REDES NEURONALES

Volver

Resultados 173 resultados LastUpdate Última actualización 13/10/2019 [19:32:00] pdf PDF xls XLS




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



Página1 de 7 nextPage   Mostrar por página


ENHANCED CONVOLUTIONAL NEURAL NETWORK FOR IMAGE SEGMENTATION

NºPublicación: WO2019194865A1 10/10/2019

Solicitante:
12 SIGMA TECH [US]

Resumen de: WO2019194865A1

This disclosure relates to digital image segmentation and region of interest identification. A computer implemented image segmentation method and system are particularly disclosed, including a predictive model trained based on a deep fully convolutional neural network. The model is trained using a loss function in at least one intermediate layer in addition to a loss function at the final stage of the full convolutional neural network. The predictive segmentation model trained in such a manner requires less training parameters and facilitates quicker and more accurate identification of relevant local and global features in the input image. In one implementation, the fully convolutional neural network is further supplemented with a conditional adversarial neural networks iteratively trained with the fully convolutional neural network as a discriminator measuring the quality of the predictive model generated by the fully convolutional neural network.



traducir


 

APPARATUS AND METHOD FOR APPLYING ARTIFICIAL NEURAL NETWORK TO IMAGE ENCODING OR DECODING

NºPublicación: WO2019194425A1 10/10/2019

Solicitante:
SK TELECOM CO LTD [KR]

Resumen de: WO2019194425A1

The present disclosure relates to image encoding or decoding and, more specifically, to an apparatus and a method for applying an artificial neural network (ANN) to image encoding or decoding. The apparatus and the method of the present disclosure are characterized by applying a CNN-based filter coefficient to a first picture and at least one of a quantization parameter map and a block division map to output a second picture.



traducir


 

METHOD FOR RANDOM SAMPLED CONVOLUTIONS WITH LOW COST ENHANCED EXPRESSIVE POWER

NºPublicación: US2019311248A1 10/10/2019

Solicitante:
INTEL CORP [US]

Resumen de: US2019311248A1

A system and method for random sampled convolutions are disclosed to efficiently boost a convolutional neural network (CNN) expressive power without adding computation cost. The method for random sampled convolutions selects a receptive field size and generates filters with a subset of the receptive field elements, the number of learnable parameters, as being active, wherein the number learnable parameters corresponds to computing characteristics, such as SIMD capability, of the processing system upon which the CNN is executed. Several random filters may be generated, with each being run separately on the CNN. The random filter that causes the fastest convergence is selected over the others. The placement of the random filter in the CNN may be per layer, per channel, or per convergence operation. The CNN employing the random sampled convolutions method performs as well as other CNNs utilizing the same receptive field size.



traducir


 

END-TO-END TEXT-TO-SPEECH CONVERSION

NºPublicación: US2019311708A1 10/10/2019

Solicitante:
GOOGLE LLC [US]

Resumen de: US2019311708A1

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating speech from text. One of the systems includes one or more computers and one or more storage devices storing instructions that when executed by one or more computers cause the one or more computers to implement: a sequence-to-sequence recurrent neural network configured to: receive a sequence of characters in a particular natural language, and process the sequence of characters to generate a spectrogram of a verbal utterance of the sequence of characters in the particular natural language; and a subsystem configured to: receive the sequence of characters in the particular natural language, and provide the sequence of characters as input to the sequence-to-sequence recurrent neural network to obtain as output the spectrogram of the verbal utterance of the sequence of characters in the particular natural language.



traducir


 

Face emotion recognition method based on dual-stream convolutional neural network

NºPublicación: US2019311188A1 10/10/2019

Solicitante:
UNIV SICHUAN [CN]

Resumen de: US2019311188A1

A face emotion recognition method based on dual-stream convolutional neural network uses a multi-scale face expression recognition network to single frame face images and face sequences to perform learning classification. The method includes constructing a multi-scale face expression recognition network which includes a channel network with a resolution of 224×224 and a channel network with a resolution of 336×336, extracting facial expression characteristics at different resolutions through the recognition network, effectively combining static characteristics of images and dynamic characteristics of expression sequence to perform training and learning, fusing the two channel models, testing and obtaining a classification effect of facial expressions. The present invention fully utilizes the advantages of deep learning, effectively avoids the problems of manual extraction of feature deviations and long time, and makes the method provided by the present invention more adaptable. Moreover, the present invention improves the accuracy and productivity of expression recognition.



traducir


 

Low- And High-Fidelity Classifiers Applied To Road-Scene Images

NºPublicación: US2019311221A1 10/10/2019

Solicitante:
FORD GLOBAL TECH LLC [US]

Resumen de: US2019311221A1

Disclosures herein teach applying a set of sections spanning a down-sampled version of an image of a road-scene to a low-fidelity classifier to determine a set of candidate sections for depicting one or more objects in a set of classes. The set of candidate sections of the down-sampled version may be mapped to a set of potential sectors in a high-fidelity version of the image. A high-fidelity classifier may be used to vet the set of potential sectors, determining the presence of one or more objects from the set of classes. The low-fidelity classifier may include a first Convolution Neural Network (CNN) trained on a first training set of down-sampled versions of cropped images of objects in the set of classes. Similarly, the high-fidelity classifier may include a second CNN trained on a second training set of high-fidelity versions of cropped images of objects in the set of classes.



traducir


 

CHARACTER RECOGNITION USING HIERARCHICAL CLASSIFICATION

NºPublicación: US2019311194A1 10/10/2019

Solicitante:
ABBYY PRODUCTION LLC [RU]

Resumen de: US2019311194A1

Aspects of the disclosure provide for mechanisms for character recognition using neural networks. A method of the disclosure includes assigning, using a first-level classifier of a grapheme classifier, an input grapheme image to a first grapheme cluster of a plurality of grapheme clusters, wherein the first grapheme cluster comprises a first plurality of graphemes; selecting, by a processing device, a classifier from a plurality of second-level classifiers of the grapheme classifier based on the first grapheme cluster, wherein the selected classifier is trained to recognize the first plurality of graphemes; and processing the input grapheme image using the selected classifier to recognize at least one character in the input grapheme image.



traducir


 

SYSTEM AND METHOD FOR USING A DATA GENOME TO IDENTIFY SUSPICIOUS FINANCIAL TRANSACTIONS

NºPublicación: US2019311367A1 10/10/2019

Solicitante:
QUANTIPLY CORP [US]

Resumen de: US2019311367A1

A system and method for using a data genome to identify suspicious financial transactions. In one embodiment, the method comprises receiving a data set of financial activity data of multiple participants; configuring a deep neural network and thresholds, wherein the thresholds enable detection of what is within abnormal range of financial activity, patterns, and behavior over a period of time; converting the data set to a genome containing a node for each participant among the multiple participants; computing threat vectors for each node within a graphical representation of the genome that represents behavioral patterns of participants in financial activities, including determining when a key risk indicator (KRI) value computed for a particular threshold within the data set falls outside of a dynamically determined range bounded by thresholds, wherein the threat vectors automatically identify one or more of suspicious participants and suspicious activities in a provided financial activity pattern; and determining a particular edge in the network whose behavior falls outside the dynamically determined range associated with normal activity as a suspicious.



traducir


 

METHOD AND DEVICE FOR IDENTIFYING PATHOLOGICAL PICTURE

NºPublicación: US2019311479A1 10/10/2019

Solicitante:
SUN YAT SEN UNIV CANCER CENTER [CN]

Resumen de: US2019311479A1

The invention discloses a method and a device for identifying pathological pictures, wherein the method comprises: obtaining sample data including a positive sample that is a pathological picture of malignant lesion and a negative sample that is a picture of normal tissue or a pathological picture of benign lesion, with a lesion area marked on the pathological picture of a malignant lesion; dividing the sample data into a training set and a testing set; training a deep neural network model using the training set; testing a trained deep neural network model using the testing set; adjusting parameters of the trained deep neural network model according to a testing result; identifying the pathological picture using the trained deep neural network model. The invention can improve the efficiency and accuracy of pathological picture identification.



traducir


 

DETERMINING STRUCTURE AND MOTION IN IMAGES USING NEURAL NETWORKS

NºPublicación: EP3549102A1 09/10/2019

Solicitante:
GOOGLE LLC [US]

Resumen de: WO2018102717A1

A system comprising an encoder neural network, a scene structure decoder neural network, and a motion decoder neural network. The encoder neural network is configured to: receive a first image and a second image; and process the first image and the second image to generate an encoded representation of the first image and the second image. The scene structure decoder neural network is configured to process the encoded representation to generate a structure output characterizing a structure of a scene depicted in the first image. The motion decoder neural network configured to process the encoded representation to generate a motion output characterizing motion between the first image and the second image.



traducir


 

L2 CONSTRAINED SOFTMAX LOSS FOR DISCRIMINATIVE FACE VERIFICATION

NºPublicación: US2019303754A1 03/10/2019

Solicitante:
UNIV MARYLAND [US]

Resumen de: US2019303754A1

Various face discrimination systems may benefit from techniques for providing increased accuracy. For example, certain discriminative face verification systems can benefit from L2-constrained softmax loss. A method can include applying an image of a face as an input to a deep convolutional neural network. The method can also include applying an output of a fully connected layer of the deep convolutional neural network to an L2-normalizing layer. The method can further include determining softmax loss based on an output of the L2-normalizing layer.



traducir


 

ARTIFICIAL INTELLIGENCE-BASED TEXT-TO-SPEECH SYSTEM AND METHOD

NºPublicación: US2019304435A1 03/10/2019

Solicitante:
TELEPATHY LABS INC [US]

Resumen de: US2019304435A1

A technique proves training and speech quality of a text-to-speech (TTS) system having an artificial intelligence, such as a neural network. The TTS system is organized as a front-end subsystem and a back-end subsystem. The front-end subsystem is configured to provide analysis and conversion of text into input vectors, each having at least a base frequency, f0, a phenome duration, and a phoneme sequence that is processed by a signal generation unit of the back-end subsystem. The signal generation unit includes the neural network interacting with a pre-existing knowledgebase of phenomes to generate audible speech from the input vectors. The technique applies an error signal from the neural network to correct imperfections of the pre-existing knowledgebase of phenomes to generate audible speech signals. A back-end training system is configured to train the signal generation unit by applying psychoacoustic principles to improve quality of the generated audible speech signals.



traducir


 

ARTIFICIAL INTELLIGENCE-BASED TEXT-TO-SPEECH SYSTEM AND METHOD

NºPublicación: US2019304434A1 03/10/2019

Solicitante:
TELEPATHY LABS INC [US]

Resumen de: US2019304434A1

A technique improves training and speech quality of a text-to-speech (TTS) system having an artificial intelligence, such as a neural network. The TTS system is organized as a front-end subsystem and a back-end subsystem. The front-end subsystem is configured to provide analysis and conversion of text into input vectors, each having at least a base frequency, f0, a phenome duration, and a phoneme sequence that is processed by a signal generation unit of the back-end subsystem. The signal generation unit includes the neural network interacting with a pre-existing knowledgebase of phenomes to generate audible speech from the input vectors. The technique applies an error signal from the neural network to correct imperfections of the pre-existing knowledgebase of phenomes to generate audible speech signals. A back-end training system is configured to train the signal generation unit by applying psychoacoustic principles to improve quality of the generated audible speech signals.



traducir


 

TEMPORAL TECHNIQUES OF DENOISING MONTE CARLO RENDERINGS USING NEURAL NETWORKS

NºPublicación: US2019304067A1 03/10/2019

Solicitante:
PIXAR [US]
DISNEY ENTPR INC [US]

Resumen de: US2019304067A1

A modular architecture is provided for denoising Monte Carlo renderings using neural networks. The temporal approach extracts and combines feature representations from neighboring frames rather than building a temporal context using recurrent connections. A multiscale architecture includes separate single-frame or temporal denoising modules for individual scales, and one or more scale compositor neural networks configured to adaptively blend individual scales. An error-predicting module is configured to produce adaptive sampling maps for a renderer to achieve more uniform residual noise distribution. An asymmetric loss function may be used for training the neural networks, which can provide control over the variance-bias trade-off during denoising.



traducir


 

TARGET DETECTION METHOD AND DEVICE, COMPUTING DEVICE AND READABLE STORAGE MEDIUM

NºPublicación: US2019303731A1 03/10/2019

Solicitante:
BOE TECHNOLOGY GROUP CO LTD [CN]

Resumen de: US2019303731A1

The present disclosure relates to a target detection method and device, a computing device and a readable storage medium. The target detection method include performing target detection using a convolutional neural network comprising a plurality of convolutional layers. The method include performing a branch convolutional process on at least one of the convolutional layers to obtain a branch detection result. The method includes performing a fusion process on the branch detection result, or on the branch detection result and a detection result of a last convolutional layer in the convolutional neural network, and transmitting a result of the fusion process to a fully connected layer.



traducir


 

Field Operations Neural Network Heuristics

NºPublicación: US2019302310A1 03/10/2019

Solicitante:
SCHLUMBERGER TECHNOLOGY CORP [US]

Resumen de: US2019302310A1

A method includes representing oilfield operational plan information as pixels where the pixels include pixels that correspond to a plurality of different state variables associated with oilfield operations; training a deep neural network based at least in part on the pixels to generate a trained deep neural network; implementing the trained deep neural network during generation of an oilfield operational plan; and outputting the oilfield operational plan as a digital plan that specifies at least one control action for oilfield equipment.



traducir


 

Using Machine Learning to Optimize Memory Usage

NºPublicación: US2019303176A1 03/10/2019

Solicitante:
QUALCOMM INC [US]

Resumen de: US2019303176A1

Disclosed are methods and apparatuses for optimizing a usage of a memory storing apps. In an aspect, an apparatus receives time data reflecting when each of the apps in the memory was used, receives location data reflecting where each of the apps in the memory was used, receives frequency data reflecting a usage frequency of each of the apps in the memory and trains a neural network to learn an app usage pattern based on the received time data, received location data and received frequency data.



traducir


 

DETECTING DATA ANOMALIES ON A DATA INTERFACE USING MACHINE LEARNING

NºPublicación: US2019303567A1 03/10/2019

Solicitante:
NVIDIA CORP [US]

Resumen de: US2019303567A1

The disclosure provides systems and processes for applying neural networks to detect intrusions and other anomalies in communications exchanged over a data bus between two or more devices in a network. The intrusions may be detected in data being communicated to an embedded system deployed in vehicular or robotic platforms. The disclosed system and process are well suited for incorporation into autonomous control or advanced driver assistance system (ADAS) vehicles including, without limitation, automobiles, motorcycles, boats, planes, and manned and un-manned robotic devices. Data communicated to an embedded system can be detected over any of a variety of data buses. In particular, embodiments disclosed herein are well suited for use in any data communication interface exhibiting the characteristics of a lack of authentication or following a broadcast routing scheme—including, without limitation, a control area network (CAN) bus.



traducir


 

INTERPRETABLE BIO-MEDICAL LINK PREDICTION USING DEEP NEURAL REPRESENTATION

NºPublicación: US2019303535A1 03/10/2019

Solicitante:
IBM [US]

Resumen de: US2019303535A1

Link prediction for biomedical entities. A neural network is trained using known associations between biomedical entities, including their vector representations and additional information-carrying content describing the biomedical entities. The trained network infers or predicts unobserved associations between two entities.



traducir


 

COMBINING CONVOLUTION AND DECONVOLUTION FOR OBJECT DETECTION

NºPublicación: US2019303715A1 03/10/2019

Solicitante:
QUALCOMM INC [US]

Resumen de: US2019303715A1

Provided are systems, methods, and computer-readable medium for operating a neural network. In various implementations, the neural network can receive an input image that includes an object to be identified. The neural network can generate a plurality of initial feature maps using a convolution layers, wherein a first initial feature maps is generated using the input image. The neural network can generate an up-sampled feature map using a de-convolution layer that takes an initial feature map as input, where the up-sampled feature map has a same resolution as the previous initial feature map. The neural network can combine the up-sampled feature map and the previous initial feature map, and use the combined feature map to more accurate identify the object.



traducir


 

Method of Neural Network Training Using Floating-Point Signed Digit Representation

NºPublicación: US2019303756A1 03/10/2019

Solicitante:
UNIV NAT TAIWAN [TW]

Resumen de: US2019303756A1

A method of training a neural network including multiple neural network weights and multiple neurons, and the method includes using floating-point signed digit numbers to represent each of the multiple neural network weights, wherein a mantissa of each of the multiple neural network weights is represented by multiple mantissa signed digit groups and an exponent of each of the multiple neural network weights is represented by an exponent digit group; and using the exponent digit group and at least one of the multiple mantissa signed digit groups to perform weight adjustment computation and neural network inference computation.



traducir


 

DEVICE PLACEMENT OPTIMIZATION WITH REINFORCEMENT LEARNING

NºPublicación: US2019303761A1 03/10/2019

Solicitante:
GOOGLE LLC [US]

Resumen de: US2019303761A1

A method for determining a placement for machine learning model operations across multiple hardware devices is described. The method includes receiving data specifying a machine learning model to be placed for distributed processing on multiple hardware devices; generating, from the data, a sequence of operation embeddings, each operation embedding in the sequence characterizing respective operations necessary to perform the processing of the machine learning model; processing the sequence of operation embeddings using a placement recurrent neural network in accordance with first values of a plurality network parameters of the placement recurrent neural network to generate a network output that defines a placement of the operations characterized by the operation embeddings in the sequence across the plurality of devices; and scheduling the machine learning model for processing by the multiple hardware devices by placing the operations on the multiple devices according to the placement defined by the network output.



traducir


 

Dataset completion

NºPublicación: US2019303774A1 03/10/2019

Solicitante:
CELECT INC [US]

Resumen de: US2019303774A1

Systems and methods for completing at least one entry in a dataset. The systems and methods described herein find a rich set of dense features for each of a plurality of entities, then combine the rich set of dense features with externally provided features to estimate a target value. The systems and methods combine these features using a neural network model in which portions of the input layer are each only connected to a portion of the hidden layer.



traducir


 

NEURAL NETWORK TRAINING SYSTEM

NºPublicación: US2019303725A1 03/10/2019

Solicitante:
FRINGEFY LTD [IL]

Resumen de: US2019303725A1

In order for the feature extractors to operate with sufficient accuracy, a high degree of training is required. In this situation, a neural network implementing the feature extractor may be trained by providing it with images having known correspondence. A 3D model of a city may be utilized in order to train a neural network for location detection. 3D models are sophisticated and allow manipulation of viewer perspective and ambient features such as day/night sky variations, weather variations, and occlusion placement. Various manipulations may be executed in order to generate vast numbers of image pairs having known correspondence despite having variations. These image pairs with known correspondence may be utilized to train the neural network to be able to generate feature maps from query images and identify correspondence between query image feature maps and reference feature maps. This training can be accomplished without requiring the capture of real images with known correspondence. Capture of real images with known correspondence is cumbersome, time and resource-intensive, and difficult to manage.



traducir


 

SYSTEMS AND METHODS FOR AUTOMATIC DETECTION OF AN INDICATION OF ABNORMALITY IN AN ANATOMICAL IMAGE

Nº publicación: US2019304092A1 03/10/2019

Solicitante:
IBM [US]

Resumen de: US2019304092A1

There is provided a method for training a deep convolutional neural network (CNN) for detecting an indication of likelihood of abnormality, comprising: receiving anatomical training images, each including an associated annotation indicative of abnormality for the whole image without an indication of location of the abnormality, executing, for each anatomical training image: decomposing the anatomical training image into patches, computing a feature representation of each patch, computing for each patch, according to the feature representation of the patch, a probability that the patch includes an indication of abnormality, setting a probability indicative of likelihood of abnormality in the anatomical image according to the maximal probability value computed for one patch, and training a deep CNN for detecting an indication of likelihood of abnormality in a target anatomical image according to the patches of the anatomical training images, the one patch, and the probability set for each respective anatomical training image.


traducir



 

Página1 de 7 nextPage Mostrar por página

Volver