International Business Machines Corporation

Yinglong Xia

Jui-Hsin Lai

Lifeng Nai

Ching-Yung Lin

Toyotaro Suzumura

Xavier Rene Guerin

Yu Ma

Wan-Yi Lin

Conglei Shi

Chia-Yu Chen

Pei-Yun S Hsueh

Fei Wang

Wei Tan

Kun-Lung Wu

Anni Rosa Coden

Ilie G Tanase

Hanghang Tong

Yanbin Liu

Toyotaro Suzumara

Lifeng Nail

A query image is obtained. In a database including a plurality of reference image graphs, at least one of the reference image graphs, with feature vectors similar to the query image, is identified. Image querying is carried out by graph traversal on the at least one of the reference image graphs with the feature vectors similar to the query image. An image from the at least one of the reference image graphs having a highest matching score in the graph traversal is returned as a response to the query image. techniques for building the database are also disclosed.

Graph-based online image queries

A data storage and retrieval system for a computer memory includes a stream graph engine extracting graph data from streaming data, the graph data occupying a sliding window and comprising a plurality of slices representing a set of contiguous graphs, and where each slice of the plurality of slices corresponds to a given graph structure and its properties at a particular time, the stream graph engine causing the computer memory to store an on-disk portion of the graph data including a key table, a timestamp table and a plurality of property files, wherein the key table comprises a plurality of pointers to corresponding entries of the timestamp table, wherein each of the entries of the timestamp table comprise a corresponding timestamp and a pointer to the properties files, wherein the properties files comprise properties of a corresponding graph of the set of contiguous graphs, an in-memory portion of the graph data having a cache data structure storing a subset of the key table, and a versioning control module storing evolving changes in the graph data in the computer memory.

Efficient property graph storage for streaming/multi-versioning graphs

Method and apparatus for monitoring eating behavior are disclosed. A wearable device including an imaging sensor, an electromyography (EMG) sensor, and a processing unit is configured to collect EMG data via the EMG sensor, transmit the EMG data to a computing device, receive a control signal to capture one or more images via the imaging sensor, capture, one or more images, and transmit the one or more images. A computing device is disclosed including a processor and a memory containing a machine learning model and computer program code that, when executed, performs an operation. The operation includes receiving EMG data, processing the EMG data as input to a trained machine learning model, transmitting the control signal to capture one or more images via the imaging sensor, receiving the one or more images, and determining nutritional content of the food or beverage using the one or more images.

System for monitoring eating habit using a wearable device

A Resilient Distributed Dataset (RDD) processing method, system, and computer program product include sending a different portion including a global identifier and different values of a message from a receiver to a sender and at a receiver side, an already-processed message is extracted from an RDD object repository corresponding to the global identifier of the different portion and the incoming differential portion is applied to the already-processed message.

Differential processing mechanism for spark-based graph computing background

A remote control system detects a directional signal at an alerting sensor package; activates an appliance to accept a control signal in response to the directional signal; accepts the control signal at the appliance from a muscle sensor package; and adjusts operation of the appliance in response to the control signal.

Remote control with muscle sensor and alerting sensor

A computer-implemented method, computer program product, and system for determination of critical parts and component correlations in a circuit using a correlation graph and centrality analysis including; receiving a circuit layout portion of a larger circuit layout, converting the circuit layout portion into a correlation graph representing components as nodes and connecting wires as edges, determining, using ground truth and Na&#xef;ve Bayes to determine correlation weighting, scaling the correlation graph to represent the larger circuit, and presenting the larger correlation graph on a graphical user interface (GUI).

Graph method for system sensitivity analyses

One or more biological signals are obtained. The one or more biological signals are converted to one or more graph structures. Correlation between two or more of the biological signals are determined using the one or more graph structures. One or more changes in the one or more graph structures within a time window are recorded. A signal graph model is generated based on the recorded changes.

Graph-based analysis for bio-signal event sensing

Identity recognition is achieved using user's living signatures, such as a heartbeat signal, from multiple devices. A user requests identity recognition for using a first registered device, such as a mobile device. The request is provided to a cloud. When the cloud confirms the mobile device is a registered device, the cloud directs the first registered device to upload the user living signature to the cloud. The cloud also sends a request to a user second registered device to upload the user living signature from the second registered device to the cloud. The user second registered device may be a headband monitor, watch, Google glasses, or the like capable of monitoring and uploading the user living signature to the cloud. The cloud compares both living signature signals from the first registered device and from the second registered device. If the signals are matched, the user identity is confirmed. If the signals are not matched, the user identity is not confirmed. Preferably the first registered device is a mobile device.

Identity recognition with living signatures from multiple devices

At a mobile electronic device, first sensor data is obtained from at least one wearable muscle sensor, suggesting that location tracking will imminently be desirable. A first indication that an external locator signal is not usable is also obtained. Responsive to the first sensor data and the indication, second sensor data, from at least one wearable muscle sensor, indicative of motion of a subject wearing the at least one muscle sensor, is recorded. A second indication that the external locator signal is now usable is obtained. Responsive to the second indication, the external locator signal is used to determine a current location of the mobile electronic device; and the second sensor data is sued to backtrack from the current location of the mobile electronic device to indicate a location of the mobile electronic device at a time when the recording of the second sensor data was initiated.

Trace/trajectory reconstruction via wearable and/or mobile sensors for indoor/outdoor location

At each of a plurality of unmanned aerial vehicles operating in a three-dimensional space, current relative locations of other ones of the plurality of unmanned aerial vehicles which are visible to each of the plurality of unmanned aerial vehicles are determined based on two-dimensional camera images from on-board dual-lens cameras. The current relative locations are wirelessly transmitted to a controller. At the plurality of unmanned aerial vehicles, specification of a path to be followed by each of the plurality of unmanned aerial vehicles is wirelessly received from the controller. At each of the plurality of unmanned aerial vehicles, on-board collision avoidance is carried out based on the two-dimensional camera images from the on-board dual-lens cameras, while following the path specification.

Multiple unmanned aerial vehicle autonomous coordination

A system and method for painting a structure. The system comprises: a computer vision processing system configured to obtain images of a target structure and generate first instructions signals for real time communication to an unmanned aerial vehicle (UAV) having a paint fluid dispensing system provided thereon; a control device at the UAV responsive to received the first instruction signals for controlling real time navigating of the UAV to a location at the target structure; and the computer vision processing system configured to generate second instruction signals for real time communication to the UAV, wherein the control device at the UAV is configured to automatically actuate the paint fluid dispensing system to apply a paint fluid at the location on the target structure in response, the received second instruction signals configuring the UAV to render a desired visual image on the target structure.

Automatic painting system with drone, user interface and computer vision

A method for clustering vertices of streaming graphs includes: maintaining one or more clusters, assigning a random number to an incoming edge, computing a sampling threshold based on the current clusters, and adjusting the current clusters based on the random number and the sampling threshold.

Adaptive sampling schemes for clustering streaming graphs

Determining a number of kernels within a model is provided. A number of kernels that include data samples of a majority data class of an imbalanced training data set is determined based on a set of generated artificial data samples for a minority data class of the imbalanced training data set. The number of kernels within the model is generated based on the set of generated artificial data samples. A likelihood of the set of generated artificial data samples being included in the majority data class of the imbalanced training data set is calculated. Parameters of each kernel in the number of kernels are updated based on the likelihood of the set of generated artificial data samples being included in the majority data class of the imbalanced training data set. Each kernel in the number of kernels is adjusted based on the updated parameters.

Determining a number of kernels using imbalanced training data sets

A system, method and computer program product for real-time recognition of individuals. The method comprises: receiving at a hardware processor, a first sensor data associated with the individual. The first sensor data associated with the individual is processed to determine an activity the individual is currently engaged in. Additionally, a second sensor data associated with the individual is received and processed to determine a feature associated with a muscle of said individual while engaged in activity. Based on the received sensor data and determined activity and muscle features, the method generates a model that correlates determined activities and corresponding muscle features of the individual user with the individual's identity. The model is subsequently used to identify the individual. By receiving and inputting real-time, sensor data from an individual into said generated model, the model generates and determines: an activity and a muscle feature for use in identifying the individual.

Growing community of inventors

Rye Brook, NY, United States of America

Yinglong Xia