65 results about "Movement recognition" patented technology

The invention relates to a simulation digital human real-time intelligent voice interaction system and a simulation digital human real-time intelligent voice interaction method based on vision and a large model, and aims to solve the problems of inaccurate target speaker recognition, high response delay and the like in digital human voice interaction in a complex scene. The system circles an effective recognition range through a camera, triggers audio collection in combination with face detection, locks a target speaker and reduces noise by using lip movement recognition and sound image fusion technologies, converts the target speaker into a text through voice wake-up, generates an answer by means of a large language model (LLM) and knowledge retrieval enhancement (RAG) technologies, generates low-delay voice through a voice synthesis technology accelerated by the vLLM, and performs voice recognition on the target speaker. And driving the preloaded digital human image to synthesize a video stream and pushing the video stream to a front end for rendering in real time. Accurate pickup, low-delay interaction and rapid digital human image switching in a complex environment are realized, the accuracy and real-time performance of intelligent voice question answering are improved, and the method is suitable for government affair halls, exhibition halls and other scenes.

The invention discloses a sign language recognition method and device based on multi-modal deep learning, and the method comprises the steps: multi-modal data input: capturing hand motions, gesture tracks and facial expressions at the same time through a camera, a motion capture sensor and other devices, and forming multi-modal data input; sign language action recognition: precise recognition of sign language actions is realized through a combined model of a deep convolutional neural network and a long-short-term memory network; facial expression and gesture track combined recognition: realizing understanding and translation of complex sign language sentences by combining the captured facial expressions and gesture tracks; and context natural language processing: generating a target statement in combination with context semantic understanding, and outputting a translation result. Through hand motion capture, facial expression analysis, gesture trajectory tracking and context natural language processing, complex sign language motions can be recognized more accurately and translated into characters or voices in real time, and low delay and high accuracy are achieved.

The invention discloses a dance movement recognition method, device and equipment based on dynamic time warping and a computer readable storage medium. The method comprises the steps of obtaining a user skeleton point sequence corresponding to a user dance movement and a standard skeleton point sequence corresponding to a standard dance movement; for each pair of frames in the user skeleton point sequence and the standard skeleton point sequence, multi-dimensional differences including joint point position differences, angle differences and motion amplitude differences are calculated; a double-layer dynamic weighting mechanism is applied to perform weighted fusion on the multi-dimensional difference values, and local cost between two frames is calculated; constructing a cost matrix based on local cost, and determining an optimal regular path and a dance motion similarity score by applying a path constraint dynamic time warping algorithm; and on the basis of the dance motion similarity score and the optimal regular path, generating an identification report including an overall score, a subitem score and visual correction information. The method has the advantages of being high in dance movement recognition precision, high in specialty and the like.

The invention belongs to the technical field of anti-unmanned aerial vehicles, and discloses an unmanned aerial vehicle capturing system based on radar search, which comprises a radar echo analysis unit, a phased array radar array is deployed to scan a target airspace in real time, radar echo signals are acquired, target detection sensitivity is adjusted in real time through a nonparametric dynamic threshold value method, and the target detection sensitivity is acquired. Target echoes and natural clutters are distinguished, frequency drift and waveform deformation of radar echo signals are analyzed, and a target dynamic trajectory sequence is output; the target motion recognition unit is used for receiving the target dynamic trajectory sequence, performing recognition analysis on a flight mode of the target based on a rule-driven trajectory fitting method, judging whether the target accords with preset unmanned aerial vehicle flight characteristics or not, and generating an unmanned aerial vehicle recognition flag bit if the target accords with the preset unmanned aerial vehicle flight characteristics; the sensing precision, the recognition reliability, the interception efficiency and the combat safety of the unmanned aerial vehicle target in a complex environment are improved.

The invention discloses a seat pre-adjustment control method based on embedded CNN waist-hip trajectory dynamic tracking, and the method can achieve the precise prediction of the motion intentions of people with different heights through the construction of a height proportion change curve, effectively eliminates the motion recognition error caused by the height and body type difference of different individuals, and improves the accuracy of the motion recognition. The method enhances the refinement of seat control, and reflects the change of the action stage of the target person by setting the track slope, thereby recognizing the transition of the target person from sitting down to stable sitting posture and from sitting posture to getting up in different action stages, predicting the next trend of the action in advance, effectively improving the prediction accuracy of the action intention, and improving the user experience. The seat system can make a response when the action starts, and the situation that the comfort is affected due to delayed adjustment is avoided.

Systems and methods described herein facilitate the tracking of events relating to property. For example, in certain embodiments, a system includes a property event detection device (e.g., smart home device, sensor device, and so forth) configured to detect activity (e.g., motion, light, sound, and so forth) proximate the property. The system also includes a property event tracking system configured to receive data (e.g., images, videos, facial recognition, movement recognition, object tracking, sound signatures, wireless data streams, and so forth) relating to the activity from the property event detection device, to determine whether the activity relates to a property event (e.g., home service, and so forth) for the property, and to store data relating to the property event in a distributed ledger (e.g., blockchain network) upon determination that the activity relates to the property event for the property.

The invention discloses an automatic classroom director method and electronic equipment. Audio data and video data of a current classroom scene are acquired; based on the audio data, sound source direction positioning is carried out through a microphone array, and the current sound source direction is determined; face key points are extracted according to the video data, lip movement recognition is carried out according to the face key points, and a lip movement recognition result is obtained; extracting human body key points according to the video data, and performing action recognition according to the human body key points to obtain a target action recognition result; based on the sound source direction, the lip movement recognition result and the target action recognition result, a current shooting picture of the camera device is controlled, and the current shooting picture of the camera device is a director picture; and outputting and displaying the director picture. According to the application, audio and video multi-mode information is fused, high-precision identification and natural and accurate automatic picture switching of the speaker and the interaction object in the classroom scene are realized, and the automation level of director and the overall classroom recording and broadcasting effect are improved.

The invention discloses a personnel state analysis method and system based on limb movement recognition, relates to the technical field of limb movement recognition, and determines the movement state of a corresponding limb part based on recognition of each movement-expression combination. And the driving response state of the personnel is determined based on the action state of each limb part, the corresponding part priority and the driving state of the truck, so that the accuracy of the driving response state of the personnel is improved. Therefore, the change event of the facial expressions is determined according to the plurality of facial expressions at different time, and the driving fatigue state of the personnel is determined according to the change event of the facial expressions and the lane changing frequency of the truck; according to the method, the voice interaction event of the person in the driving process is collected, the multiple state key contents are determined according to recognition of the voice interaction event, the state analysis system of the person is determined according to the multiple state key contents, the driving response state of the person and the driving fatigue state, and the accuracy of the state analysis system of the person is improved.

The invention relates to a voice interaction method and system based on lip movement and voice fusion recognition and a garbage truck, and the method comprises the following steps: collecting a high-definition lip movement image, and carrying out the preprocessing and lip region positioning of the lip movement image, and obtaining a lip image sequence; the method comprises the following steps: acquiring a voice signal, and carrying out noise reduction processing on the voice signal to obtain a voice signal after noise reduction; extracting lip movement features of the lip image sequence, and extracting voice features of the voice signal after noise reduction; respectively recognizing the lip movement features and the voice features to obtain a lip movement recognition result and a voice recognition result; the lip movement recognition result and the voice recognition result are fused, and a final operation instruction is generated; and executing corresponding operation based on the operation instruction. According to the method, the recognition accuracy in a high-noise environment can be remarkably improved.

This invention discloses a Tai Chi movement recognition method based on multi-rate compressive convolution. First, electromyographic (EMG) data of Tai Chi movements is collected and preprocessed. Then, temporal features of the source signals are extracted to construct a feature matrix. A multi-rate compressive convolution algorithm is used to extract cross-channel compressed feature sequences, which are then fused to obtain a multi-dimensional fused feature matrix. Temporal relationship features are then obtained through a long short-term memory (LSTM) network. Finally, a fully connected layer is used to obtain the movement recognition result, thus completing the Tai Chi movement recognition. This invention achieves multi-scale cross-channel feature extraction of EMG signals through multi-rate dilated convolution, overcoming the limitation of ordinary convolution which can only extract spatial relationships between adjacent channels. It can compress the information contained in EMG signal sequences of arbitrary sizes into a one-dimensional compressed feature sequence to extract highly fused features, reducing data redundancy, the number of parameters, and model computation, thereby improving the accuracy of Tai Chi movement recognition.

The invention provides a backpack multi-sensor data fusion-based motion attitude anomaly real-time detection method, which comprises the following steps of: acquiring multi-channel original data through a backpack-type integrated triaxial accelerometer, a gyroscope, a magnetometer and a barometer, implementing synchronous sampling and time-space alignment, and constructing a standardized time sequence data stream through wavelet denoising and multi-dimensional normalization preprocessing; high-dimensional features representing motion differences are extracted, nonlinear dimensionality reduction is achieved through principal component and Laplacian feature mapping, and a motion mode prototype library is generated; according to the method, a new motion situation can be continuously and adaptively learned, and the motion recognition accuracy, the system robustness and the attitude anomaly detection capability in a complex scene are effectively improved.

Aspects relate to a virtual reality management technique for facilitating precise long-range and short-range movement in a virtual environment while enabling complex multi-task actions. The virtual reality management technique includes determining, by analyzing the set of video data using a movement recognition model, a short-range movement action for an avatar in the virtual reality environment in a case that a set of movement factors detected for both a first arm and a second arm of the user satisfy a predetermined movement threshold and executing the short-range movement action, determining, by analyzing the set of video data using the movement recognition model, a long-range movement action for the avatar in the virtual reality environment in a case that a first hand of the user corresponds to a pointing gesture and a confirmation action is received from the user, and executing the long-range movement action.

This invention relates to the interdisciplinary field of image recognition and intelligent medical technology, specifically disclosing a method for orthopedic rehabilitation movement recognition and assessment based on image recognition. This method acquires patient motion images using a non-invasive multi-view visual sensor, extracts the three-dimensional coordinates of key skeletal points, and constructs a parametric biomechanical digital twin model integrating joint range of motion, muscle force lines, and ligament tension mechanisms. The key point data is input into a graph neural network for posture recognition, and the digital twin model is simultaneously invoked for physical feasibility verification and posture correction. Finally, the spatiotemporal trajectory similarity is calculated using a standard rehabilitation movement template, generating a comprehensive assessment result including movement completion, safety level, and injury risk. This invention, through the above technical solution, improves the robustness, physiological rationality, and clinical safety of movement recognition.

This invention relates to a scoring system and method for aerobics competitions, belonging to the field of sports competition scoring. The system includes a data acquisition module, a first data processing module, a second data processing module, a core calculation module, and an output module. The data acquisition module acquires scoring data input by the art, execution, and difficulty judging groups and the chief judge. The first data processing module performs arithmetic mean calculations on the art, execution, and difficulty scores respectively to generate average scores for each item. The second data processing module sums the points deducted by the chief judge and the points deducted based on the line of sight to generate the total points deducted by the chief judge. The core calculation module sums the average scores for each item with the total points deducted to generate the total score for the routine. The output module displays the total score. Through automated data calculation, combined with difficulty suggestion scoring provided by the movement recognition module, data compliance ensured by the verification module, and full traceability achieved through the storage module, the system improves the accuracy, efficiency, and objectivity of scoring, and enhances the transparency and credibility of the scoring.

The present application takes the teacher's hand gesture and upper limb movement as the research object, and discloses a teacher teaching intention detection method and system based on staggered attention for the auxiliary role of computer vision technology in the teaching scene. The method comprises the following steps: 1) respectively acquiring the teacher's hand gesture and upper limb movement video stream under the RGB-D camera and 3D structured light camera in the teaching environment; 2) respectively pre-processing the image data stream. The collected video stream is converted into teacher gesture, teacher upper limb movement key point heat map data stream and teacher gesture, teacher upper limb movement structured light image data stream; 3) the pre-processed data stream is respectively input into the teacher hand gesture recognition model and the upper limb movement recognition model. 4) after model prediction, the teacher hand gesture recognition result and the teacher upper limb movement recognition result are respectively output. The teacher gesture and the upper limb movement recognition result are combined to judge the teaching gesture. The present application uses a teacher teaching gesture estimation model based on a staggered attention mechanism, so that the network can recognize 3D teaching gestures according to the key point heat map when predicting the teacher gesture. At the same time, by combining the teacher upper limb movement recognition result with the gesture recognition result, the detection of the teacher's teaching intention is realized, and reference support is provided for the non-verbal ability evaluation of the teacher's teaching process.

This application provides a method, system, and device for monitoring intelligent knee braces. The method is applied to an intelligent knee brace, which includes a motion data acquisition device comprising multiple sensors. The method includes: acquiring motion data of a target user (a user wearing the intelligent knee brace) collected by each sensor; based on the motion data, identifying basic movement types using a basic classification algorithm model to obtain basic movement recognition results; if the basic movement recognition results are not composite movements, using the basic movement recognition results as the target user's movement state; if the basic movement recognition results are composite movements, further identifying subdivided movement types using a subdivided movement algorithm model to obtain the target user's movement state.

The invention relates to a physical education teaching and training all-in-one machine integrating AI motion recognition and voice interaction, belongs to the field of physical education teaching and intelligent motion equipment, and solves the problems that existing campus motion equipment is single in function and lacks intelligent guidance. The device comprises a machine body shell, a core function module, an AI motion recognition module, an AI voice large model dialogue module, a data statistics module, a display module and a storage module. The core function module covers free training, classroom teaching and other scenes; the AI motion recognition module collects images and corrects postures; the AI voice large model dialogue module realizes intelligent dialogue; the data statistics module analyzes the score; the display module displays the information; and the storage module is used for storing data. The all-in-one machine integrates multiple functions, supports multi-person interaction, improves the standardization and interestingness of sports, and meets the requirements of campus physical education.

A method for lower limb movement recognition based on surface electromyography (EMG) decoding using S-transform energy concentration is proposed. First, the raw EMG signal is preprocessed. The preprocessed signal is then truncated to a useful time segment using endpoint detection. Next, the signal within this time segment undergoes segmented S-transform and energy concentration calculation. Segmented operations are used to extract signal features of a specified dimension. Movement pattern classification is performed using SVM, and multi-channel signal features are fused and analyzed for lower limb movement recognition. This invention performs segmented S-transform on the signal to preserve its temporal information, while optimizing the energy concentration calculation process to improve movement classification accuracy. An SVM multi-classifier is built, and the accuracy of lower limb movement pattern recognition is further improved through S-transform energy concentration and feature fusion.

The application discloses a kind of based on deep learning's hip exoskeleton movement mode identification method and system.The method includes: data acquisition and processing: through the inertial measurement unit and joint motor of hip exoskeleton, heterogeneous sensor data are acquired;Offline / real-time calibration and time synchronization are carried out to data, and data set is constructed in combination with motion label;Model construction and training: splice multimodal sensor data as input, construct deep learning network, fuse heterogeneous features and identify terrain and movement mode, train model and save optimal model;Lightweight deployment: through ablation experiment optimization sensor configuration, adopt channel pruning and compress model parameter quantity, and generate light weight model in combination with INT8 quantization technology.The system includes three major modules of data acquisition, mode identification, quantization deployment.The application solves the problem of heterogeneous sensor data fusion and real-time deployment, significantly improves the precision and efficiency of exoskeleton movement recognition.

The application discloses a lip movement recognition method and device based on three-dimensional lip reconstruction, and the method comprises the following steps: acquiring a lip reading video to be recognized; generating a lip image to be recognized; inputting the lip image to be recognized into a target three-dimensional lip reconstruction model to obtain three-dimensional lip data to be recognized; and inputting the three-dimensional lip data to be recognized into a target lip movement recognition model to obtain a lip movement recognition result. Wherein, the method comprises the following steps: acquiring a lip language video; generating a lip image to be trained and text information; constructing an initial three-dimensional lip reconstruction model; constructing an initial lip movement recognition model; training the initial three-dimensional lip reconstruction model to obtain a target three-dimensional lip reconstruction model; inputting the lip image to be trained into the target three-dimensional lip reconstruction model to obtain three-dimensional lip data to be trained; and training the initial lip movement recognition model to obtain a target lip movement recognition model. The application realizes lip movement recognition, improves the accuracy and robustness, and can be widely applied to the technical field of visual speech recognition.

The present invention belongs to the field of computer application technologies and relates to a method for recognizing and evaluating limb rehabilitation movements based on deep learning. This method collects video data and skeleton data of limb functional rehabilitation movements, constructs training sets and validation sets for the video data and skeleton data after preprocessing; constructs a hybrid model of SE-ResBlock3D and LSTM, and the hybrid model realizes the recognition and classification of limb functional rehabilitation movements by sequentially passing the input video data through a convolutional layer, a sequence of 4 residual blocks, and an LSTM module; constructs a deep learning network SGP-Net, and the deep learning network SGP-Net sequentially passes the input preprocessed skeleton data through an STGCN module, a GAT module, and a PMTC module to extract and evaluate the spatial and temporal features of limb functional rehabilitation movements, so as to realize the quality evaluation of limb functional rehabilitation movements.

The present application relates to a device and method for controlling vehicle motion. In the vehicle motion control device and its control method, it can provide vehicle motion synchronized with passenger action in the vehicle based on passenger action, the recognizer is configured to obtain passenger motion information from the image in response to the case that the image including the passenger in the vehicle is input, and determine whether there is passenger dance action; the determiner is configured to obtain vehicle state information in response to the case that there is passenger dance action in the image, to determine whether vehicle motion control can be carried out, based on the determination that vehicle motion control can be carried out, the characteristics of passenger dance action are extracted; the controller is configured to generate target vehicle motion based on the characteristics, and control the vehicle motion by driving the actuator corresponding to the target vehicle motion.

The application provides a kind of eight-sections palm-chi movement evaluation method and system based on dynamic programming and deep learning, including collecting eight-sections palm-chi movement video data;Through the eight-sections palm-chi movement recognition model based on MediaPipe and DTW, the matching path, DTW distance and single frame score are obtained, the feature sequence of pivot action output by the eight-sections palm-chi movement recognition model is uniformly frame extracted and processed and then input into the eight-sections palm-chi movement evaluation model based on LSTM to obtain the classification label of user pivot action, and the error of user in key action is reflected according to the classification label. The skeleton points are extracted through the Mediapipe framework, the total score is obtained according to the DTW distance, the similarity between the overall movement of user and the standard action is evaluated according to the rhythm score, and the rhythm stability of the movement can be reflected in detail according to the single frame score;The problem of unstable length of video sequence is solved by using the uniform frame extraction method, and the model effectively learns and recognizes the action features.

The invention discloses a stroke patient hand rehabilitation action recognition method based on machine learning, and relates to the technical field of artificial intelligence medical treatment. The invention discloses implementation of a stroke patient hand rehabilitation action recognition model based on machine learning, and a construction method of the stroke patient hand rehabilitation action recognition model based on machine learning comprises the following steps: S1, data acquisition; s2, data preprocessing; s3, feature extraction; s4, performing correlation analysis; s5, constructing and training a model; and S6, evaluating the performance of the model. According to the method, the Leap Motion Controller is used for collecting hand kinematics data, common hand rehabilitation actions are accurately recognized and classified through a machine learning algorithm, interference of human factors is reduced, the objectivity and accuracy of evaluation are improved, and hand function quantitative data provide a basis for making a clinical rehabilitation scheme and determining prognosis.

1. Name of the designed product: Livestock mandibular movement detection device. 2. Use of the designed product: Livestock mandibular movement identification and analysis application. 3. Design points of the designed product: In shape. 4. Picture or photo best indicating the design points: Perspective view.

The invention relates to the technical field of medicine and rehabilitation, and discloses an automatic recognition-based whole body exercise quality evaluation and suggestion generation system, which comprises a data acquisition module, a data processing module, an exercise recognition module, an evaluation analysis module, a suggestion generation module and a user interaction module which are in communication connection in sequence. According to the method, data are collected through a camera, human body key point labeling and association are performed in combination with growth data, motion features are extracted by applying a growth association learning algorithm, a motion quality score is obtained based on a GMS standard evaluation method and growth data analysis, personalized rehabilitation suggestions are generated based on a big data expert knowledge base, and the personalized rehabilitation suggestions are displayed through a graphical interface. The defects that a traditional manual evaluation mode consumes manpower and time and is high in subjectivity are effectively overcome, the efficiency, accuracy and individuation degree of whole-body exercise quality evaluation and suggestion generation are improved, and the urgent requirement for exercise function evaluation in the field of medical treatment and rehabilitation is met.

The utility model relates to a clamping identification mechanism of an electroplating hanger, which comprises a frame body used for being connected with an external manipulator so as to move along with the external manipulator; the clamping assembly is connected with the frame body; the hanging tool assembly comprises a hanging plate, the hanging plate can bear a wafer, and the hanging plate is configured to be capable of being clamped by the clamping assembly so as to move along with the frame body; the recognition assembly is arranged on the clamping assembly and used for recognizing various types of hanging plates; wherein the top end of each hanging plate is provided with an identification part, the identification assembly comprises a scanning piece connected with the controller, the scanning piece is configured to be capable of scanning the identification part, and due to the fact that the scanning piece is arranged on the clamping assembly, when the hanging plate is selected, the scanning piece can firstly identify the identification part at the top end of the hanging plate; then the hanging plate is directly clamped through the clamping assembly, so that the wafer electroplating step is reduced, the time is saved, and the working efficiency is improved; and moreover, an independent hanger identification assembly does not need to be arranged, and the space of the whole electroplating hanger is saved due to the integrated arrangement.

The present application relates to a dance posture and movement recognition and interaction method and system based on visual computing, comprising the following steps: upon receiving a recognition start command, obtaining a student's dance movement information and dance music rhythm information through a camera device; extracting the student's posture profile information at several preset moments from the dance movement information based on the dance music rhythm information; comparing the extracted posture profile information with pre-stored standard posture profile information and outputting the comparison results; based on the comparison results, screening out posture profile information with a degree of overlap with the standard posture profile information below a preset threshold, extracting the standard posture profile information corresponding to the posture profile information, generating a movement correction data packet; and sending the movement correction data packet to an interactive terminal. The present application has the effect of accurately determining whether a student's dance posture movement is standard and providing timely feedback to the student.

The application discloses a kind of based on LSTM model identification human motion and electric stimulation's wearing training equipment, it is related to medical health technical field, including: intelligent insole sensor system is used to collect and upload the sensor time series data of foot when human activity;LSTM network model human motion recognition system is used to receive data and the intelligent insole sensor time series data is preprocessed, the data after processing is input to LSTM network model, and the motion mode of the motion of human motion and the behavior of kicking are identified;Electric stimulation system releases current pulse to foot bottom part according to the motion mode of the motion of human motion and the behavior of kicking identified.The application innovatively introduces LSTM neural network technology, gives the intelligent motion mode identification ability of electric stimulation equipment, can dynamically adjust electric stimulation parameter according to real-time motion state, can also predict kicking action in advance, through accurate electric stimulation auxiliary wearer optimizes kicking effect, improves training efficiency and motion performance.

The utility model discloses an eight-segment brocade motion identification auxiliary device based on multi-modal data fusion, which comprises a wearing bandage, an integrated module, a power supply module, a feedback module and an image analysis module, the wearing bandage is provided with a magic tape, a mounting cavity and a buffer layer are arranged in the wearing bandage, and the integrated module comprises a nine-axis attitude sensor, a data processing unit and a Bluetooth unit. The motion data can be collected, processed and sent to an external terminal, the image analysis module is composed of a camera and a processor and can capture motion images, position joint points and calculate the similarity with standard motion, the power supply module is a bendable lithium polymer battery, and the feedback module is provided with a vibration unit and a terminal graphical interface and achieves tactile and visual feedback. And the data processing unit and the image analysis module synchronize data through Bluetooth and fuse the data to generate a comprehensive evaluation result. In use, the device can assist a user in accurately identifying eight-segment brocade actions, provides comprehensive action evaluation through multi-modal data fusion, helps the user to correct the actions in time, and improves the exercise effect.