REDES NEURONALES

VolverVolver

Resultados 154 resultados LastUpdate Última actualización 26/03/2019 [17:47:00] pdf PDF

Solicitudes publicadas en los últimos 60 días / Applications published in the last 60 days

Página1 de 7 nextPage   por página


VIDEO CLASSIFICATION METHOD, INFORMATION PROCESSING METHOD AND SERVER

NºPublicación: WO2019052301A1 21/03/2019

Solicitante:

TENCENT TECH SHENZHEN CO LTD [CN]

Resumen de: WO2019052301A1

Disclosed is an information processing method, comprising: acquiring a video to be processed; sampling, according to a time characteristic sampling rule, the video to be processed, and acquiring at least one video frame characteristic sequence, the time characteristic sampling rule being a correlation between a time characteristic and the video frame characteristic sequence; processing the at least one video frame characteristic sequence by means of a first neural network model in order to obtain a characteristic expression result of the at least one video frame characteristic sequence, the first neural network model being a recursive neural network model; and processing the characteristic expression result of the at least one video frame characteristic sequence by means of a second neural network model in order to obtain a prediction result corresponding to the at least one video frame characteristic sequence, with the prediction result being used for determining the class of the video to be processed. In the present application, during the process of classifying videos, a characteristic change of the videos in a time dimension is also considered, so as to better express the content of the videos, improve the accuracy rate of video classification, and improve the video classification effect.

traducir

PATTERN RECOGNITION APPARATUS, PATTERN RECOGNITION METHOD, AND STORAGE MEDIUM

NºPublicación: WO2019053898A1 21/03/2019

Solicitante:

NEC CORP [JP]

Resumen de: WO2019053898A1

Provided is a pattern recognition apparatus to provide classification robustness to any kind of domain variability. The pattern recognition apparatus 500 based on Neural Network (NN) includes: NN training unit 501 that trains an NN model to generate NN parameters, based on at least one first feature vector and at least one domain vector indicating one of subsets in a specific domain, wherein, the first feature vector is extracted from each of the subsets, the domain vector indicates an identifier corresponding to the each of the subsets; and NN verification unit 502 that verifies a pair of second feature vectors in the specific domain to output whether the pair indicates same individual or not, based on a target domain vector and the NN parameters.

traducir

METHOD AND APPARATUS FOR DETECTING HUMAN FACE

NºPublicación: US2019087686A1 21/03/2019

Solicitante:

BAIDU ONLINE NETWORK TECHNOLOGY BEIJING CO LTD [CN]

CN_107644209_A

Resumen de: US2019087686A1

The present disclosure discloses a method and apparatus for detecting a human face. A specific embodiment of the method comprises: acquiring a to-be-detected image; inputting the to-be-detected image into a pre-trained first convolutional neural network to obtain facial feature information, the first convolutional neural network being used to extract a facial feature; inputting the to-be-detected image into a pre-trained second convolutional neural network to obtain semantic feature information, the second convolutional neural network being used to extract a semantic features of the image; and analyzing the facial feature information and the semantic feature information to generate a face detection result. This embodiment improves accuracy of a detection result of a blurred image.

traducir

METHOD AND APPARATUS FOR FACIAL RECOGNITION

NºPublicación: US2019087647A1 21/03/2019

Solicitante:

BAIDU ONLINE NETWORK TECHNOLOGY BEIJING CO LTD [CN]

CN_107622240_A

Resumen de: US2019087647A1

Embodiments of the present disclosure disclose a method and apparatus for facial recognition. A specific embodiment of the method includes: acquiring a to-be-recognized image; inputting the to-be-recognized image into a pre-trained first convolutional neural network to obtain complete facial feature information and partial facial feature information, the first convolutional neural network being used to extract a complete facial feature and a partial facial feature; and inputting the complete facial feature information and the partial facial feature information into a pre-trained second convolutional neural network to obtain a facial recognition result, the second convolutional neural network being used to represent a correlation between the facial recognition result, and the complete facial feature information and the partial facial feature information. This embodiment improves the accuracy of the recognition result in a situation where a face is partially covered.

traducir

METHODS AND SYSTEMS FOR BALL GAME ANALYTICS WITH A MOBILE DEVICE

NºPublicación: US2019087661A1 21/03/2019

Solicitante:

NEX TEAM INC [US]

Resumen de: US2019087661A1

Methods and systems for ball shot attempt detection and game analytics generation are disclosed. The methods and systems perform steps to receive an input video of a ball gameplay, wherein the input video was captured using a stationary camera, and wherein frames of the input video comprises a goal; identify a Region of Interest (ROI) surrounding the goal by performing a first computer vision algorithm on the input video; detect a ball within the ROI during a shot attempt and determining a trajectory of the ball by performing a second computer vision algorithm on the input video; and identify a player relevant to the shot attempt based on the trajectory of the ball. In some embodiments, the computer vision algorithms comprise a convolution neural network (CNN). The present invention uses computer vision techniques to enable a resource-limited mobile device such as a smartphone to efficiently execute the new process.

traducir

INFERRING PETROPHYSICAL PROPERTIES OF HYDROCARBON RESERVOIRS USING A NEURAL NETWORK

NºPublicación: WO2019055774A1 21/03/2019

Solicitante:

SAUDI ARABIAN OIL CO [SA]
ARAMCO SERVICES CO [US]

Resumen de: WO2019055774A1

Received image data is enhanced to create enhanced image data using image processing to remove artifacts and to retrieve information associated with a desired target output. Image segmentation is performed on useable enhanced image data to created segmented image data by partitioning the enhanced image data into coherent regions with respect to a particular image-based criterion. Useable segmented image data and auxiliary data is pre-processing for input into a neural network as pre-processed data. The pre-processed data is divided into training, validation, and testing data subsets. A neural network architecture is determined to process the pre-processed data and the determined neural network architecture is executed using the pre-processed data. Output of the determined neural network is post-processed as post-processed data. The post-processed data is compared to a known value range associated with the post-processed data to determine if the post-processed data satisfies a desired output result.

traducir

METHOD AND APPARATUS FOR IDENTIFYING TRAFFIC LIGHT

NºPublicación: US2019087673A1 21/03/2019

Solicitante:

BAIDU ONLINE NETWORK TECHNOLOGY BEIJING CO LTD [CN]

Resumen de: US2019087673A1

The disclosure discloses a method and apparatus for identifying a traffic light. An embodiment of the method comprises: zooming a to-be-processed image acquired by an image acquisition device by at least one preset ratio to obtain at least one zoomed image; inputting the at least one zoomed image into a pre-trained convolutional neural network to obtain location information and category information of a traffic light corresponding to each zoomed image of the at least one zoomed image, wherein the convolutional neural network is used for retrieving location information and category information of a traffic light displayed in an image; and analyzing the obtained location information and category information to generate at least one candidate traffic light identification result, and fusing the generated candidate traffic light identification result to generate a traffic light identification result corresponding to the to-be-processed image. The embodiment improves the accuracy in identifying a traffic light.

traducir

METHOD FOR LOCATING ONE OR MORE CANDIDATE DIGITAL IMAGES BEING LIKELY CANDIDATES FOR DEPICTING AN OBJECT

NºPublicación: US2019087687A1 21/03/2019

Solicitante:

AXIS AB [SE]

EP_3457324_A1

Resumen de: US2019087687A1

A method for finding one or more candidate digital images being likely candidates for depicting a specific object comprising: receiving an object digital image depicting the specific object; determining, using a classification subnet of a convolutional neural network, a class for the specific object depicted in the object digital image; selecting, based on the determined class for the specific object depicted in the object digital image, a feature vector generating subnet from a plurality of feature vector generating subnets; determining, by the selected feature vector generating subnet, a feature vector of the specific object depicted in the object digital image; locating one or more candidate digital images being likely candidates for depicting the specific object depicted in the object digital image by comparing the determined feature vector and feature vectors registered in a database, wherein each registered feature vector is associated with a digital image.

traducir

METHOD OF OPERATING ARTIFICIAL NEURAL NETWORK WITH NONVOLATILE MEMORY DEVICES

NºPublicación: US2019087715A1 21/03/2019

Solicitante:

UNIV CHUNG YUAN CHRISTIAN [TW]

Resumen de: US2019087715A1

The present invention discloses a method of operating artificial neural network with nonvolatile memory devices having at least one artificial neural nonvolatile memory network. In the present invention, a plurality of nonvolatile memory devices or a nonvolatile memory array comprising the nonvolatile memory devices and necessary circuit units are integrated into an artificial neural network. By such arrangement, it is able to perform feedforward and recurrent operations in the M×N number of nonvolatile memory devices in the nonvolatile memory array, so as to adjust or correct the weights stored in the M×N number of nonvolatile memory devices through the operating function of the artificial neural network with nonvolatile memory devices.

traducir

METHOD AND SYSTEM FOR CONVERTING AN IMAGE TO TEXT

NºPublicación: US2019087677A1 21/03/2019

Solicitante:

RAMOT AT TEL AVIV UNIV LTD [IL]

WO_2017163230_PA

Resumen de: US2019087677A1

In a method of converting an input image patch to a text output, a convolutional neural network (CNN) is applied to the input image patch to estimate an n-gram frequency profile of the input image patch. A computer-readable database containing a lexicon of textual entries and associated n-gram frequency profiles is accessed and searched for an entry matching the estimated frequency profile. A text output is generated responsively to the matched entries.

traducir

METHOD AND APPARATUS FOR FACIAL RECOGNITION

NºPublicación: US2019087648A1 21/03/2019

Solicitante:

BAIDU ONLINE NETWORK TECHNOLOGY BEIJING CO LTD [CN]

CN_107491771_A

Resumen de: US2019087648A1

Embodiments of the present disclosure disclose a method and apparatus for facial recognition. A specific embodiment of the method comprises: extracting a to-be-recognized dark light image captured in a dark light environment; inputting the dark light image into a pre-trained first convolutional neural network to obtain a target image after the dark light image is preprocessed, the first convolutional neural network being used to preprocess the dark light image; and inputting the target image into a pre-trained second convolutional neural network to obtain a facial recognition result, the second convolutional neural network being used to represent a corresponding relationship between the image and the facial recognition result. This embodiment improves accuracy of the facial recognition on the image captured in the dark light environment.

traducir

INFERRING PETROPHYSICAL PROPERTIES OF HYDROCARBON RESERVOIRS USING A NEURAL NETWORK

NºPublicación: US2019087939A1 21/03/2019

Solicitante:

SAUDI ARABIAN OIL CO [SA]

Resumen de: US2019087939A1

Received image data is enhanced to create enhanced image data using image processing to remove artifacts and to retrieve information associated with a desired target output. Image segmentation is performed on useable enhanced image data to created segmented image data by partitioning the enhanced image data into coherent regions with respect to a particular image-based criterion. Useable segmented image data and auxiliary data is pre-processing for input into a neural network as pre-processed data. The pre-processed data is divided into training, validation, and testing data subsets. A neural network architecture is determined to process the pre-processed data and the determined neural network architecture is executed using the pre-processed data. Output of the determined neural network is post-processed as post-processed data. The post-processed data is compared to a known value range associated with the post-processed data to determine if the post-processed data satisfies a desired output result.

traducir

METHOD AND APPARATUS FOR OUTPUTTING INFORMATION

NºPublicación: US2019087683A1 21/03/2019

Solicitante:

BAIDU ONLINE NETWORK TECHNOLOGY BEIJING CO LTD [CN]

CN_107679466_A

Resumen de: US2019087683A1

Embodiments of the present disclosure disclose a method and apparatus for outputting information. The method comprises: acquiring a first image and a second image, the first image including a first face image region, and the second image including a second face image region; generating an image matrix of the first image and an image matrix of the second image; inputting respectively the image matrix of the first image and the image matrix of the second image into a pre-trained convolutional neural network to obtain a characteristic vector of the first image and a characteristic vector of the second image; calculating a distance between the characteristic vector of the first image and the characteristic vector of the second image; and outputting, based on the calculated distance, information of an object relationship between an object the first face image region belongs and an object the second face image region belongs.

traducir

METHODS AND PROCESSES OF ENCRYPTED DEEP LEARNING SERVICES

NºPublicación: US2019087689A1 21/03/2019

Solicitante:

NOVUMIND LTD [KY]

Resumen de: US2019087689A1

A computer system may provide Encrypted Deep Learning Service (EDLS) to a client. The computer system includes one or more processors and memory storing instructions. When instructions are executed by the one or more processors, the instructions cause the computer system to perform acts including: receiving training data from the client, where the training data comprise cipher images that are encrypted using an orthogonal transformation that hides sensitive information in original images. The acts further include training a deep neural network using the training data in the computer system; and producing cipher inference using the deep neural network when the computer system receives new data including new images encrypted using the orthogonal transformation.

traducir

METHOD FOR LOCATING ONE OR MORE CANDIDATE DIGITAL IMAGES BEING LIKELY CANDIDATES FOR DEPICTING AN OBJECT

NºPublicación: EP3457324A1 20/03/2019

Solicitante:

AXIS AB [SE]

Resumen de: EP3457324A1

The disclosure relates to a method for finding one or more candidate digital images being likely candidates for depicting a specific object, the method comprising: receiving (S102) an object digital image (205) depicting the specific object; determining (S106), using a classification subnet (220) of a convolutional neural network (210), a class for the specific object depicted in the object digital image (205); selecting (S108), based on the determined class for the specific object depicted in the object digital image, a feature vector generating subnet from a plurality of feature vector generating subnets (230a,230b,230c); determining (S110), by the selected feature vector generating subnet, a feature vector of the specific object depicted in the object digital image (205); locating (S110) one or more candidate digital images being likely candidates for depicting the specific object depicted in the object digital image (205) by comparing the determined feature vector and feature vectors (242) registered in a database (240), wherein each registered feature vector (242) is associated with a digital image.

traducir

NEURAL NETWORK ACCELERATOR INCLUDING BIDIRECTIONAL PROCESSING ELEMENT ARRAY

NºPublicación: US2019079801A1 14/03/2019

Solicitante:

ELECTRONICS & TELECOMMUNICATIONS RES INST [KR]

Resumen de: US2019079801A1

Provided is a neural network accelerator which performs a calculation of a neural network provided with layers, the neural network accelerator including a kernel memory configured to store kernel data related to a filter, a feature map memory configured to store feature map data which are outputs of the layers, and a Processing Element (PE) array including PEs arranged along first and second directions, wherein each of the PEs performs a calculation using the feature map data transmitted in the first direction from the feature map memory and the kernel data transmitted in the second direction from the kernel memory, and transmits a calculation result to the feature map memory in a third direction opposite to the first direction.

traducir

TRAINING AND TESTING OF A NEURAL NETWORK SYSTEM FOR DEEP ODOMETRY ASSISTED BY STATIC SCENE OPTICAL FLOW

NºPublicación: US2019079536A1 14/03/2019

Solicitante:

TUSIMPLE [US]

Resumen de: US2019079536A1

A system for visual odometry is provided. The system includes: an internet server, comprising: an I/O port, configured to transmit and receive electrical signals to and from a client device; a memory; one or more processing units; and one or more programs stored in the memory and configured for execution by the one or more processing units, the one or more programs including instructions for: in response to images in pairs, generating a prediction of static scene optical flow for each pair of the images in a visual odometry model; generating a set of motion parameters for each pair of the images in the visual odometry model; training the visual odometry model by using the prediction of static scene optical flow and the motion parameters; and predicting motion between a pair of consecutive image frames by the trained visual odometry model.

traducir

TRAINING AND TESTING OF A NEURAL NETWORK METHOD FOR DEEP ODOMETRY ASSISTED BY STATIC SCENE OPTICAL FLOW

NºPublicación: US2019079535A1 14/03/2019

Solicitante:

TUSIMPLE [US]

Resumen de: US2019079535A1

A method of visual odometry for a non-transitory computer readable storage medium storing one or more programs is disclosed. The one or more programs include instructions, which when executed by a computing device, causes the computing device to perform the following steps comprising: in response to images in pairs, generating a prediction of static scene optical flow for each pair of the images in a visual odometry model; generating a set of motion parameters for each pair of the images in the visual odometry model; training the visual odometry model by using the prediction of static scene optical flow and the motion parameters; and predicting motion between a pair of consecutive image frames by the trained visual odometry model.

traducir

CONVOLUTIONAL NEURAL NETWORK ARCHITECTURE WITH ADAPTIVE FILTERS

NºPublicación: US2019079915A1 14/03/2019

Solicitante:

NEC LAB AMERICA INC [US]

US_2019079999_A1

Resumen de: US2019079915A1

A computer-implemented method for employing input-conditioned filters to perform natural language processing tasks using a convolutional neural network architecture includes receiving one or more inputs, generating one or more sets of filters conditioned on respective ones of the one or more inputs by implementing one or more encoders to encode the one or more inputs into one or more respective hidden vectors, and implementing one or more decoders to determine the one or more sets of filters based on the one or more hidden vectors, and performing adaptive convolution by applying the one or more sets of filters to respective ones of the one or more inputs to generate one or more representations.

traducir

Parallel Neural Processor for Artificial Intelligence

NºPublicación: US2019080227A1 14/03/2019

Solicitante:

SETH ROHIT [CA]

US_2019080228_A1

Resumen de: US2019080227A1

Systems and/or devices for efficient and intuitive methods for implementing artificial neural networks specifically designed for parallel AI processing are provided herein. In various implementations, the disclosed systems, devices, and methods complement or replace conventional systems, devices, and methods for parallel neural processing that (a) greatly reduce neural processing time necessary to process more complex problem sets; (b) implement neuroplasticity necessary for self-learning; and (c) introduce the concept and application of implicit memory, in addition to explicit memory, necessary to imbue an element of intuition. With these properties, implementations of the disclosed invention make it possible to emulate human consciousness or awareness.

traducir

METHOD AND SYSTEM FOR PROVIDING A HIGHLY-PERSONALIZED RECOMMENDATION ENGINE

NºPublicación: US2019080012A1 14/03/2019

Solicitante:

HUANG YU [CN]
CHEN FANG [US]

Resumen de: US2019080012A1

Various embodiments of a deep learning (DL)-based face perception engine for constructing, providing, and applying a highly-personalized face perception model for an individual through a deep learning process are disclosed. In some embodiments, a disclosed face perception engine includes a deep neural network configured for training a personalized face perception model for a unique individual based on a standard set of training images and a corresponding set of decisions on the set of training images provided by the unique individual. When sufficiently trained using the standard set of training images and the corresponding set of decisions, the personalized face perception model for the unique individual perceives a new face photo/image as if through the eyes of that unique individual. Hence, the trained face perception model can be used an “agent” or “representative” of the associated person in making very personal decisions, such as to decide if a given face photo/image includes a desirable face in the eyes of that person.

traducir

METHOD AND APPARATUS FOR MACHINE LEARNING

NºPublicación: US2019080236A1 14/03/2019

Solicitante:

FUJITSU LTD [JP]

Resumen de: US2019080236A1

A machine learning apparatus calculates second values, based on first values each assigned to one variable value of each term in relation to a neuron in a layer following an input layer of a neural network, where the second values are assigned to variable-value combination patterns in relation to each following-layer neuron. Each second value is represented by a product of first values each assigned to a variable value included in the combination pattern in relation to the following-layer neuron. The apparatus then applies the second values as weights each to a numerical value when it is entered to the corresponding following-layer neuron, to calculate an output value of the neural network with the numerical values arranged in an input order. The apparatus updates reference values in a reference pattern and the first values based on input error that the output value exhibits with respect to training data.

traducir

BRIDGE IMPACT DETECTION AND CLASSIFICATION SYSTEMS AND METHODS

NºPublicación: US2019080237A1 14/03/2019

Solicitante:

SENSR MONITORING TECH LLC [US]
SOUTHERN METHODIST UNIV [US]

Resumen de: US2019080237A1

A method for classifying a response signal of acceleration data of a structure includes obtaining at least one signal feature of a response signal, inputting the at least one signal feature into an artificial neural network, and classifying, using the artificial neural network, the response signal as an impact event or a non-impact event. One or more signal features may be used, including a response length feature, a number of peaks feature, a spectral energy feature, a dominant frequency feature, a maximum response feature, a center of mass feature, a slope feature, an average peak power feature, a response symmetry feature, or combinations thereof. One or more artificial neural networks may be used. The artificial neural networks may be trained using different combinations of signal features.

traducir

METHOD AND APPARATUS FOR MACHINE LEARNING

NºPublicación: US2019080235A1 14/03/2019

Solicitante:

FUJITSU LTD [JP]

Resumen de: US2019080235A1

A machine learning apparatus generates a reference pattern including an array of reference values to provide a criterion for ordering numerical values to be entered to a neural network. The reference values correspond one-to-one to combination patterns of variable values of terms among a first term group and combination patterns of variable values of terms among a second term group. Next the machine learning apparatus calculates numerical input values corresponding one-to-one to the combination patterns of variable values of the terms among the first term group and the combination patterns of variable values of the terms among the second term group. Then the machine learning apparatus determines an input order of the numerical input values based on the reference pattern, calculates an output value of the neural network, calculates an input error, and updates the reference pattern based on the input error.

traducir

ELECTRONIC MESSAGE CLASSIFICATION AND DELIVERY USING A NEURAL NETWORK ARCHITECTURE

Nº publicación: US2019079999A1 14/03/2019

Solicitante:

NEC LAB AMERICA INC [US]

US_2019079915_A1

Resumen de: US2019079999A1

A system for electronic message classification and delivery using a neural network architecture includes one or more computing devices associated with one or more users, and at least one computer processing system in communication with one or more computing devices over at least one network. The at least one computer processing system includes at least one processor operatively coupled to a memory device and configured to execute program code stored on the memory device to receive one or more inputs associated with one or more e-mails corresponding to the one or more users across the at least one network, classify the one or more e-mails by performing natural language processing based on one or more sets of filters conditioned on respective ones of the one or more inputs, and permit the one or more users access to the one or more classified e-mails via the one or more computing devices.

traducir

Página1 de 7 nextPage por página

punteroimgVolver