98 results about "Video learning" patented technology

The present invention discloses a method for solving open-type long video question and answer by using a hierarchical codec network mechanism, which mainly comprises the following steps: 1) training an adaptive hierarchical coding neural network for a group of video, question and answer training set, segmenting the long video based on a question and video learning adaptive segmentation mechanism,thereby obtaining a joint expression of the video segment and the question; 2) for the output of the coded neural network obtaining the joint expression of the video and the question, combining the idea of reinforcement learning together with a related answer to train a decoding neural network, directed at a natural language answer corresponding to the output of the joint expression of the video and the question. Compared with the general video question and answer solution, the method for solving open-type long video question and answer by using a hierarchical codec network mechanism in the present invention utilizes question-based adaptive layering, which can better lock the segments favorable for answering questions in long videos, can better reflect the characteristics of the video, andutilizes the reinforcement learning mechanism to train the decoder, so that a more powerful decoder can be obtained, and an answer better satisfying the requirement is produced. The effect achieved by the method in the present invention in the long video question and answer questions is better than the traditional method.

The invention discloses an interactive education method, which comprises the following steps that: audio and video display equipment receives two-dimensional code information from image scanning equipment; the audio and video display equipment obtains an address of an audio and video resource corresponding to the two-dimensional code information according to the two-dimensional code information; and the audio and video display equipment acquires the audio and video resource according to the address of the audio and video resource, and plays the audio and video resource. The invention also provides the audio and video display equipment and a corresponding system. According to the interactive education method, printed data learning and video learning are combined for interactive education, so that the method has the depth and systematicness of the printed data learning and the vividness and interestingness of the video learning, and the learning enthusiasm and efficiency of a user are improved; and moreover, the conventional function mode of the audio and video display equipment is changed, so that a good experience can be brought to the user.

The embodiment of the invention relates to the technical field of human-computer interaction, and discloses an interactive learning method and device in a video playing process, and terminal equipment. The method comprises the following steps of: detecting a first interactive switching instruction in a process of outputting a video in an operation interface of a video learning application, wherein the first interactive switching instruction is used for indicating switching of interface contents output by the operation interface; determining knowledge point information corresponding to the video and obtaining test question information corresponding to the knowledge point information according to the first interactive switching instruction, pausing video output in the operation interface and outputting the test question information for users to respond in the operation interface, detecting a second interactive switching instruction in a process of outputting the test question information in the operation interface, wherein the second interactive switching instruction is used for indicating switching of interface contents output by the operation interface, and continuously outputting the video in the operation interface according to the second interactive switching instruction. By means of the embodiment of the invention, the learning effect and the learning initiative of users can be improved.

The invention is applicable to the technical field of a computer and provides an examination question pushing method and device. The method comprises the steps of receiving a examination question generation request input by a user, wherein the examination question generation request comprises an identification number of a video learning file, and the identification number is used for identifying the video learning file; obtaining a referred subject of the video learning file and matching corresponding examination questions in a preset examination question library according to the referred subject; and selecting the examination questions of the corresponding number for the referred subject from the matched examination questions and outputting the selected examination questions to the user. The examination question pushing automatic and intelligent degree is improved, the availability of the output examination questions is effectively improved, and the learning efficiency of the user is improved.

The invention discloses a teaching video learning method as well as a system. The method comprises the following steps: when a teaching video is played, a background server acquires information markedby a user; according to the marked information, intercepting a corresponding video clip; and after playing of the teaching video is finished, locating the intercepted teaching video clip according tothe input information marked by the user, and performing playback on the intercepted teaching video clip. The method disclosed by the invention has the advantages that marking is performed on playedvideo information, a marked video is stored in the background server or a cloud server, retrieval is performed according to needs of the user, and then the playback is performed again, so that the problems that the whole video is viewed for multiple times and time is consumed for finding key information in the prior art are solved, and learning efficiency is further improved.

An individuation video learning system and method are applied to an electronic device or a network platform. First, individual teaching videos are made to set up a video database comprising a first storage unit used for storing individual videos of first learning and teaching and a second storage unit used for storing individual videos of non-first learning and teaching. The first storage unit and the second storage unit both comprise A level videos, B level videos and C level videos at least. For one teaching content, teaching models of individual teaching videos are different, sorts of teachers chosen are different, and teaching design made according to characteristics of personality of students is different. Next, a student judges whether the teaching content is learned by the student for the first time. At last, the student chooses individual teaching videos appropriate for the student according to the individual characteristics such as the learning condition, the learning level, the learning feature, the personality, and the interest of the student. The individuation video learning system and the method enable the students to choose the individual teaching videos according to the individual characteristics of the students to achieve the effect of individuation video learning.

The invention discloses a video playing method and device, terminal equipment and a computer readable storage medium, and relates to the field of video playing. The video playing method comprises thefollowing steps: receiving a video playing instruction; obtaining video information and text information corresponding to the video playing instruction; playing the video information in the first display area, and playing the text information in the second display area; and when a text page operation instruction is received, playing content corresponding to the text page operation instruction in the text information in a second display area. According to the invention, video information and text information are played on one terminal device at the same time; When a user only needs to check thecharacter information, the user can directly look up the character information through the operation of controlling the character information without influencing the playing of course contents in thevideo information, so that the video learning efficiency is greatly improved on the premise that the normal video learning is not influenced.

The invention discloses a management system of a real classroom and a manufacturing method thereof. According to the system, a projection device is used to project courseware image onto a pure color background. A control device is used to emit a control command to a courseware and control information is formed. Drawing track information of sketching, rubbing and writing on the courseware by a teacher in teaching is collected and compiled in real time through an interactive device. The class audio information of the teacher is collected and compiled in real time through a pickup device. Through a camera, pictures that the teacher teaches in coordination with the courseware projected on project cloth in front of the pure color background is compiled into class video information. Generated information is subjected to collection and image matting processing through a server device, combined with courseware information, the synchronous marking and storage in the time are carried out, and completed real classroom information is formed through synthesis. The system can display content more clearly and vividly, a user can feel the atmosphere of the on-site teaching, the effect of video learning is improved, thus the pictures are more hierarchical and more vivid, and more information can be displayed than actual class.

The invention discloses a teaching video display method, a device and a system, and the method comprises the steps: collecting the teaching video and voice of a teacher in a classroom, and forming teaching image data and audio data; recognizing the teacher teaching voice content, the student questioning voice content and the teacher answering voice content, and separating to form independent multi-track teaching audio data; obtaining teaching audio-visual data; generating a teaching display video, and uploading the teaching display video to a teaching platform; and selecting one teaching display video for playing to realize classroom reproduction by the process that students communicate with a platform through a student terminal. The invention provides a perfect teaching video learning method, combines an Internet teaching platform, gives play to the advantages of Internet teaching, integrates the advantages of offline and online learning, and facilitates the remarkable improvement ofstudent scores.

The invention discloses an automatic video learning effect testing method based on subtitles. The method comprises the steps of (1) playing a learning video; (2) setting parameters with a learning control unit; (3) selecting n subtitles from corresponding subtitle files in the video in the time period between t1 and t2 by means of a random function to serve as test answers, and selecting m subtitles from corresponding subtitle files in the video in other time periods to serve as confusion answers to form test questions; (4) conducting test through a subtitle test question generating unit, wherein video playing is suspended when the video is played to the t2 moment, and the test questions are displayed on the screen for a learner to answer; (5) processing the test result through a feedback control unit, continuing to conduct video learning of the next stage and repeating the steps (3) to (5) if the learner finishes the test questions, and processing the test result with a failure processing program by regarding the result as a failure if the learner does not finish the test questions, replaying the video sections from t1 moment to t2 moment under the default setting of the failure processing program and repeating the steps (3) to (5). According to the method, questions are automatically generated according to the subtitles appearing in the video for the learner to answer, and it shows that the learning effect meets the requirements if the answers are correct.

The invention discloses a personalized video learning system and method. The system comprises a judging module which allows a student to judge whether to firstly learn a course in a video according tohis or her previous learning experiences, a first level test module which provides comprehensive test questions according to previous courses to determine the current learning level of the student, asecond level test module which determines the current learning level according to the usual performance of the student, a selection module which comprises a teaching level selection module, a teaching mode selection module, a teacher selection module and a teaching design selection module, a storage module which comprises a first storage unit which stores videos within a first video level range and a second storage unit which stores videos within a second video level range, and a post-test module. The student can choose the most suitable teaching mode, teacher and teaching design according tohis or her interests, hobbies and learning habits and characteristics. Through personalized teaching, the learning efficiency is improved.

The embodiment of the invention discloses a learning video subtitle adding method and device, terminal equipment and a storage medium, and the method comprises the steps: extracting each target to-be-processed video segment, which comprises an audio signal, in a to-be-processed video, and each target to-be-processed video segment comprises first timestamp information; for each target to-be-processed video clip, identifying each audio signal in the current target to-be-processed clip as a target text, and combining each target text in combination with second timestamp information of each audiosignal to obtain each target video clip with subtitles; and displaying the target video clips with the subtitles according to the corresponding first timestamp information. The subtitle adding speed and accuracy are improved, the subtitle adding method is applied to the online learning video, the advantages of online video learning are brought into full play, students better follow the teaching progress of teachers by watching the subtitles corresponding to the learning video, and the understanding ability and learning efficiency of the students are improved.

The invention relates to a safe swimming trainer which comprises a waterproof mask. Transparent lenses are arranged on an eye portion of the waterproof mask, a head-wearing displayer is arranged on the outer side of each transparent lens, a hollow-type mouthpiece is arranged on a mouth portion of the waterproof mask, an air inlet and an air outlet are arranged on the mouthpiece and connected with an air incoming pipe and an air outgoing pipe respectively, the air outgoing pipe is connected to the upper portion of the side face of the waterproof mask, the other end of the air incoming pipe is connected with an air pressure ball, a hemispherical bottom shell is arranged in the pressure ball and provided with a storage battery and an electric air suction pump, and an air suction port of the electric air suction pump is positioned outside the pressure ball. The safe swimming trainer is simple to operate and convenient to use, and a breathing system is combined with the head-wearing displayer, so that a swimming learner can practice swimming underwater while performing video learning, training effect is improved, and signals can be sent out timely for help when dangerous circumstances occur. The safe swimming trainer has the advantages of good training effect, high safety and the like and is suitable for the swimming learner to practice swimming in water areas small in depth.

The invention discloses a video learning effect evaluation method based on artificial intelligence, and particularly relates to the field of digital learning, and the method comprises the steps: comprehensive testing before learning, wrong question arrangement, video adjustment, information recording playing, in-class testing, judgment model building, score judgment, and single-person learning result model building. According to the invention, process records of all users are collected and include data such as online visual evaluation watching time and playback times, after-learning test difficulty question answering accuracy and answering time, and the like, and the data are used as training data sources after a new teaching video is online, whether the teaching video is appropriate or not is predicted through the collected user data by taking the ability result of the user after the test as the classification basis, so that whether the video has the effect of improving the learning effect or not is evaluated.

A digital language teaching system is disclosed, in which through an audio control manner, the learner, by utilizing a learning sub-machine, can focus on listening and speaking practice under a simple hardware control and operation, so that the teaching staff and the learner can have a more direct and faster interaction for improving learning efficiency. Moreover, through the output devices of the video device and teaching providing the learner the teaching image and the interactive teaching pictures, the video learning manner can be provided to the learner through the supplemental multimedia teaching materials, so that the teaching staff can teach the learners in accordance with their aptitudes, and the learner can find the most suitable learning method.

The invention discloses a mother-and-baby all-life-cycle health medical service system based on artificial intelligence. The mother-and-baby all-life-cycle health medical service system comprises a data acquisition system, a data management system and a data exchange system. The data acquisition system is used for acquiring examination reports for the all life cycles of a large number of mothers and babies, diagnosis results which correspond with the examination reports and corresponding treatment plans. The data management system learns the data which are acquired by the data acquisition system through a neural network for obtaining a diagnosis model. The diagnosis module is used for obtaining an accurate diagnosis result and the corresponding treatment plan according to the input examination report. The data exchange system is used for inputting the examination report by the user for obtaining the diagnosis result and the corresponding treatment plan, and further is used for online transaction of mother-and-baby medical products. The service system of the invention can supply a health diagnosis service, an online mother-and-baby product purchasing service, a video learning platform and a community communication platform for the pregnant women.

The invention discloses a computer aided education system and method. The system comprises a classroom teaching subsystem, an after-class teaching subsystem and a central processing unit; the classroom teaching subsystem comprises a sign-up module, a roll call module, an interactive recording module and a video collection module; the after-class teaching subsystem comprises a free application module, an authorization code acquisition module, a video learning module, a payment learning module and a courseware sharing module. The system conducts recording on students who sign up for the class bychecking class attendance, the student who signs up for the class but does not attend the class can conduct classroom teaching video free learning application after class through the personal information from enrollment, the personal information of the student who initiates the classroom teaching video free learning application is matched with the personal information of the student who is absentfrom the class, when the matching is successful, and an authorization code is sent to a mobile phone of the student who initiates the classroom teaching video free learning application, so that a free classroom teaching video learning opportunity can be provided for the student who signs up for the class but dose not attend the class.

The invention provides a kindergarten morning inspection medical care integrated machine, which belongs to the field of health inspection in kindergarten. The integrated machine comprises an integrated machine main body, and the integrated machine main body integrates the tools and equipment required for the morning inspection and medical treatment to a portable and easy-to-use multifunctional platform. The integrated machine can be used as a child's morning inspection bed when a combination state is closed; the kindergarten morning inspection process can collect all aspects including the child's height, weight, identity, heart rate, pulse, body temperature and illness; the collected information can be uploaded to a cloud or a local server; the school and the teacher can query the morninginspection results and print the report through a display device or a smart terminal APP; Parents can use the smart terminal APP to query the data sent back by the morning inspection machine in real time and view the nursing advice for the patient; and the integrated machine integrates the common disease inspection and nursing knowledge video learning materials of the kindergarten. A height measuring rod adopts a rotating integrated design, and a weight scale is arranged in a sliding groove at the bottom of an accommodating groove in a sliding mode for convenient use and placement.

The invention provides a robot behavior teaching method based on meta-learning. The robot behavior teaching method is characterized by comprising the steps: acquiring a teaching video; and learning the teaching video by using the trained neural network model. The training process of the neural network model comprises the following steps: collecting training content; preprocessing the training content to obtain a preprocessed comparison video, a preprocessed teaching video and a preprocessed motion video; constructing an initial neural network model; taking the preprocessed teaching video as aninput, obtaining a demonstration action, and calculating the loss of the demonstration action; updating the initial neural network model according to the demonstration action loss to obtain an updated model; taking the preprocessed motion video and the track action as input to obtain a predicted track action, demonstrating semantics, motion semantics and comparison semantics, and calculating target action loss and semantic loss so as to construct total loss; updating the updated model based on the total loss; and until the total loss is stably converged to a total loss threshold, obtaining atrained neural network model.