296 results about "Activity recognition" patented technology

Methods and apparatus proved for: simultaneously monitoring at least one of nerve and muscle electrical activity on respective sides of a target plexus of a patient; identifying desired electrical activity of at least one of nerves and muscles on both sides of the target plexus based on the monitored activity; and modulating the electrical activity of the at least one of nerves and muscles on both sides of the target plexus to achieve a therapeutic result.

There is provided an activity recognition apparatus for detecting an activity of a subject. The apparatus includes: a sensor unit including a plurality of linear motion sensors configured to detect linear motions and a plurality of rotational motion sensors, the linear motions being orthogonal to each other, the rotational motions being orthogonal to each other; and a computational unit configured to receive and process signals from the sensors included in the sensor unit so as to detect an activity of the subject. The sensor unit is directly or indirectly supported by the subject with an arbitrary orientation with respect to the subject. The computational unit performs a calculation that uses the signals from both linear motion sensors and rotational motion sensors to determine the activity of the subject independent of the orientation of the sensor unit.

A system, method, and computer program product for providing a user with a virtual environment in which the user can perform guided activities and receive feedback are described. The user is provided with guidance to perform certain movements. The user's movements are captured in an image stream. The image stream is analyzed to estimate the user's movements, which is tracked by a user-specific human model. Biomechanical quantities such as center of pressure and muscle forces are calculated based on the tracked movements. Feedback such as the biomechanical quantities and differences between the guided movements and the captured actual movements are provided to the user.

There is provided an activity recognition apparatus for detecting an activity of a subject. The apparatus includes: a sensor unit including a plurality of linear motion sensors configured to detect linear motions and a plurality of rotational motion sensors, the linear motions being orthogonal to each other, the rotational motions being orthogonal to each other; and a computational unit configured to receive and process signals from the sensors included in the sensor unit so as to detect an activity of the subject. The sensor unit is directly or indirectly supported by the subject with an arbitrary orientation with respect to the subject. The computational unit performs a calculation that uses the signals from both linear motion sensors and rotational motion sensors to determine the activity of the subject independent of the orientation of the sensor unit.

Video and corresponding metadata is accessed. Events of interest within the video are identified based on the corresponding metadata, and best scenes are identified based on the identified events of interest. A video summary can be generated including one or more of the identified best scenes. The video summary can be generated using a video summary template with slots corresponding to video clips selected from among sets of candidate video clips. Best scenes can also be identified by receiving an indication of an event of interest within video from a user during the capture of the video. Metadata patterns representing activities identified within video clips can be identified within other videos, which can subsequently be associated with the identified activities.

System and method for automatic computer aided optical recognition of toys, for example, construction toy elements, recognition of those elements on digital images and associating the elements with existing information is presented. The method and system may recognize toy elements of various sizes invariant of toy element distance from the image acquiring device for example camera, invariant of rotation of the toy element, invariant of angle of the camera, invariant of background, invariant of illumination and without the need of predefined region where a toy element should be placed. The system and method may detect more than one toy element on the image and identify them. The system is configured to learn to recognize and detect any number of various toy elements by training a deep convolutional neural network.

New activity recognition, recording, analysis and control techniques, systems and sensors are provided. In some aspects of the invention, multiple sensory tags with unique identification and data transfer attributes create positional, movement, orientation and acceleration data and supply it to a control system. The tags may be placed at location(s) on the user's body, clothing, personal effects, exercise equipment and other activity-relevant locations to enhance activity recognition and mapping. For example, a specialized PDA may comprise part of the control system, and may provide, reflect and / or receive modified data signals to and from such tags.In other aspects, a system defines a personal activity space, samples data preferentially from that space, and performs a simplified form of object-recognition to determine, record and analyze personal activities.A GUI interface within the control system may also aid in recognizing and recording activity patterns, in addition to tags.

A platform is provided for generating, transmitting, and processing sensor data. The platform can provide various components with which a sensor can interface to allow the sensor to provide sensor data using a common mechanism. The platform can include an API which provides a common interface for communicating with a wearable sensor device. The API can allow an application to receive activity recognition and biometric data from one or more sensors in a processed and usable format thereby facilitating the usage of such data. The platform can also comprise a sensor data processor configured to process raw sensor data into a usable format for consumption by other applications, and a sensor data interface for transmitting the raw sensor data to a mobile device for processing by the sensor data processor.

One embodiment of the present invention provides a system that provides a user interface for a context-aware recommendation system which can be operated with one hand. The user interface includes a touch-screen display and a list of recommendations presented on the touch-screen display. The user interface further includes a plurality of graphic elements presented on the touch-screen display which allow a user to configure preferences associated with the activities. A layout of the leisure activities and the graphic elements allows the user to hold the touch-screen display and perform operations on the leisure activities and / or the graphic elements with one hand.

A system and method for researching an identity of a source of activity that is indicative of pestware is described. In one embodiment the method comprises monitoring the computer for activity that is indicative of pestware, identifying, based upon the activity, an object residing on the computer that is a suspected pestware object; and accessing at least a portion of a recorded history of sources that the computer received files from so as to identify a reference to an identity of a particular source that the suspected pestware object originated from.

Binary motion events are detected by individual motion sensors placed in a physical environment. The motions events are transmitted to a cluster leader, each motion detector being a cluster leader of immediately spatially adjacent motion sensors. Movements of objects are detected by the cluster leaders according to the motion events. The movements are transmitted to supercluster leaders, each motion detector being a supercluster leader of immediately spatially adjacent motion clusters of sensors. Activities of the objects are detected by the supercluster leaders, and actions of the objects are detected according to the activities.

Various embodiments for creating media clips are disclosed. Media clips are created by content capture and tagging devices as events are recorded by the content capture and tagging devices. Users select one or more predefined activity indictors displayed on their content capture and tagging devices as activities occur during events, or as recordings are viewed after events have concluded. Media clips are then created based on the recordings and the activity type associated with each activity indicator. The media clips are then provided to a server, where they are stored in association with activity identification information, such as an activity type based on the selected activity indicator. The media clips may then be searched and accessed by providing key words to the server.

Described is a system for visual activity recognition that includes one or more processors and a memory, the memory being a non-transitory computer-readable medium having executable instructions encoded thereon, such that upon execution of the instructions, the one or more processors perform operations including detecting a set of objects of interest in video data and determining an object classification for each object in the set of objects of interest, the set including at least one object of interest. The one or more processors further perform operations including forming a corresponding activity track for each object in the set of objects of interest by tracking each object across frames. The one or more processors further perform operations including, for each object of interest and using a feature extractor, determining a corresponding feature in the video data. The system may provide a report to a user's cell phone or central monitoring facility.

The invention discloses a digital device control method based on somatosensory recognition. The digital device control method includes a user authority management step, a user action recognition step and a control step. The user authority management step includes displaying an operation control frame corresponding to users within a display area of a digital device when detecting requests of the users to acquire control authority, and distributing the control authority of the digital device to the users on the basis of authority management mechanisms. The user action recognition step includes judging whether to perform key operation or not according to the action recognition within the operation control frame, if yes, laying out a keyboard within the operation control frame, and recognizing on the basis of the first type of actions transmitted by the users through the keyboard, if not, recognizing the second type of actions transmitted by the users; the control step includes controlling the digital device on the basis of recognition results according to the preset correspondence between the actions and the operation of the digital device. The digital device control method is complicated in operation of the digital device, simple to operate and good in user experience.

Method and system for objects surveillance and real-time activity recognition is based on analysis of spatio-temporal images of individuals under surveillance where a spatio-temporal volume occupied by each individual is decomposed by crossing the same at specific heights to form 2-dimensional slices, each containing representation of trajectory of the motion of corresponding portions of the individual body. The symmetry of the trajectories (Gait DNA) is analyzed and classified to generate data indicative of a type of activity of the individual based on the symmetry or asymmetry of the Gait DNA in each 2-dimensional slice. An effective occlusion handling ability is implemented which permits to restore the occluded silhouette of an individual.

Multi-source multi-modal activity recognition for conducting aerial video surveillance comprising detecting and tracking multiple dynamic targets from a moving platform, representing FMV target tracks and chat-messages as graphs of attributes, associating FMV tracks and chat-messages using a probabilistic graph based mapping approach; and detecting spatial-temporal activity boundaries.

The invention discloses a user activity recognition method and a user activity recognition device. The user activity recognition method is used for an intelligent terminal with a sensor, and includes: obtaining first data of the sensor; calculating the first data according to a first algorithm so as to obtain a first activity recognition result; using an amending model built according to training data to amend the first activity recognition result, and obtaining a second activity recognition result, wherein the training data includes at least one group of user relevance information and history data, the history data comprises a user activity recognition result and a user activity of a user label corresponding to the user activity recognition result, and the user relevance information at least includes one item of obtaining time, an obtaining position, user personal information and application use state information of sensor data used to confirm the user activity recognition result; confirming the second activity recognition result as the user activity. The user activity recognition method improves recognition accuracy for the user activity.

The invention discloses an information push method based on internet-surfing log mining and user activity recognition.The method includes a data preprocessing stage, a potential pushing user screening stage and a timely push stage.The data preprocessing stage comprises a webpage portrait, webpage level classification and a user portrait.In the potential pushing user screening stage, similarity between new information and historical web pages which users have access to is solved through a method based on matrix decomposition, and potential pushing users who may be interested in the new information are screened out.The timely push stage includes situational information acquisition and activity recognition based on a mobile intelligent device, and information timely pushing adaptive to the activity situation.By mining internet-surfing log data, potential users interested in information are found, when new information is generated, the new information is pushed to the potential users in good time, and therefore information push efficiency is improved.

A system and method for implementing a machine learning-based system for generating event intelligence from video image data that: collects input of the live video image data; detects coarse features within the live video image data; constructs a coarse feature mapping comprising a mapping of the one or more coarse features; receives input of the coarse feature mapping at each of a plurality of distinct sub-models; identify objects within the live video image data based on the coarse feature mapping; identify one or more activities associated with the objects within the live video image data; identify one or more interactions between at least two of the objects within the live video image data based at least on the one or more activities; composes natural language descriptions based on the one or more activities associated with the objects and the one or more interactions between the objects; and constructs an intelligence augmented live video image data.

A system and method for researching an identity of a source of activity that is indicative of pestware is described. In one embodiment the method comprises monitoring the computer for activity that is indicative of pestware, identifying, based upon the activity, an object residing on the computer that is a suspected pestware object; and accessing at least a portion of a recorded history of sources that the computer received files from so as to identify a reference to an identity of a particular source that the suspected pestware object originated from.

The disclosure provides an approach for recognizing and analyzing activities. In one embodiment, a learning application trains parameters of a hierarchical model which represents human (or object) activity at multiple levels of detail. Higher levels of detail may consider more context, and vice versa. Further, learning may be optimized for a user-preferred type of inference by adjusting a learning criterion. An inference application may use the trained model to answer queries about variable(s) at any level of detail. In one embodiment, the inference application may determine scores for each possible value of the query variable by finding the best hierarchical event representation that maximizes a scoring function while fixing the value of the query variable to its possible values. Here, the inference application may approximately determine the best hierarchical event representation by iteratively optimizing one level-of-detail variable at a time while fixing other level-of-detail variables, until convergence.

The invention belongs to the technical field of wireless communication, and discloses an indoor personnel activity identification method based on channel state information and a man-machine interaction system. The method comprises steps of preprocessing the acquired CSI data to extract environmental change information including correlation coefficients and variances; sensing the surrounding environment by using the decision tree, monitoring the personnel activity, if the personnel activity is not detected, continuing to collect the CSI data, keeping the activity monitoring state, and otherwise, triggering an activity feature extraction module; carrying out extreme removal processing and low-pass filtering processing on the collected CSI data in the module; and performing activity matchingand heaviest personnel activity identification by using a dynamic time planning algorithm based on principal component analysis. According to the invention, indoor personnel activity monitoring and activity identification are combined, and PCA-DTW is adopted to carry out indoor personnel activity identification, the step of offline data collection training in the prior art is omitted, and the system performance and the identification rate are improved.

The invention provides a method for processing motion pattern recognition and action recognition of a moving target based on a light stream direction sequence. The method comprises the following steps of: detecting the target by using background subtraction, tracking the target by using mean shift tracking algorithm, calculating a light flow value, respectively acquiring light stream directions of a rigid body and a non-rigid body, acquiring time series of the light stream directions after quantization, then building a Hidden Markov Model by using the time series, and finally processing the motion pattern recognition and the action recognition by using the Hidden Markov Model. The method provided by the invention can simutaneously process the motion pattern recognition and the action recognition of the moving target through the light stream directions, and can be widely applied to various civil or military systems such as a video monitoring system, a video conference system, an industrial product testing system, a robot visual navigation system, a military target testing classification system, thus having broad market prospect and application value.

Various embodiments of the invention provide methods, systems, and computer program products for generating a roster template used in scheduling agent resources to conduct a first campaign and a second campaign involving two different channels of communication. Specifically, forecasts are generated for the first and second campaigns and agent resources needed to handle communications forecasted for each campaign are identified. A roster template is generated that includes one or more shifts to schedule based on the agent resources identified for the first campaign and evaluated to identify whether a deficiency exists with respect to the one or more shifts not meeting the agent resources identified for the second campaign. If so, the one or more shifts are revised so as not to cause a second deficiency with respect to the first campaign.

The invention provides a system of and a method for real-time magic-type stage performance based on augmented reality and action recognition, comprising a backstage input control system and a magic-type stage performance system, wherein the backstage input control system is composed of a traditional performance effect input device and a virtual character control input device, and the magic-type stage performance system is preassembled in a computer, receives data from an input module and outputs through a holographic image device, a sound device and a projection device. The system and the method can enhance stage performance effect, and enable co-performance of simulated virtual characters and actors and interaction with the audience, making the traditional stage performance more magic, more intelligent and more interactive.

The invention belongs to the field of computer vision and specifically relates to a human action recognition method based on a noise reduction automatic encoder and particle filter. The human action recognition method randomly divides a to-be-sorted action video set into a training video set and a test video set, the training video set and the test video set are respectively used for calculating training tracks and test tracks of the action, distances between the training tracks and the test tracks are calculated, a track distance set is obtained and is inputted to a support vector machine, and a sort result of the action is obtained. According to the invention, the method has an advantage of substantial human action recognition effect, accuracy of action recognition can be effectively improved, and the method has certain robustness on shielding and view change and can further be used for human body abnormal behavior recognition in video monitoring.

The invention relates to a human action recognition method. The method comprises the following steps: S1, continuous image frames comprising human action recognition feature data are acquired; S2, an effective image frame is screened, and feature data in the effective image frame serve as to-be-detected data; S3, a static posture feature vector Fcc, a continuous action feature vector Fcp and an overall action feature vector Fco of the to-be-detected data are built as action features; S4, a final feature vector Fc is built, and Fc=[Fcc, Fcp, Fco]; S5, the final feature vector Fc is subjected to dimension reduction; and S6, a well-trained model classifier is used for action recognition on the final feature vector Fc after dimension reduction, and a recognition result is obtained. Compared with the prior art, the method of the invention has the advantages of high recognition rate and small calculation amount.

The invention discloses a rail transit abnormal person detection method based on action recognition, and belongs to the field of aviation monitoring. The method comprises: inspecting railways by usingan unmanned aerial vehicle, extracting frames of videos, training and using an SSD detection model, obtaining position information of each person in each video frame image, intercepting a local areacontaining the person, and a key point detection model and a Reset-are detected. The 18-clothes classification model is trained; predicting joint coordinates of each person by using the key point detection model, forming a human skeleton sequence by using the joint coordinates in a certain time period, inputting the human skeleton sequence into an LSTM action recognition model, and recognizing theaction category of each person; and performing clothing classification of the personnel through the Resnet-18 clothing classification model; and judging whether the person is a worker or not according to the action type of each person and the corresponding appearance clothes. The method solves the problem of high cost of traditional manual inspection or roadbed inspection, and has high practicability and robustness.