227 results about "Action model" patented technology

The invention discloses a physical education assisting system and method based on human body posture recognition. The system comprises a depth shooting device, a standard action storage unit, a display terminal, an action comparison unit and a prompting device, wherein the depth shooting device is used for obtaining user action depth image sequences comprising movement action of a user following the demonstration action; the standard action storage unit is used for storing a standard action model and standard action demonstration image sequences corresponding to the standard action model; the display terminal is used for displaying the standard action demonstration image sequences for the user; the action comparison unit is used for comparing the user action depth image sequence at a scheduled time point or time period with the standard action model at the corresponding time point or the time period, and obtaining the difference between the user action and the standard action model; the prompting device is used for outputting action evaluation information or action correction prompting information to a display device according to the difference. According to the physical education assisting system and method, the user does not need to wear a special recognition device, the manufacturing cost of the system is lowered, and the user experience is improved.

Embodiments of a dialog system employing a discriminative action selection solution based on a trainable machine action model. The discriminative machine action selection solution includes a training stage that builds the discriminative model-based policy and a decoding stage that uses the discriminative model-based policy to predict the machine action that best matches the dialog state. Data from an existing dialog session is annotated with a dialog state and an action assigned to the dialog state. The labeled data is used to train the discriminative model-based policy. The discriminative model-based policy becomes the policy for the dialog system used to select the machine action for a given dialog state.

The invention discloses a travelling robot positioning system based on a single camera. In the system, a point model is adopted for representing an environmental model, an off-line map is calibrated, an action model and an observation model are established for carrying out position updating and road sign calibration, Monte Carto particle filter technology is adopted, particle swarm distribution with weights is adopted for representing robot pose estimation, the weights and the distribution of particles are updated through road sign matching processing, field particle distribution is rasterized, and a sub-region with maximal particle density is selected for positioning the robot. As an independent platform, the system greatly improves the map calibration technology on the environmental model, has great flexibility on the identification of ground line sections, and just needs re-calibration but not needs model re-establishment after environment change. The model reduces the matching complexity in visual processing and improves calculation efficiency. The Monte Carlo localization algorithm is expanded on the information integration processing technology, thus ensuring the real-time property and robustness of the system.

Described and shown herein is simple dynamic model for insulin time-action profiles. Various aspects are suitable for implementation in an embedded system such as an insulin pump, or a handheld device such as a pump remote control. Various aspects can be used as an off-line modeling tool running on a smartphone to assist patients in their treatment, or can be used as part of a closed-loop control algorithm for an artificial pancreas.

The invention provides a system for adjusting the amount of data bandwidth provided to a mobile device comprising a bandwidth adjustment facilitator arranged to receive a request from the mobile device for adjusting the amount of data bandwidth; a bandwidth throttler in communication with the bandwidth adjustment facilitator; the bandwidth throttler adapted to cap or allocate excess available data bandwidth to the mobile device; wherein on receipt of the request, the bandwidth adjustment facilitator process the request and if the request is successfully processed, adjusts the data bandwidth provided to the mobile device via the bandwidth throttler. The system may further be adapted for billing / charging based on either pay-per-specified-time model or pay per action model.The invention is conveniently suited for use in telecommunications system and does not require modifications to be made to existing telecommunications system.

The invention relates to a method for capturing human motion by the aid of fused depth images and three-dimensional models. By the aid of the method, problems of adhesion requirement on mark points, inconvenience in interaction and easiness in confusing and shielding the mark points of a method for optically capturing motion can be solved. The method includes acquiring depth information of human actions; removing backgrounds of motion objects; acquiring complete depth information of the human actions; converting the complete depth information of the human actions into three-dimensional point-cloud information of human bodies; acquiring three-dimensional human action models; establishing databases; enabling the databases to be in one-to-one correspondence with data of human action skeleton databases; extracting depth information of to-be-identified human actions to build the three-dimensional models; then matching the similarity of the three-dimensional models with the similarity of human actions in the databases of the three-dimensional models; outputting human action skeletons according to similarity sequences. The human action skeletons are motion capturing results. The method has the advantages that sensors or added mark points can be omitted on the human bodies, the method is easy to implement, motion sequences are matched with one another by the aid of a regular time wrapping process, the matching precision of each two sequences can be improved, the matching time can be greatly shortened, and the motion capturing speed and the motion capturing precision can be guaranteed.

The invention relates to a shooting action training method. The shooting action training method comprises the following steps: obtaining the shooting association data of a shooting action and a basketball moving state as well as shooting average by combining the basketball moving state information which is fed back by a basketball, the shooting actions of arms which are fed back by arm sleeves when the basketball is shot, and the shooting average which is fed back by a basket; comparing the shooting association data with a preset shooting action model which serves as a training standard, and finding and feeding back the difference between the shooting association data and the shooting action model to obtain training data, wherein the basketball moving state information comprises shooting trajectory, flying speed, rotating speed, and shooting angle; the shooting action comprises muscle power strength, power generating order, the postures of arms and fingers, and the action trajectory of the arms. The method is used for analyzing the coherent postures and muscle effectiveness of a basketball player in a shooting practice exercise, and giving posture correction and muscle power generation suggestions.

The invention relates to a hand action identification method and apparatus, a computer device and a readable storage medium. The hand action identification method comprises the following steps of collecting an image of a to-be-identified object through an image collection apparatus, and inputting the image of the to-be-identified object to an initial human body model to obtain pose information ofthe to-be-identified object; selecting position information of wrist joints from the pose information, and according to the position information of the wrist joints, obtaining a hand region image; andinputting the hand region image to a preset hand action model to obtain a first hand action of the to-be-identified object. According to the method, an overall pose of an identified person is identified firstly and then a hand is located; and when multiple to-be-identified objects exist in an identification region and are mutually shielded, the hand positions can be judged according to the posesof the to-be-identified objects, so that the hand action identification accuracy is improved.

Systems and methods are provided for a computer-implemented method of providing an interactive avatar that reacts to a communication from a communicating party. Data from an avatar characteristic table is provided to an avatar action model, where the avatar characteristic table is a data structure stored on a computer-readable medium that includes values for a plurality of avatar personality characteristics. A communication with the avatar is received from the communicating party. A next state for the avatar is determined using the avatar action model, where the avatar action model determines the next state based on the data from the avatar characteristic table, a current state for the avatar, and the communication. The next state for the avatar is implemented, and the avatar characteristic table is updated based on the communication from the communicating party, where a subsequent state for the avatar is determined based on the updated avatar characteristic table.

The invention provides an accelerometer data modeling and motion recognizing method. The method comprises the steps of acquiring action data by use of an accelerometer and building a plurality of action models, including actions such as moving, sleeping, palming / palm dropping and knocking, acquiring the action data in real time by use of the accelerometer and performing smoothing processing on the acquired data by use of a smoothing filter so as to remove noise, analyzing the filtered data and recognizing actions by comparing with target action data models. After action recognition is completed, subsequent data, such as motion related information like calorie consumption, distance, motion pattern and sleeping quality, are calculated according to different action recognition requirements. The accelerometer data modeling and motion recognizing method is capable of realizing real-time analysis on a plurality of actions according to the difference between various actions while guaranteeing high accuracy of action recognition; as a result, a high-accuracy and high-intelligence human body action recognition method can be provided for a user intelligently.

The invention provides an iron tower bolt defect display method and system based on a three-dimensional live-action model. The method comprises: achieving automatic flight of an unmanned aerial vehicle through air route planning, collecting iron tower bolt installation image information, shooting position coordinates (planes and elevations) and shooting angle information, and carrying out POS calculation, aerial triangulation encryption, three-dimensional modeling and the like to construct a live-action three-dimensional model of an iron tower; placing the three-dimensional model of the iron tower on a three-dimensional earth, and analyzing and recognizing the image by using a deep learning intelligent image recognition algorithm; locating the location of a defect, and marking a red colorblock in the three-dimensional scene; and according to the marked defect position on the three-dimensional model, calculating the position coordinates of the bolt defect by using the principle of forward intersection in combination with the position and attitude information of all photos, and carrying out manual checking. According to the invention, with the method and the system, detection of thebolt integrity rate of the power transmission line iron tower is guaranteed, on-site construction acceptance is assisted, the construction quality of the power transmission line is improved, and safety risks caused by human factors are reduced.

The invention discloses a flood model building and dynamic displaying method based on BIM and a three-dimensional live-action model. A three-dimensional design platform, a geographic information platform, a hydrodynamic modeling platform and a three-dimensional live-action modeling platform are subjected to cross-platform technological fusion, the three-dimensional live-action model is applied tobuild a high-precision flood model, and three-dimensional dynamic displaying is performed. The method includes: 1, producing the three-dimensional live-action model, a DEM and an orthoimage; 2, building a flood analysis model and calculating; 3, exporting and converting a flood calculation result; 4, integrating the live-action model, a BIM design model and the flood calculation result; 5, three-dimensionally and dynamically displaying a flood result. Cross-platform technological fusion is performed on the three-dimensional design platform, the geographic information platform, the hydrodynamicmodeling platform and a three-dimensional visual rendering platform, the flood model is built on the basis of live-action model data generated by inclined photographing of an unmanned aerial vehicleand overlaid with the live-action model for three-dimensional dynamic display of flood, so that accuracy of the flood model is improved substantially.

Embodiments of techniques or systems for usage modeling or gesture based authentication are provided herein. Actions of a user may be captured, classified, and utilized to generate an authentication scheme for a device based on gesture recognition, gesture analysis, or action analysis. Training data may be determined based on actions deemed to be training actions and one or more action models may be generated based on the training data. An action model may be indicative of how a user performs a particular or predetermined action. When a security mode is enabled, usage actions may be recorded and usage data may be extracted or determined based on the usage actions. A usage action may be identified and corresponding usage data may be compared with training data from an appropriate action model. An identity of a user associated with the usage action may be determined based on this comparison.

The invention discloses a heterogeneous network of wireless service and a switching method thereof. The inside of each network is provided with a node i-Node; the inside of the i-Node comprises a strategy database for storing various communication strategy data, a knowledge database for storing a vertical switching action model counted and formed according to vertical switching action and occupancy rate of a current network channel, a cognition level interface for collecting parameters of each protocol layer, and a cognition platform, wherein the cognition platform is used for judging whether the vacancy rate of the current channel is more than a threshold value after a switching signaling of an outer network is received, if so, a connected wireless node accessed to the network is triggered; the wireless node utilizes the parameters of the cognition layer interface and the vertical switching model to implement the switching process, and otherwise, the cognition platform replies a signaling of refusing switching. The invention also provides the method for switching the heterogeneous network of the wireless service. Through the strategy and the model pre-stored in the knowledge database and the strategy database, the heterogeneous network and the switching method can rapidly improve the service switching process among the heterogeneous networks and reduce the time delay of the switching.

The invention provides an intelligent trouble diagnosis method for a tractive power supply system, which comprises the following steps: firstly establishing a system description library for a system to be diagnosed; then collecting observed value data required by the diagnosis in real time, carrying out pretreatment, acquiring system predicted values in accordance with a system element action description library and the observed value data, and performing a fuzzy consistency check on the predicted values and the observed values to judge the difference, if not, performing a layering consistency diagnosis on the system to generate a fault candidate collection; selecting a fault action model to perform an abductive diagnosis to determine fault reasons and fault positions, thereby predicating relay protection action and breaker action; and finally comparing the consistency of a predicated action result and real action to obtain the fault reasons and the results, and carrying out alarm output. The invention also provides an intelligent trouble diagnosis system for a tractive power supply system. The invention is suitable for the tractive power supply system, can immediately and exactly find out the fault elements and the fault reasons, overcomes computational complexity and realizes real-time monitoring and fault diagnosis for the tractive power supply system.

The invention discloses an intelligent wrist watch capable of automatically rotating interface along with gesture and a control method thereof. The method comprises the following steps: A, obtaining the movement data of an intelligent wrist watch through a sensor arranged in the intelligent wrist watch, and pre-processing the movement data; B, performing data mining operation on the pre-processed movement data and establishing an action model, identifying and outputting the rotating angle of the intelligent wrist watch; C, correspondingly rotating a graphical interface according to the output rotating angle by an upper layer. The sensor arranged in the intelligent wrist watch is adopted for obtaining the movement data change information of the intelligent wrist watch; a bottom layer is adopted for driving and algorithm processing for outputting the rotating angle, and the upper layer APP is used for rotating the graphical interface according to the selected angle.

The invention discloses a wireless multimedia sensor network design method driven by a subjective interest model, is mainly used for the design and application of a wireless multimedia sensor network and pertains to the cross filed of distributed computing and network communication techniques. The method of the invention designs a network intelligent model according to an action model abstracted from the action characteristics of biocenosis in the nature, can be applied to the design of the multimedia sensor network triggered by multiple events and is characterized by reasonably utilizing and distributing recourses in distributed computing and obtaining effective data of event information sources at a relatively low cost and relatively low time delay, carrying out activity cooperation with other crunodes through an interest degree measurement method and being conducive to centralizing network resources to do an in-depth reporting on an interest event source, effectively carrying out data fusion and abandoning redundant data, carrying out natural evolution towards the network, forming a friend circle facing the event and optimizing the structure of the network.

The invention provides a TBM tunneling parameter prediction method based on multi-target learning, which can solve the problems of subjectivity of artificial decision and singleness of decision rulesin a traditional control strategy. The method comprises: acquiring TBM tunneling parameters through an intelligent data acquisition system; performing noise reduction and enhancement processing on theTBM tunneling parameters to obtain data of each cyclic in a rising section anda stable section; extracting time domain characteristics of the rising section and the stable section in each cycle, andconstructing an evaluation index reflecting the health state of the TBM; utilizing an artificial intelligence algorithm to effectively represent a nonlinear mapping relation between the rising sectionTBM characteristic index and the stable section multi-target variable, and establishing a multi-target learning-based rock machine action model; carrying out self-adaptive adjustment on parameters ofthe rock machine action model through a nonlinear optimization method; according to the multi-target parameter information estimated by the rock machine action model, enabling the TBM driver to optimize the TBM control strategy by integrating the geological conditions of the surrounding rock and the particle distribution of the slag pieces. According to the method, TBM tunneling parameters can beestimated online, and meanwhile normal operation of the TBM is not interfered.

The invention discloses an interactive physical education teaching auxiliary system and method based on human body posture recognition. The system comprises a depth camera device, which is used for acquiring a user action depth image sequence and a standard action depth image sequence; a conventional camera device for acquiring a standard motion demonstration image sequence; A standard movement training unit for establishing a standard movement model; a standard action storage unit for storing a standard action model and a standard action demonstration image sequence; a display terminal for displaying a standard action demonstration image sequence to a user; An action comparison unit configured to compare a user action depth image sequence at a predetermined time point or time period witha standard action model at a corresponding time point or time period, and obtain a difference between the user action and the standard action model; a presentation means for outputting action evaluation or action correction presentation information to the display means according to the difference. The invention does not require a user to wear a special identification device, and the system manufacturing cost is reduced and the user experience is improved.

A system and method for provisioning Internet access to a computing device comprising an Internet access adjustment facilitator arranged to receive a request from the computing device for adjusting Internet access from a paid Internet access mode to a toll-free Internet access mode; and a whitelist in communication with the Internet access adjustment facilitator, the whitelist maintaining a list of web resources available for toll free access by the computing device; wherein upon successful processing of the request, the list of web resources are toll free for access by the computing device. The system may further be adapted for billing / charging based on either pay-per-specified-time model or pay per action model.

The invention relates to an entity behavioral modeling assembling method and system and belongs to the field of generating troops by a computer. The assembling system comprises a behavior editing component, a model base and an operation-time task managing device. The behavior editing component is used for modeling for an entity task, an atom behavior model represents basic actions and the atom behavior model and / or a combination action model is assembled by means of a flow chart or a finite-state machine or a rule set. Sub-engines of the atom behavior model and the operation-time task managing device achieve a uniform interface calling specification. The assembling method is based on the system and a field expert decomposes entity tasks from top to bottom layer by layer through the behavior editing component and performs modeling. When the tasks are allocated to an operation-time simulation entity, the operation-time task managing device analyzes and performs a task behavior model. The entity behavioral modeling assembling method and system simplify a modeling process, improves simulation development efficiency and reduces development cost. The model is repeatedly used so as to reduce stored resources of a computer and reduce working loads in maintaining the model.

The invention discloses an Event-Action conflict resolution method and an Event-Action conflict resolution device. The method comprises the following steps: when an Event is triggered, determining an Action matched with the Event; acquiring the type of a currently-executed conflict operation which corresponds to the Action, wherein types of the conflict operation comprise manual trigger and Event trigger; and executing the Action by use of different modes based on different types. According to the method and the device, the problem that poor user experience is caused by the Event-Action conflict operation in the prior art is solved, near or reverse Action conflict caused by different Events in an Event-Action model is avoided, and the intelligence of a household appliance is improved, and the user experience is improved and furthermore the equipment operation is safer.

The invention discloses a road slope three-dimensional visualization modeling method, and particularly relates to the technical field of road slope modeling, and the method comprises the following modeling steps: S1, constructing a roadbed slope three-dimensional visualization model; S2, constructing a slope structure model and displaying vector data information. According to the invention, technologies such as geographic information, virtual reality, big data management, roadbed slope engineering and the like are comprehensively applied; the construction of a roadbed slope three-dimensional visual live-action model and a navigation technology are researched. a three-dimensional visualization scheme of the slope is constructed and the technical problem of compatible display of the model ona 'cloud eye' system platform is solved; meanwhile, structure vector data on the side slope body is subjected to integrated processing; the structure three-dimensional visualization model is constructed, roadbed slope basic information, a protection and reinforcement structure, a construction process and safety monitoring and maintenance information are fused and visually, clearly and selectivelydisplayed through the three-dimensional model, the loading speed during browsing is high, and the user experience is good.

The invention discloses a case-related news viewpoint sentence sentiment classification method based on a CNN-BiLSTM + action model. The model comprises a word embedding layer, a convolution layer, apooling layer, a BiLSTM layer, an Attention attention layer and a Softmax classification layer. The method comprises the following steps: step 1, preprocessing viewpoint sentences of case-related news, and then encoding all words into word vectors through a word embedding layer; 2, inputting the word vector obtained in the step 1 into a convolution layer, and carrying out convolution operation; 3,the feature vectors obtained after convolution are input into a pooling layer to be subjected to maximum pooling operation; 4, inputting the feature vector obtained after the maximum pooling into a BiLSTM layer; step 5, performing Attention operation on each hidden state vector obtained by the BiLSTM layer; and step 6, classifying the output obtained by the Attention operation through a softmax classification layer to obtain a target emotion category probability. According to the method, complex feature engineering and redundant manual work can be effectively solved, and the workload and complexity can be reduced to a great extent.

Disclosed is an operating environment system and method for an improved efficiency of a workflow executed on a computer by a user, the method being allowed to be executed on the operating environment system, comprising a plurality of soft applications and an integrated operating environment, comprising an action model module described in a data structure form, a plug-in component module monitoring the user's operating status and intercepting and transmitting an issued instruction, a command center module interpreting the possible action in response to the instruction based on the data structure and a service module directed by the command center module to execute the possible action.

The invention discloses an intelligent mirror design method based on automatic image recognition and action model comparison, belonging to the technical field of image recognition. The design method collects human body images through a Kinect sensor; Acquiring a human skeleton image according to the human body image; Human skeleton points are recognized according to human skeleton images. According to the skeleton point, calculate the skeleton angle, establish the actual movement model; Comparing the actual action model with the initialized standard action, the action fitting percentage is obtained. Actual actions and standard actions are displayed on the display screen and prompted in voice, text or image form. The intelligent mirror provided by the invention has the advantages of timelyand accurate feedback, and provides various prompt information such as voice and image.

The method includes following steps: (1) recording request of accessing file resources generated in computer system in real time; (2) carrying out pretreatment for the recorded requests; (3) using relation tree of file access records relation of file access described in access records of normal files; (4) relation tree of file access is divided into two parts of fixed part and variation part based on rarity of time t-Rarity; (5) analyzing records of file access, and selecting normal actions of file access, and reporting abnormal actions of file access to safety officer; (6) updating relation tree of file access periodically, and deleting out outdated relation of file access recorded in relation tree of file access. Mechanisms of selective self-learning and self-adapting forgetting are adopted in the invention. Thus, safe and reliable normal action model is built automatically without manual intervention. The invention detects lots of malicious acts with high precision.

The invention relates to a virtual anchor method based on a non-humanoid character of three-dimensional animation, which comprises the following steps: creating a three-dimensional model of the non-humanoid virtual anchor of three-dimensional animation, adding a texture file, and establishing an action model; importing the data file created by non-human virtual anchor in 3D animation into 3D real-time engine; matching a moving capture actor to an impersonal virtual anchor, capturing real actor motion information in real time, and then associating and driving the working model of the impersonalvirtual anchor; superimposing the real-time composite animation of a virtual anchor of a non-humanoid character on the recorded video picture and transmitted live. The invention also relates to a device and a system for a virtual anchor based on a non-humanoid character of three-dimensional animation. The invention has the beneficial effects that the actions of non-human characters in the three-dimensional animation are generated by capturing the actions or even expressions of actors in real time, and the interaction between the virtual anchor and the audience is improved.

The invention discloses a multi-target behavior action recognition and prediction method. The method comprises the following steps: firstly, establishing a behavior action model according to characteristic data information of an individual or an individual animal, sensor data collected by various portable monitoring devices carried by the individual or the individual animal and behavior action information of the individual or the individual animal; processing the sensor data or the action flow data of the individual or the individual animal in a preset time period to generate a data fragment graph; and calculating the data fragment graph by using the corresponding behavior action model to realize recognition of behavior actions of individuals or individual animals within a preset time and / or prediction of behavior actions after a preset time period. The method can achieve the recognition or prediction of various types of behavior actions, and provides an intelligent recognition, prediction and recording method for daily life behavior activities for individuals or individual animals. The invention further discloses electronic equipment and a storage medium.

A flexible part orientation acquiring method under attitude orbit control maneuvering action includes: performing structural finite element modeling on a reflective plane antenna to build a satellite rigid-flexible coupling dynamics model, and acquiring a vibration mode vector of the reflective plane antenna; building a satellite attitude control action model, and enable the same and the satellite rigid-flexible coupling dynamics model to form a satellite dynamics model; establishing an expression formula of antenna radiation field intensity under antenna vibration mode space, and building a whole satellite system on-orbit state dynamics-attitude control-antenna radiation comprehensive model; simulating according to satellite on-orbit working excitation data to acquire time change mode coordinates of antenna vibration, substituting the time change mode coordinates into the radiation field mode space expression formula, and acquiring dynamic influence change of on-orbit working mode on antenna radiation field intensity.