage operation in response to a pause request, caching operations
- H04N21/4334—Recording operations
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Databases & Information Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The application discloses a training video management method and a system, wherein the method comprises the following steps: obtaining instructor information in a training video to be played; acquiring a first video of the lecturer during historical live broadcasting according to the identifier of the lecturer; extracting a second video of the instructor from the first video; playing the training video, and judging whether a picture of the instructor exists in the training video; and starting a picture display control under the condition that the picture of the instructor does not exist in the training video, wherein the picture display control is a control superposed on a training video playing interface and is used for playing the second video. Through the method and the device, the problem that no instructor goes out of the mirror in the training video recorded in advance in the prior art is solved, so that the images of the instructor can be added when the training video is played, and the training effect is improved.
Description
Technical Field
The application relates to the field of video processing, in particular to a training video management method and system.
Background
Traditional offline educational training is tied to geographical locations and cannot cover more audiences. The online education training can well solve the problem, is an organization or online learning system for informationizing knowledge education resources, mainly comprises live broadcast training and network online training, and is relatively excellent in effect because teachers directly carry out online live broadcast. The online network training is generally a training video recorded in advance, and the training video recorded in advance is directly played during online network training. In the training video, if all recorded courseware forms are recorded, the trainees cannot see the lecturers in the course of listening to the lectures, and the trainees are easy to be distracted, so that the training effect is influenced.
Disclosure of Invention
The embodiment of the application provides a training video management method and system, which are used for at least solving the problem caused by the fact that no instructor goes out of a mirror in a training video recorded in advance in the prior art.
According to one aspect of the application, a training video management method is provided, including: obtaining instructor information in a training video to be played, wherein the instructor information comprises an identifier for uniquely identifying an instructor, and the training video is a pre-recorded video; acquiring a first video of the lecturer during historical live broadcasting according to the identification of the lecturer, wherein the first video is obtained by recording live broadcasting of the lecturer during live broadcasting; extracting a second video of the instructor from the first video, wherein the second video is obtained by performing background blurring on the parts of the first video except for the face and the body of the instructor; playing the training video, and judging whether a picture of the instructor exists in the training video; and starting a picture display control under the condition that the picture of the instructor does not exist in the training video, wherein the picture display control is a control superposed on a training video playing interface and is used for playing the second video.
Further, the step of judging whether the lecturer picture exists in the training video comprises the following steps: acquiring the duration of the training video, and determining a time interval for extracting key frames according to the duration, wherein the number of the key frames extracted according to the time interval is greater than a preset threshold value; extracting a plurality of key frames according to the time interval; sequentially judging whether the lecturer exists in the picture corresponding to each key frame in the plurality of key frames; and determining whether the pictures of the lecturer exist in the training video according to the judgment result.
Further, determining whether a picture of the instructor exists in the training video according to the judgment result comprises: determining that the picture of the instructor does not exist in the training video under the condition that the instructor does not exist in the picture corresponding to each key frame in the plurality of key frames, and determining that the picture of the instructor exists in the training video under the condition that the instructor exists in at least one key frame in the plurality of key frames.
Further, the determining whether the instructor exists in the picture corresponding to each key frame includes: extracting an image corresponding to the lecturer from a database according to the lecturer information; and comparing the image corresponding to the instructor with the picture corresponding to each key frame, and judging whether the picture corresponding to each key frame has the image corresponding to the instructor or not so as to determine whether the picture corresponding to the key frame has the instructor or not.
Further, comparing the image corresponding to the instructor with the picture corresponding to each key frame comprises: inputting the image corresponding to the instructor and the picture corresponding to each key frame into a machine learning model trained in advance for comparison, wherein the machine learning model is obtained by using multiple groups of training data for training, each group of training data in the multiple groups of training data comprises input data and output data, the input data comprises a first picture and a second picture, and the output data is label information used for identifying whether the second picture comprises a person in the first picture; and acquiring label information output by the machine learning model, wherein the label information is used for indicating whether an image corresponding to the instructor exists in a picture corresponding to each key frame.
According to another aspect of the present application, there is also provided a training video management system including: the system comprises a first acquisition module, a second acquisition module and a display module, wherein the first acquisition module is used for acquiring instructor information in a training video to be played, the instructor information comprises an identifier for uniquely identifying an instructor, and the training video is a pre-recorded video; a second obtaining module, configured to obtain, according to the identifier of the instructor, a first video during live broadcast historically by the instructor, where the first video is obtained by recording live broadcast during live broadcast by the instructor; an extraction module, configured to extract a second video of the instructor from the first video, where the second video is a video obtained by performing background blurring on the other parts of the first video except for the face and the body of the instructor; the judging module is used for playing the training video and judging whether a picture of the lecturer exists in the training video; and the starting module is used for starting a picture display control under the condition that the picture of the instructor does not exist in the training video, wherein the picture display control is a control superposed on a training video playing interface, and the picture display control is used for playing the second video.
Further, the determining module is configured to: acquiring the duration of the training video, and determining a time interval for extracting key frames according to the duration, wherein the number of the key frames extracted according to the time interval is greater than a preset threshold value; extracting a plurality of key frames according to the time interval; sequentially judging whether the lecturer exists in the picture corresponding to each key frame in the plurality of key frames; and determining whether the lecturer picture exists in the training video according to the judgment result.
Further, the determining module is configured to: determining that the picture of the instructor does not exist in the training video under the condition that the instructor does not exist in the picture corresponding to each key frame in the plurality of key frames, and determining that the picture of the instructor exists in the training video under the condition that the instructor exists in at least one key frame in the plurality of key frames.
Further, the determining module is configured to: extracting an image corresponding to the lecturer from a database according to the lecturer information; and comparing the image corresponding to the instructor with the picture corresponding to each key frame, and judging whether the picture corresponding to each key frame has the image corresponding to the instructor or not so as to determine whether the picture corresponding to the key frame has the instructor or not.
Further, the determining module is configured to: inputting the image corresponding to the instructor and the picture corresponding to each key frame into a machine learning model trained in advance for comparison, wherein the machine learning model is obtained by using multiple groups of training data for training, each group of training data in the multiple groups of training data comprises input data and output data, the input data comprises a first picture and a second picture, and the output data is label information used for identifying whether the second picture comprises a person in the first picture; and acquiring label information output by the machine learning model, wherein the label information is used for indicating whether an image corresponding to the instructor exists in a picture corresponding to each key frame.
In the embodiment of the application, the method comprises the steps of obtaining instructor information in a training video to be played, wherein the instructor information comprises an identifier for uniquely identifying an instructor, and the training video is a pre-recorded video; acquiring a first video of the lecturer during historical live broadcasting according to the identifier of the lecturer, wherein the first video is obtained by recording live broadcasting of the lecturer during live broadcasting; extracting a second video of the instructor from the first video, wherein the second video is obtained by performing background blurring on the parts of the first video except for the face and the body of the instructor; playing the training video, and judging whether a picture of the instructor exists in the training video; and starting a picture display control under the condition that the picture of the instructor does not exist in the training video, wherein the picture display control is a control superposed on a training video playing interface and is used for playing the second video. Through the method and the device, the problem that no instructor goes out of the mirror in the training video recorded in advance in the prior art is solved, so that the images of the instructor can be added when the training video is played, and the training effect is improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, are included to provide a further understanding of the application, and the description of the exemplary embodiments of the application are intended to be illustrative of the application and are not intended to limit the application. In the drawings:
fig. 1 is a flow chart of a training video management method according to an embodiment of the present application.
Detailed Description
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer-executable instructions and that, although a logical order is illustrated in the flowcharts, in some cases, the steps illustrated or described may be performed in an order different than presented herein.
In this embodiment, a training video management method is provided, and fig. 1 is a flowchart of a training video management method according to an embodiment of the present application, as shown in fig. 1, the method includes the following steps:
step S102, obtaining instructor information in a training video to be played, wherein the instructor information comprises an identifier for uniquely identifying an instructor, and the training video is a pre-recorded video;
step S104, acquiring a first video of the lecturer during historical live broadcasting according to the identifier of the lecturer, wherein the first video is obtained by recording live broadcasting of the lecturer during live broadcasting;
step S106, extracting a second video of the instructor from the first video, wherein the second video is obtained by carrying out background blurring on the other parts except the face and the body of the instructor in the first video;
as an optional implementation manner, if the first video of the lecturer during historical live broadcasting cannot be acquired according to the identifier of the lecturer, extracting audio information of the training video, and determining attribute information of the lecturer according to the extracted audio information, wherein the attribute information includes an age range and a gender of the lecturer; searching a preset lecturer according to the attribute information of the lecturer, wherein the attribute information of the preset lecturer is matched with the attribute information of the lecturer, and the preset lecturer has a video recorded during historical live broadcasting; using the recorded video of the predetermined lecturer as the first video.
Under the condition that a preset lecturer is found to fail according to the attribute information of the lecturer, generating an avatar according to the attribute information of the lecturer, wherein the avatar is matched with the attribute information of the lecturer; generating an animation of the avatar according to audio information extracted from the training video, wherein a mouth shape in the avatar in the animation corresponds to the audio information; and configuring the time axis of the animation according to the time axis of the training video, wherein the configured time axis of the animation corresponds to the time axis of the training video, and the vacant seat after the time axis is configured is used as the second video.
Step S108, playing the training video, and judging whether a picture of the instructor exists in the training video;
and step S110, starting a picture display control under the condition that the picture of the instructor does not exist in the training video, wherein the picture display control is a control superposed on a training video playing interface and is used for playing the second video.
In order to save the operation resources in this step, the following processing method may be adopted: playing the training video by using a first process, judging whether a picture of the lecturer exists in the training video by using a second process, and starting a picture display control by using a third process, wherein the second process is suspended when judging that the picture of the lecturer does not exist in the training video; and reserving resources for the first process and a third process in advance, wherein the reserved resources are used for running the first process and the third process, the reserved resources comprise a cache shared by the first process and the third process, and the third process stores the second video which is displayed on a picture and used for playing in the shared cache after being started. By the method, sufficient resources can be ensured to run the picture display control in the live broadcasting process.
In another optional embodiment, after the screen display control is started, the method may further include: recording a video played by a training video playing interface with the picture display control, obtaining a recorded third video with the second video after the training video is played, and storing the third video, wherein the storing of the third video comprises: acquiring identification information of the training video, wherein the identification information of the training video is used for uniquely identifying the training video; adding the identification information of the training video, the recording time of the third video and label information together to serve as the identification information of the third video, wherein the label information is used for indicating that a second video of the lecturer is displayed in the third video; and storing the identification information and the third video in a database, and adding the identification information of the third video serving as second identification information of the training video to establish a corresponding relationship between the third video and the training video.
After the corresponding relation between the training video and the third video is established, when the training video is played next time, whether the training video has the second identification information or not is searched to determine whether the training video has the corresponding third video or not, and under the condition that the corresponding third video exists, the third video is used for replacing the training video to be played.
Through the steps, the problem that no instructor goes out of the mirror in the training video recorded in advance in the prior art is solved, so that the images of the instructor can be added during the playing of the training video, and the training effect is improved.
In step S108, it may be determined whether a frame of the instructor exists in the training video by extracting a key frame.
In the optional mode, the duration of the training video can be obtained, and the time interval for extracting the key frames is determined according to the duration, wherein the number of the key frames extracted according to the time interval is greater than a preset threshold value; extracting a plurality of key frames according to the time interval; sequentially judging whether the lecturer exists in the picture corresponding to each key frame in the plurality of key frames; and determining whether the pictures of the lecturer exist in the training video according to the judgment result. For example, if the duration of a training video is 1 hour, the number of the preconfigured key frames is 10 key frames, and then one key frame is extracted every 6 minutes, so that the number of the extracted key frames is equal to or greater than 10, and the key frames are evenly distributed on the time axis.
In this case, when the lecturer is not present in the picture corresponding to each of the plurality of key frames, it is determined that the lecturer's picture is not present in the training video, and when the lecturer is present in at least one of the plurality of key frames, it is determined that the lecturer's picture is present in the training video. On one hand, the judgment is carried out by adopting the judgment mode, so that the pictures of a plurality of key frames need to be compared without comparing the whole video, and the resources are saved; on the other hand, the key frames are extracted from different time points, and the accuracy of judgment can be ensured.
Judging whether the lecturer exists in the picture corresponding to each key frame can be carried out in an image comparison mode. In this manner, the image corresponding to the instructor may be extracted from the database according to the instructor information; and comparing the image corresponding to the instructor with the picture corresponding to each key frame, and judging whether the picture corresponding to each key frame has the image corresponding to the instructor or not so as to determine whether the picture corresponding to the key frame has the instructor or not.
When the images are compared, a machine learning mode may be adopted for comparison, in which the images corresponding to the instructor and the pictures corresponding to each key frame are input into a machine learning model trained in advance for comparison, wherein the machine learning model is obtained by training using multiple sets of training data, each set of training data in the multiple sets of training data includes input data and output data, the input data includes a first picture and a second picture, and the output data is label information for identifying whether the second picture includes a person in the first picture; and acquiring label information output by the machine learning model, wherein the label information is used for indicating whether an image corresponding to the instructor exists in a picture corresponding to each key frame.
The number of the machine learning models can be two, wherein the first machine learning model is used for carrying out face recognition from the picture corresponding to the key frame and extracting a face image obtained by face recognition from the picture corresponding to the key frame; and the second machine learning model is used for comparing the person in the first picture with the face image extracted from the picture corresponding to the key frame so as to determine whether the person in the first picture is the same as the face image extracted from the picture corresponding to the key frame.
The second machine learning model is mainly used for comparing two face images, and the following alternative implementation is provided in the embodiment.
In the optional implementation mode, a single training sample face recognition method is mainly adopted, and the method comprises the following steps: 1) inputting a human face sub-feature training sample material: preparing a group of face photos, wherein the capacity is M = M [1] + M [2] +. + M [ N ], N is the number of people participating in the training sample and shooting the sample, and M [ i ] (1 ≦ i ≦ N, and M [ i ] ≧ 1) is the total number of photos of the ith person under given different shooting conditions; 2) constructing a training sample: m training materials are paired pairwise to generate training samples of M multiplied by M face photos; 3) extracting P sub-features of each training sample, and further obtaining P sub-feature measurement modules of each training sample through the difference value between the corresponding sub-features of the two photos in each training sample; 4) giving an arbitrary training sample, calculating a difference value of two images in the training sample according to the P sub-feature measurement modules, constructing a P-dimensional sample feature data vector v of the sample, wherein if two pictures in the training sample represent the same person, the response value of v is r =1, otherwise, r = 0; 5) obtaining a training result data set of machine learning by a machine learning method for the M multiplied by M training vectors and the corresponding response values in the step 4); 6) inputting two face photos to be identified and compared, calling P sub-feature measurement modules to calculate P distances under the topological distance space meaning to form a vector v to be tested, and predicting and judging a value r corresponding to v' according to a machine learning algorithm and a training result data set; when r =1, judging that the two photos correspond to the same person; let r =0, judge that two photos correspond to different persons.
Optionally, before the step 2), a step of normalizing the sample material scale is further included: unifying the pupil average coordinates of people on all the photos, unifying the distance between two pupils on each photo, and regulating the photos into the same size. The method also includes the step of graying the sample material after the sample material is standardized in size.
The method may further comprise the step of normalizing the brightness of the obtained grayed picture. Brightness normalization is to perform face detection, cut out face regions, and then normalize the face average brightness and contrast. For example, the standard for the average brightness of the face is 127, and the standard for the contrast normalization is a mean square error of brightness of 32.
The normalized size of the photo in the step 2) is 240 × 320 pixels, and the pupil distance is 64 pixels. In the above step, the number of the sub-features is not less than 6 and not more than 38.
There are many ways of machine learning. For example, the machine learning method in this alternative embodiment may be selected from an artificial neural network algorithm, a support vector machine algorithm, a bayesian classification algorithm, and a decision tree algorithm.
In the embodiment, the trained video information and the trained sound information are processed through the live broadcast client, the information is processed and displayed on a broadcast page of the terminal by using the broadcast terminal, meanwhile, the information data of the live broadcast client are collected by the signal collector, the live broadcast information is collected and matched with the main broadcast information of the database, and meanwhile, the live broadcast content is monitored through the live broadcast management server.
In this embodiment, an electronic device is provided, comprising a memory in which a computer program is stored and a processor configured to run the computer program to perform the method in the above embodiments.
The programs described above may be run on a processor or may also be stored in memory (or referred to as computer-readable media), which includes both non-transitory and non-transitory, removable and non-removable media, that implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
These computer programs may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks, and corresponding steps may be implemented by different modules.
Such an apparatus or system is provided in this embodiment. The system is called a training video management system and comprises: the first acquisition module is used for acquiring instructor information in a training video to be played, wherein the instructor information comprises an identifier for uniquely identifying an instructor, and the training video is a pre-recorded video; the second obtaining module is used for obtaining a first video of the lecturer during historical live broadcasting according to the identifier of the lecturer, wherein the first video is obtained by recording live broadcasting of the lecturer during live broadcasting; an extraction module, configured to extract a second video of the instructor from the first video, where the second video is obtained by performing background blurring on the other parts of the first video except for the face and the body of the instructor; the judging module is used for playing the training video and judging whether a picture of the lecturer exists in the training video; and the starting module is used for starting a picture display control under the condition that the picture of the instructor does not exist in the training video, wherein the picture display control is a control superposed on a training video playing interface, and the picture display control is used for playing the second video.
Preferably, the second obtaining module is further configured to, if the first video during live broadcast of the instructor in history cannot be obtained according to the identifier of the instructor, extract audio information of the training video, and determine attribute information of the instructor according to the extracted audio information, where the attribute information includes an age range and a gender of the instructor; searching a preset lecturer according to the attribute information of the lecturer, wherein the attribute information of the preset lecturer is matched with the attribute information of the lecturer, and the preset lecturer has a video recorded during historical live broadcasting; using the recorded video of the predetermined instructor as the first video.
The system or the apparatus is used for implementing the functions of the method in the foregoing embodiments, and each module in the system or the apparatus corresponds to each step in the method, which has been described in the method and is not described herein again.
For example, the determining module is configured to: acquiring the duration of the training video, and determining a time interval for extracting key frames according to the duration, wherein the number of the key frames extracted according to the time interval is greater than a preset threshold value; extracting a plurality of key frames according to the time interval; sequentially judging whether the lecturer exists in the picture corresponding to each key frame in the plurality of key frames; and determining whether the pictures of the lecturer exist in the training video according to the judgment result.
For another example, the determining module is configured to: determining that the picture of the instructor does not exist in the training video under the condition that the instructor does not exist in the picture corresponding to each key frame in the plurality of key frames, and determining that the picture of the instructor exists in the training video under the condition that the instructor exists in at least one key frame in the plurality of key frames.
For another example, the determining module is configured to: extracting an image corresponding to the lecturer from a database according to the lecturer information; and comparing the image corresponding to the instructor with the picture corresponding to each key frame, and judging whether the picture corresponding to each key frame has the image corresponding to the instructor or not so as to determine whether the picture corresponding to the key frame has the instructor or not.
For another example, the determining module is configured to: inputting the image corresponding to the instructor and the picture corresponding to each key frame into a machine learning model trained in advance for comparison, wherein the machine learning model is obtained by training by using multiple groups of training data, each group of training data in the multiple groups of training data comprises input data and output data, the input data comprises a first picture and a second picture, and the output data is label information for identifying whether the second picture comprises a person in the first picture; and acquiring label information output by the machine learning model, wherein the label information is used for indicating whether an image corresponding to the instructor exists in a picture corresponding to each key frame.
The system may further include: the storage module is configured to record a video played by a training video playing interface with the picture display control, obtain a recorded third video with the second video after the training video is played, and store the third video, where storing the third video includes: acquiring identification information of the training video, wherein the identification information of the training video is used for uniquely identifying the training video; adding the identification information of the training video, the recording time of the third video and the label information together to serve as the identification information of the third video, wherein the label information is used for indicating that a second video of the lecturer is displayed in the third video; and storing the identification information and the third video in a database, and adding the identification information of the third video to the training video as second identification information of the training video to establish a corresponding relationship between the third video and the training video.
Through the embodiment, the problem that no instructor goes out of the mirror in the training video recorded in advance in the prior art is solved, so that the images of the instructors can be added during the playing of the training video, and the training effect is improved.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A training video management method, comprising:
obtaining instructor information in a training video to be played, wherein the instructor information comprises an identifier for uniquely identifying an instructor, and the training video is a pre-recorded video;
acquiring a first video of the lecturer during historical live broadcasting according to the identifier of the lecturer, wherein the first video is obtained by recording live broadcasting of the lecturer during live broadcasting;
extracting a second video of the instructor from the first video, wherein the second video is obtained by performing background blurring on the parts of the first video except for the face and the body of the instructor;
playing the training video, and judging whether a picture of the lecturer exists in the training video;
and starting a picture display control under the condition that the picture of the instructor does not exist in the training video, wherein the picture display control is a control superposed on a training video playing interface and is used for playing the second video.
2. The method of claim 1, wherein determining whether a frame of the instructor exists in the training video comprises:
acquiring the duration of the training video, and determining a time interval for extracting key frames according to the duration, wherein the number of the key frames extracted according to the time interval is greater than a preset threshold value;
extracting a plurality of key frames according to the time interval;
sequentially judging whether the lecturer exists in the picture corresponding to each key frame in the plurality of key frames;
and determining whether the pictures of the lecturer exist in the training video according to the judgment result.
3. The method of claim 2, wherein determining whether a frame of the instructor exists in the training video according to the determination comprises:
determining that the picture of the instructor does not exist in the training video under the condition that the instructor does not exist in the picture corresponding to each key frame in the plurality of key frames, and determining that the picture of the instructor exists in the training video under the condition that the instructor exists in at least one key frame in the plurality of key frames.
4. The method of claim 3, wherein determining whether the instructor exists in the picture corresponding to each key frame comprises:
extracting an image corresponding to the lecturer from a database according to the lecturer information;
and comparing the image corresponding to the instructor with the picture corresponding to each key frame, and judging whether the picture corresponding to each key frame has the image corresponding to the instructor or not so as to determine whether the picture corresponding to the key frame has the instructor or not.
5. The method of claim 4, wherein comparing the image corresponding to the instructor with the picture corresponding to each key frame comprises:
inputting the image corresponding to the instructor and the picture corresponding to each key frame into a machine learning model trained in advance for comparison, wherein the machine learning model is obtained by using multiple groups of training data for training, each group of training data in the multiple groups of training data comprises input data and output data, the input data comprises a first picture and a second picture, and the output data is label information used for identifying whether the second picture comprises a person in the first picture;
and acquiring label information output by the machine learning model, wherein the label information is used for indicating whether an image corresponding to the instructor exists in a picture corresponding to each key frame.
6. A training video management system, comprising:
the system comprises a first acquisition module, a second acquisition module and a display module, wherein the first acquisition module is used for acquiring instructor information in a training video to be played, the instructor information comprises an identifier for uniquely identifying an instructor, and the training video is a pre-recorded video;
the second obtaining module is used for obtaining a first video of the lecturer during historical live broadcasting according to the identifier of the lecturer, wherein the first video is obtained by recording live broadcasting of the lecturer during live broadcasting;
an extraction module, configured to extract a second video of the instructor from the first video, where the second video is a video obtained by performing background blurring on the other parts of the first video except for the face and the body of the instructor;
the judging module is used for playing the training video and judging whether a picture of the lecturer exists in the training video;
and the starting module is used for starting a picture display control under the condition that the picture of the instructor does not exist in the training video, wherein the picture display control is a control superposed on a training video playing interface, and the picture display control is used for playing the second video.
7. The system of claim 6, wherein the determination module is configured to:
acquiring the duration of the training video, and determining a time interval for extracting key frames according to the duration, wherein the number of the key frames extracted according to the time interval is greater than a preset threshold value;
extracting a plurality of key frames according to the time interval;
sequentially judging whether the lecturer exists in the picture corresponding to each key frame in the plurality of key frames;
and determining whether the pictures of the lecturer exist in the training video according to the judgment result.
8. The system of claim 7, wherein the determination module is configured to:
determining that the picture of the instructor does not exist in the training video under the condition that the instructor does not exist in the picture corresponding to each key frame in the plurality of key frames, and determining that the picture of the instructor exists in the training video under the condition that the instructor exists in at least one key frame in the plurality of key frames.
9. The system of claim 8, wherein the determination module is configured to:
extracting an image corresponding to the lecturer from a database according to the lecturer information;
and comparing the image corresponding to the instructor with the picture corresponding to each key frame, and judging whether the picture corresponding to each key frame has the image corresponding to the instructor or not so as to determine whether the picture corresponding to the key frame has the instructor or not.
10. The system of claim 9, wherein the determination module is configured to:
inputting the image corresponding to the instructor and the picture corresponding to each key frame into a machine learning model trained in advance for comparison, wherein the machine learning model is obtained by training by using multiple groups of training data, each group of training data in the multiple groups of training data comprises input data and output data, the input data comprises a first picture and a second picture, and the output data is label information for identifying whether the second picture comprises a person in the first picture;
and acquiring label information output by the machine learning model, wherein the label information is used for indicating whether an image corresponding to the instructor exists in a picture corresponding to each key frame.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210688101.7A CN114786032B (en) | 2022-06-17 | 2022-06-17 | Training video management method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210688101.7A CN114786032B (en) | 2022-06-17 | 2022-06-17 | Training video management method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114786032A CN114786032A (en) | 2022-07-22 |
| CN114786032B true CN114786032B (en) | 2022-08-23 |
Family
ID=82420261
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210688101.7A Active CN114786032B (en) | 2022-06-17 | 2022-06-17 | Training video management method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114786032B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115866340B (en) * | 2023-01-18 | 2023-05-09 | 北京思想天下教育科技有限公司 | On-line training data processing method and system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002271769A (en) * | 2001-03-08 | 2002-09-20 | Toyo Commun Equip Co Ltd | Video distribution system for lecture presentation by the internet |
| JP2011234122A (en) * | 2010-04-27 | 2011-11-17 | Keio Gijuku | Lecture video creating system, lecture video creating method and program |
| WO2020045837A1 (en) * | 2018-08-28 | 2020-03-05 | 김영대 | Method for smart-remote lecturing using automatic scene-transition technology having artificial intelligence function in virtual and augmented reality lecture room |
| CN112312066A (en) * | 2020-11-06 | 2021-02-02 | 上海钦文信息科技有限公司 | Synchronous teaching technology based on multi-terminal real-time green screen synthesis and application |
| CN113191184A (en) * | 2021-03-02 | 2021-07-30 | 深兰科技（上海）有限公司 | Real-time video processing method and device, electronic equipment and storage medium |
| WO2021192931A1 (en) * | 2020-03-27 | 2021-09-30 | ソニーグループ株式会社 | Information processing device, information processing method, and information processing program |
| CN113778595A (en) * | 2021-08-25 | 2021-12-10 | 维沃移动通信有限公司 | Document generation method and device and electronic equipment |
-
2022
- 2022-06-17 CN CN202210688101.7A patent/CN114786032B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002271769A (en) * | 2001-03-08 | 2002-09-20 | Toyo Commun Equip Co Ltd | Video distribution system for lecture presentation by the internet |
| JP2011234122A (en) * | 2010-04-27 | 2011-11-17 | Keio Gijuku | Lecture video creating system, lecture video creating method and program |
| WO2020045837A1 (en) * | 2018-08-28 | 2020-03-05 | 김영대 | Method for smart-remote lecturing using automatic scene-transition technology having artificial intelligence function in virtual and augmented reality lecture room |
| WO2021192931A1 (en) * | 2020-03-27 | 2021-09-30 | ソニーグループ株式会社 | Information processing device, information processing method, and information processing program |
| CN112312066A (en) * | 2020-11-06 | 2021-02-02 | 上海钦文信息科技有限公司 | Synchronous teaching technology based on multi-terminal real-time green screen synthesis and application |
| CN113191184A (en) * | 2021-03-02 | 2021-07-30 | 深兰科技（上海）有限公司 | Real-time video processing method and device, electronic equipment and storage medium |
| CN113778595A (en) * | 2021-08-25 | 2021-12-10 | 维沃移动通信有限公司 | Document generation method and device and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114786032A (en) | 2022-07-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109698920B (en) | Follow teaching system based on internet teaching platform | |
| CN112818737B (en) | Video identification method, device, storage medium and terminal | |
| CN111523469B (en) | Pedestrian re-identification method, system, equipment and computer readable storage medium | |
| TWI764240B (en) | Intelligent video editing method and system | |
| CN116541538B (en) | Intelligent learning knowledge point mining method and system based on big data | |
| CN114786032B (en) | Training video management method and system | |
| CN113408554A (en) | Data processing method, model training method, device and equipment | |
| CN112215700A (en) | Credit face audit method and device | |
| CN112381099A (en) | Question recording system based on digital education resources | |
| CN110347866A (en) | Information processing method, device, storage medium and electronic equipment | |
| CN116561347A (en) | Question recommending method and system based on user learning portrait analysis | |
| CN115797349A (en) | Defect detection method, device and equipment | |
| CN115546692A (en) | Remote education data acquisition and analysis method, equipment and computer storage medium | |
| CN116109964A (en) | Intelligent extraction method and device for video map, storage medium and computer equipment | |
| CN113822907B (en) | Image processing method and device | |
| CN114492803A (en) | Knowledge graph-based question and answer generation method and device and automatic examination question generation system | |
| CN116437139B (en) | Classroom video recording method, device, storage medium and equipment | |
| CN114173191B (en) | Multi-language answering method and system based on artificial intelligence | |
| CN115511670A (en) | Online education method and online education platform | |
| CN115659221A (en) | Teaching quality assessment method and device and computer readable storage medium | |
| CN110730342B (en) | Video quality analysis method and device, server and terminal | |
| CN113923516A (en) | Video processing method, device and equipment based on deep learning model and storage medium | |
| CN112528790A (en) | Teaching management method and device based on behavior recognition and server | |
| CN114329050A (en) | Visual media data deduplication processing method, device, equipment and storage medium | |
| CN112712738A (en) | Student display processing method and device and electronic device |
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 |