27 results about "Video instruction" patented technology

The invention discloses a teaching corpus construction method and system based on deep learning, and a medium, and belongs to the cross technical field of medical education, artificial intelligence and multi-modal data processing. The method comprises the following steps: a data acquisition and annotation stage: acquiring medical inquiry video data of different scenes; in the model training stage, a multi-modal deep learning model architecture is used for carrying out cross-modal alignment on multi-modal data; then semantic understanding and question classification are carried out based on the pre-training language model after fine tuning; in the corpus construction and optimization stage, multi-modal data are integrated, a corpus management system is built, dynamic updating and self-adaptive optimization are carried out on a corpus, meanwhile, a typical non-language posture library is constructed for a regional culture background, and teaching is carried out in combination with the corpus. According to the invention, in combination with medical professional knowledge, an efficient and intelligent medical inquiry video teaching corpus is constructed through deep fusion of natural language processing, computer vision, speech recognition and multi-modal data fusion technologies.

The invention provides an abnormal event video instruction data generation method and device, and the method comprises the steps: classifying sample videos contained in a sample video set, and obtaining an abnormal video set and a normal video set; processing the abnormal video set based on a pre-trained multi-modal large model to obtain an abnormal event segment; performing frame extraction on the abnormal event segment, performing splicing based on a frame extraction result to obtain an image sequence, and processing the image sequence through the multi-modal large model to generate an abnormal description; based on a pre-trained large language model, constructing video level description corresponding to the sample video; and processing the exception description through the large language model to obtain an exception event instruction, and processing the video-level description through the large language model to obtain 16 types of question and answer type video-level instructions. A high-quality video anomaly detection instruction data set is constructed by taking a video frame as a detection granularity.

The invention discloses a motion video understanding method and device based on a multi-modal large model, and the method comprises the steps: (1) carrying out the uniform sampling of an inputted motion video, extracting K video frames, and extracting the visual features of each video frame through a visual encoder; (2) mapping the visual features to hidden space dimensions of a large language model through a multi-layer perceptron, and generating mapped visual features; (3) splicing the mapped visual features and the user query into a comprehensive prompt, and inputting a large language model to generate a text response; and (4) adopting a multi-stage training strategy to train the multi-modal large model, wherein the multi-stage training strategy comprises visual language alignment pre-training, video instruction following fine-tuning training and sports knowledge fine-tuning training. According to the method, the key visual information in the video can be effectively extracted, so that the ability of understanding a dynamic complex scene in the motion video is improved, and the time dependency relationship between the video frames is fully utilized, so that the accuracy and efficiency of understanding the motion video are remarkably improved.

The invention relates to the technical field of education, in particular to a health education interaction system and method based on virtual reality and deep learning. The health education interaction system creatively integrates motion sensing games, VR video teaching, personalized reading and a body measurement motion module, and all-around and immersive learning experience is created; the system motivates the participation degree of a user through a motion sensing game, interactive learning is achieved through VR video teaching, personalized reading meets the requirements of the user, and the health state is monitored through motion measurement. The intelligent analysis module is used for deeply mining data and providing personalized health assessment and suggestions; the self-learning unit continuously optimizes a knowledge base and a recommendation algorithm, and constructs a dynamic health knowledge graph; the system breaks through the limitation of traditional health education, improves the knowledge propagation efficiency and acceptance, effectively promotes the formation and maintenance of health behaviors, and brings revolutionary change to the field of health education.

The invention relates to the technical field of medical treatment and health information, in particular to a visual interaction-based self-injection success rate evaluation method for elderly diabetics, which comprises the following steps of: constructing an aging-suitable structured video teaching module, presenting an injection process in stages and embedding a standardized question node; a closed loop type Teach-back interaction feedback process is deployed; analyzing the multi-modal behavior data by using a computer vision and natural language processing technology, and extracting key operation parameters and cognitive integrity indexes; establishing a dynamic scoring model, and quantitatively evaluating the self-injection capacity in combination with the link weight and a standard threshold value; and personalized education intervention and post-discharge periodic follow-up visit management are implemented based on an evaluation result. Objective quantification and continuous education of the self-injection skill of the elderly patient are realized, the education effect is improved, and the operation related risk is reduced.

The invention relates to the technical field of video splitting, in particular to an English video splitting learning method based on a large model. According to the method, a teaching text and a voice transcription text of a teaching video are obtained through ASR and OCR technologies respectively, the text is segmented into single sentences, vocabularies are extracted and stored in a difficulty vocabulary library, and an integrated video meeting the requirements of an examination outline is generated through integration according to the expected learning level and difficulty quantitative indexes of a user. Segmenting the video length according to the learning duration demand of the user; a user playing record, a repetition rate, a complete playing rate and interaction feedback data are collected, a user learning level and video teaching difficulty are comprehensively evaluated through a big data model, and video content is dynamically adjusted. According to the scheme, the limitation of a traditional static splitting method is filled, the effects of effectively reducing knowledge omission and improving the learning efficiency are achieved through dynamic optimization and personalized customization of the teaching content, meanwhile, accurately matched teaching resources are provided for learners of different levels, and the learning efficiency is remarkably improved.

The application discloses an intelligent system capable of realizing individualized makeup teaching and application thereof, and the intelligent system comprises a video library provided with a plurality of classified makeup teaching video items with semantic classification labels, a classified makeup image library composed of a plurality of video frame images from the classified makeup teaching video items in the video library and with corresponding semantic classification labels, a face image acquisition module, a face feature analysis module, an adaptive makeup analysis module, a virtual makeup module and a video teaching module. The application can not only realize customization of individualized makeup, but also ensure that the customized individualized makeup has accurate corresponding teaching videos, and the customized individualized makeup can be effectively implemented, and the customized makeup effect is highly consistent with the learned makeup effect, so that the problem of difference between the actual makeup and the recommended makeup can be avoided, and people can easily and quickly obtain individualized makeup guidance suitable for themselves, and the application has a significant application prospect.

The invention discloses a holographic remote interactive teaching system, and relates to the technical field of teaching systems, and the system comprises an integrated teaching system which comprises a three-dimensional dynamic capturing module, a holographic imaging module, an intelligent interactive processing module and a low-delay communication module. The three-dimensional dynamic capture module is used for collecting three-dimensional holographic data of teachers and students in real time, and the three-dimensional holographic data comprises multi-angle depth images and skeleton joint point data. According to the invention, natural eye contact, body language capture and real-time feedback between teachers and students are realized, a dynamic holographic image with a real space depth is generated, the immersion and interactivity of teaching are effectively improved, the problems of eye contact deficiency and student attention distraction in traditional video teaching are solved, and remote teaching is closer to real classroom experience.

The present application relates to the technical field of video splitting, and particularly relates to a method for English video splitting learning based on a large model. The method acquires teaching text and voice transcription text of a teaching video through ASR and OCR technologies respectively, cuts the text into single sentences, extracts vocabulary and stores it in a difficult vocabulary library, integrates videos in accordance with examination syllabus requirements according to user expected learning level and difficulty quantification indexes, and cuts the video length according to user learning duration requirements; collects user playback records, repetition rates, complete play rates and interactive feedback data, comprehensively evaluates user learning level and video teaching difficulty through a big data model, and dynamically adjusts the video content. The scheme fills the limitations of traditional static splitting methods, effectively reduces knowledge omission and improves learning efficiency through dynamic optimization and personalized customization of teaching content, and provides precise matching teaching resources for learners of different levels, significantly improving learning efficiency.

The present application relates to the field of artificial intelligence technology and provides a motion video generation method, device, electronic device and storage medium. The method obtains a reference video and a base model for motion video generation, injects LoRA components into the spatial module and time module of the base model respectively, trains the LoRA component using an asynchronous training strategy, then receives an instruction to generate a target motion video, determines the target weight of the LoRA component, fuses the LoRA component with the base model using the target weight, and generates the target motion video based on the motion video instruction using the fused base model. This method generates a motion video using the motion information provided by the reference video without the need for trajectory guidance, and sequentially trains and fine-tunes the parameters of the spatial module LoRA component and the time module LoRA component through an asynchronous training strategy, thereby improving the controllability and robustness of the target motion video.

The invention relates to the technical field of artificial intelligence, and provides a motion video generation method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a reference video, acquiring a base model generated by a motion video, respectively injecting LoRA components into a space module and a time module of the base model, training the LoRA components by using an asynchronous training strategy, receiving an instruction for generating a target motion video, determining a target weight of the LoRA components, and determining the target weight of the LoRA components. The LoRA component and the base model are fused by using the target weight, and the target motion video is generated by using the fused base model based on the motion video instruction, so that the motion video is generated by using the motion information provided by the reference video without track guidance; and the parameters of the spatial module LoRA component and the time module LoRA component are sequentially trained and finely adjusted through an asynchronous training strategy, so that the controllability and robustness of the target motion video can be improved.

The application provides a multimodal embodied intelligent robot control method and device, wherein the method comprises: acquiring a current modal instruction, the current modal instruction comprising at least one of a video instruction, a text instruction, a picture instruction and an audio instruction; based on a pre-trained multimodal large model, acquiring a collapsed representation according to the current modal instruction; based on a pre-trained robot imitation learning network, predicting an output target action according to the collapsed representation and a current environment observation; controlling the multimodal embodied intelligent robot to operate according to the target action; wherein the robot imitation learning network is obtained by training and optimizing a training sample set composed of single modal data samples, corresponding environment observation data and real robot actions. The method acquires a collapsed representation by using a pre-trained multimodal large model, and predicts a target action by using the collapsed representation, which avoids the need for multimodal artificial labeled data when training the robot imitation learning network, and can achieve multimodal reasoning effect only by training single modal data, thereby reducing data collection and labeling costs, and improving the reasoning ability of the model.

The application relates to an intelligent online video teaching effect management system which comprises self-adaptive capturing devices, multiple optimization devices, target identifying devices, area storage mechanisms, visual judgment mechanisms and warning triggering mechanisms. The intelligent online video teaching effect management system is compact in structure and simple in operation.

The invention provides a teaching recommendation method, system and device based on multi-modal fusion and a storage medium, which can be applied to video teaching in the fields of finance and medical treatment, and the method comprises the following steps: obtaining teaching behavior data of a lecturer in a teaching video; synchronously acquiring learning interaction data and physiological signal data of each student during teaching; establishing time correlation mapping among the teaching behavior data, the learning interaction data and the physiological signal data, and detecting abnormal features in the teaching behavior data; judging whether the proportion of synchronously triggering learning interaction in the student group reaches a preset proportion or not when the abnormal feature appears, and generating an intervention instruction when the proportion reaches the preset proportion; performing real-time adjustment on teaching behaviors and teaching resource recommendation strategies of a lecturer based on the intervention instruction; according to the invention, teaching abnormal features can be detected in real time and group feedback can be quantified, so that teaching behaviors and resource recommendation can be optimized, and the teaching efficiency can be improved.

1. The name of the design product: microphone (TT60). 2. The use of the design product: wireless microphone for occasions such as live network, video shooting, program interview, interview recording, video teaching, network meeting, e-sports game, etc. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view.

1. The name of the design product: dance teaching machine. 2. The use of the design product: for dance video teaching. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view 1.

The invention discloses an interactive numerical control machine tool maintenance method and system, and the method comprises the steps: S1, building a three-dimensional model of a machine tool, and a preset threshold value and a maintenance video corresponding to each part, S2, collecting the real-time wear information of each part of the machine tool, and recording the real-time wear information as a dynamic parameter, and S3, comparing the dynamic parameter with the preset threshold value, and S4, if the prompt information is displayed to be qualified, recording operation historical information, if the prompt information is displayed to be unqualified, performing maintenance according to the three-dimensional model and the maintenance video, and repeating the steps S2 to S4 after maintenance. According to the interactive maintenance method for the numerical control machine tool, when maintenance is needed, the three-dimensional model or the maintenance video can be triggered according to the prompt information to guide maintenance, the maintenance is ensured to be smooth and in place through video teaching guided by three-dimensional simulation, and even if multiple persons rotate, each person can quickly master the state of the machine tool; and a maintainer is guided to carry out maintenance.

This application discloses an information processing method and apparatus based on interactive commands in an interactive product. Using the method of this application, when the first intent information corresponding to the first interactive video input by the current user does not match the node identifier of at least one candidate plot node, a first prompt message is output to receive a confirmation or denial command input for the first intent information, thus confirming whether the first intent information is correct. This helps improve the user's logical thinking and language expression abilities. When a denial command is received for the first intent information, a second prompt message including the trigger identifier of each of the at least one candidate plot node is output, thereby entering the interactive video teaching process to output the corresponding interactive teaching video to the user. This allows the user to learn and consolidate operational knowledge of interactive commands during the interaction process, which helps improve the user's interactive ability, learning ability, and proactive thinking ability.

A system and method for multi-media instruction authoring is disclosed. The multi-media instruction authoring system and method provide a unified and efficient system that supports the simultaneous creation of instructional media in AR, VR, and video-based media formats by subject matter experts. The multi-media instruction authoring system and method require only a single demonstration of the task by the subject matter expert and does not require any technical expertise of the subject matter expert. The multi-media instruction authoring system and method incorporate a 3D editing interface for free hand, in-headset interaction to create 2D videos from 3D recordings, by exploring novel virtual camera interactions.

1. Name of the product in this design: Microphone (TT40). 2. Application of this design: Microphone for use in online live streaming, video shooting, program interviews, interview recording, video teaching, online meetings, e-sports games, and other occasions. 3. The key design feature of this product is its shape. 4. The image or photograph that best illustrates the design's key points: a 3D model.

The present invention makes it possible to easily create a moving image guide that, in combination with a moving image of a worker's key operation, promotes recognition of the pressed key. This moving image guide creation device comprises: a machine operation panel registration unit that registers an image of the operation panel of a machine; a signal registration unit that registers each key on the operation panel in the image registered by the machine operation panel registration unit and a signal for controlling the operation panel by an operation of each key on the operation panel in conjunction with each other; a signal data change time extraction unit that extracts a time of change of the signal registered in the signal registration unit;a frame detection unit that detects, based on the time extracted by the signal data change time extraction unit, a frame in which a key corresponding to the changed signal was pressed by the worker from the moving image including an operation performed by the worker on the operation panel; and a moving image generation unit that generates a moving image indicating the key pressed by the worker on the image of the operation panel registered by the machine operation panel registration unit when the frame detected by the frame detection unit is displayed.

The invention provides a teaching method and device, electronic equipment and a storage medium. The method comprises the steps of collecting a target image of teaching materials; arranging a corresponding teaching audio and video according to the target image; in the teaching audio and video playing process, determining a first position of the current teaching content in the teaching material; and controlling the signal generator to emit the optical signal to the first position. Therefore, according to the scheme, teaching audio and video generation can be carried out on teaching materials in different formats, and the generality of teaching audio and video generation is improved. In the teaching audio and video playing process, the optical signal is transmitted to the first position of the teaching content in the teaching material, so that the teaching intuition is enhanced, and the learning effect is improved.

The invention relates to an intelligent VR teaching method, system and device for enhancing a video teaching effect. The method comprises the following steps: initializing a VR video playing scene structure and a VR model display scene structure; obtaining user behavior data and target teaching video stream data in the VR video playing scene structure, and loading a VR interaction model component group corresponding to the target teaching video stream data; generating a VR interaction model component control instruction based on the user behavior data and / or currently played target teaching video stream data; and generating and displaying a real-time VR interaction result model in the VR model display scene structure according to the VR interaction model component group in combination with the VR interaction model component control instruction. By adopting the method, one-way propagation of traditional video teaching can be converted into dynamic exploration, through dynamic adjustment and display of the VR interaction model, the learning efficiency and immersion of a learner can be remarkably improved, and a brand new normal form is provided for education digitization.

The vacuum cleaner includes: a vacuum motor configured to draw an airflow through an airflow path of the vacuum cleaner; a dirt separator having a dirt container; a display screen; and a controller configured to control an image displayed on the screen. The dirt container has a closed configuration in which the dirt container can receive dirt separated from the airflow, and an open configuration in which dirt contained therein can be emptied from the dirt container. The controller is configured to display video instructions on the screen.

The invention relates to an interactive teaching method and system, computer equipment and a storage medium, and the interactive teaching method comprises the steps: monitoring a level signal of a photoelectric sensor array on a tool wall in real time, generating a jump edge in response to the level signal of the photoelectric sensor array, and displaying the jump edge on the tool wall; determining a target tool or a target measuring tool according to the jump edge of the level signal; determining a target video corresponding to the target tool or the target measuring tool from video resources according to the target tool or the target measuring tool; and playing the target video and prompting the operation information of the target tool or the target measuring tool through the target video, thereby realizing video teaching and interactive teaching of taking and placing the tool from the tool wall by a user, and improving cognitive competence and learning effect of students.

The present invention relates to the technical field of fitness equipment, and specifically to a fitness dumbbell with video teaching, including a fitness dumbbell with video teaching, comprising a dumbbell rod, a dumbbell head and a counterweight block, wherein dumbbell heads are installed at both ends of the dumbbell rod, and the dumbbell heads are in a group of two, and a counterweight block is installed on the outer side of the dumbbell head, and a display component is installed on the dumbbell head, and one end of the display component is connected to the outer end of the dumbbell head. The display component rotates periodically through an internal display screen, so that the video rotates cyclically along the path of the display screen. The present invention deflects the gravity block inside the dumbbell head following the change of the center of gravity of the dumbbell, and then uses a rotating module to toggle the display screen so that the display angle of the display screen rotates periodically, thereby solving the problem that beginners cannot perform dumbbell fitness training with video follow-up anytime and anywhere.