Siamese reconstruction convolutional neural network for pose-invariant face recognition

    公开(公告)号:US10474883B2

    公开(公告)日:2019-11-12

    申请号:US15803292

    申请日:2017-11-03

    Abstract: A computer-implemented method, system, and computer program product is provided for pose-invariant facial recognition. The method includes generating, by a processor using a recognition neural network, a rich feature embedding for identity information and non-identity information for each of one or more images. The method also includes generating, by the processor using a Siamese reconstruction network, one or more pose-invariant features by employing the rich feature embedding for identity information and non-identity information. The method additionally includes identifying, by the processor, a user by employing the one or more pose-invariant features. The method further includes controlling an operation of a processor-based machine to change a state of the processor-based machine, responsive to the identified user in the one or more images.

    Gas concentration measurement by 2F signal trough distance

    公开(公告)号:US10466106B2

    公开(公告)日:2019-11-05

    申请号:US16219136

    申请日:2018-12-13

    Abstract: A computer-implemented method for measuring gas concentration from a 2f signal in wavelength modulation spectroscopy is presented. The computer-implemented method includes emitting a beam of light from a laser to pass through a gas sample, calculating a gas measurement value from the gas sample via a trough distance calculator using a trough distance of a gas absorption line's 2f signal, calibrating the gas measurement value via a multi-point calibration process, and outputting the gas measurement value to a user interface of a computing device.

    SIGNATURE-BASED RFID LOCALIZATION
    645.
    发明申请

    公开(公告)号:US20190311162A1

    公开(公告)日:2019-10-10

    申请号:US16376039

    申请日:2019-04-05

    Abstract: A Radio Frequency Identification (RFID) localization system is provided. The system includes a set of passive RFID tags, each for reflecting transmitted signals. The system further includes an RFID reader for detecting the reflected signals by the passive RFID tags. The system also includes a processor for localizing an object in an area based on the reflected signals by computing signatures using probabilistic macro-channels between the RFID reader and locations of the passive RFID tags. The transmitted signals form inputs to the probabilistic macro-channels, and the signatures form outputs from the probabilistic macro-channels.

    DECENTRALIZED ENERGY MANAGEMENT UTILIZING BLOCKCHAIN TECHNOLOGY

    公开(公告)号:US20190288513A1

    公开(公告)日:2019-09-19

    申请号:US16257399

    申请日:2019-01-25

    Abstract: A system and methods are provided for a decentralized transactive energy management. The method includes calculating, by a processor-device, power balancing at one of a plurality of nodes responsive to current statistics at the one of a plurality of nodes. The method also includes estimating, by the processor-device, a present energy demand for the one of a plurality of nodes responsive to the current statistics. The method additionally includes obtaining, by the processor-device, an amount of excess energy available another of the plurality of nodes. The method further includes optimizing, by the processor-device, a power flow between the one of the plurality of nodes and the another of the plurality of nodes to satisfy the present energy demand for the one of the plurality of nodes. The method also includes transferring the excess energy from the another of the plurality of nodes to the one of the plurality of nodes.

    Discovering critical alerts through learning over heterogeneous temporal graphs

    公开(公告)号:US10409669B2

    公开(公告)日:2019-09-10

    申请号:US15810960

    申请日:2017-11-13

    Abstract: A method is provided that includes transforming training data into a neural network based learning model using a set of temporal graphs derived from the training data. The method includes performing model learning on the learning model by automatically adjusting learning model parameters based on the set of the temporal graphs to minimize differences between a predetermined ground-truth ranking list and a learning model output ranking list. The method includes transforming testing data into a neural network based inference model using another set of temporal graphs derived from the testing data. The method includes performing model inference by applying the inference and learning models to test data to extract context features for alerts in the test data and calculate a ranking list for the alerts based on the extracted context features. Top-ranked alerts are identified as critical alerts. Each alert represents an anomaly in the test data.

Patent Agency Ranking