REDES NEURONALES

VolverVolver

Resultados 258 resultados LastUpdate Última actualización 28/05/2020 [07:03:00] pdf PDF xls XLS

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



Página1 de 11 nextPage   por página


METHOD FOR WARNING VEHICLE OF RISK OF LANE CHANGE AND ALARM DEVICE USING THE SAME

NºPublicación: EP3657382A1 27/05/2020

Solicitante:

STRADVISION INC [KR]

JP_2020077393_A

Resumen de: EP3657382A1

A method for warning a vehicle of a risk of lane change is provided. The method includes steps of: (a) an alarm device, if at least one rear image captured by a running vehicle is acquired, segmenting the rear image by using a learned convolutional neural network (CNN) to thereby obtain a segmentation image corresponding to the rear image; (b) the alarm device checking at least one free space ratio in at least one blind spot by referring to the segmentation image, wherein the free space ratio is determined as a ratio of a road area without an object in the blind spot to a whole area of the blind spot; and (c) the alarm device, if the free space ratio is less than or equal to at least one predetermined threshold value, warning a driver of the vehicle of the risk of lane change.

traducir

DEEP LEARNING-BASED IMAGE ANALYSIS METHOD, SYSTEM, AND PORTABLE TERMINAL

NºPublicación: WO2020101121A1 22/05/2020

Solicitante:

TUAT CO LTD [KR]

Resumen de: WO2020101121A1

The present invention relates to a deep learning-based image analysis method, system, and portable terminal. A deep learning-based image analysis system according to an embodiment of the present invention comprises: an image analysis portable terminal for capturing and transmitting an image; and an image analysis server for receiving the captured image to generate analysis result values through multiple function-specific neural networks, and applying weighted priorities to probability values of the generated analysis result values to provide same to the image analysis portable terminal.

traducir

IMAGE PROCESSING APPARATUS AND METHOD OF OPERATING THE SAME

NºPublicación: WO2020101143A1 22/05/2020

Solicitante:

SAMSUNG ELECTRONICS CO LTD [KR]

US_2020160494_A1

Resumen de: WO2020101143A1

An image processing apparatus processes an image by using one or more neural networks, and includes a memory storing one or more instructions and data structures for a main neural network and a sub-neural network, and a processor configured to execute the one or more instructions stored in the memory to process an input image by using the main neural network to obtain intermediate result data and a final output image, and to process the intermediate result data by using the sub-neural network to output an intermediate image while the input image is being processed by using the main neural network.

traducir

IMAGE CLASSIFICATION, GENERATION AND APPLICATION OF NEURAL NETWORKS

NºPublicación: WO2020099854A1 22/05/2020

Solicitante:

RPPTV LTD [GB]

Resumen de: WO2020099854A1

The present invention relates to generating and training neural networks for application in numerous fields including object or event recognition in images including visual sequences. There is provided a computerised method of generating a neural network (NN), comprising generating successive candidate NNs (310, 545) using an optimisation algorithm (530), each NN having a number of connected blocks of layers (205), the layers having a plurality of neurons with connections having associated weights. Each block comprises fixed and variable architectural parameters (210X, 210Y), the or each variable architectural parameter being determined by an optimisation algorithm. Each candidate NN is trained using training data in order to update the weights of the candidate NN (320), and a fitness function score of the trained candidate NN is determined using validation data (330). If there is a block having the same architectural parameters from a previously trained candidate NN, the weights associated with layers of said block are inherited prior to training and fitness score determination (425).

traducir

KEY POINT DETECTION METHOD AND APPARATUS, ELECTRONIC DEVICE AND STORAGE MEDIUM

NºPublicación: WO2020098225A1 22/05/2020

Solicitante:

BEIJING SENSETIME TECH DEVELOPMENT CO LTD [CN]

CN_109614876_A

Resumen de: WO2020098225A1

Embodiments of the present disclosure relate to a key point detection method and apparatus, an electronic device and a storage medium, the method comprising: obtaining first characteristic images having multiple scales for an input image, the scale of each first characteristic image having a multiple relationship; performing forward processing on each first characteristic image by using a first pyramid neural network to obtain a second characteristic image corresponding one-to-one to each first characteristic image, wherein the second characteristic image and the first characteristic images to which the second characteristic image corresponds one-to-one have the same scale; performing backward processing on each second characteristic image by using a second pyramid neural network to obtain a third characteristic image corresponding one-to-one to each second characteristic image, wherein the third characteristic image and the second characteristic images to which the third characteristic image corresponds one-to-one have the same scale; fusing the characteristics of each third characteristic image, and obtaining the position of each key point in the input image by using the characteristic image after characteristic fusion. In the present disclosure, the position of a key point may be precisely extracted.

traducir

LEARNING TO GENERATE SYNTHETIC DATASETS FOR TRAINING NEURAL NETWORKS

NºPublicación: WO2020102733A1 22/05/2020

Solicitante:

NVIDIA CORP [US]

US_2020160178_A1

Resumen de: WO2020102733A1

In various examples, a generative model is used to synthesize datasets for use in training a downstream machine learning model to perform an associated task. The synthesized datasets may be generated by sampling a scene graph from a scene grammar - such as a probabilistic grammar - and applying the scene graph to the generative model to compute updated scene graphs more representative of object attribute distributions of real-world datasets. The downstream machine learning model may be validated against a real-world validation dataset, and the performance of the model on the real-world validation dataset may be used as an additional factor in further training or fine-tuning the generative model for generating the synthesized datasets specific to the task of the downstream machine learning model.

traducir

METHOD OF ANOMALY DETECTION AND SYSTEM THEREOF

NºPublicación: WO2020100136A1 22/05/2020

Solicitante:

UVEYE LTD [IL]

Resumen de: WO2020100136A1

There are provided a system and method of training a neural network system for anomaly detection, comprising: obtaining a training dataset including a set of original images and a set of random data vectors; constructing a neural network system comprising a generator, and a first discriminator and a second discriminator operatively connected to the generator; training the generator, the first discriminator and the second discriminator together based on the training dataset, such that: i) the generator is trained, at least based on evaluation of the first discriminator, to generate synthetic images meeting a criterion of photo-realism as compared to corresponding original images; and ii) the second discriminator is trained based on the original images and the synthetic images to discriminate images with anomaly from images without anomaly with a given level of accuracy, thereby giving rise to a trained neural network system.

traducir

COMPUTER, NEURAL NETWORK CONSTRUCTION METHOD, AND COMPUTER SYSTEM

NºPublicación: US2020160118A1 21/05/2020

Solicitante:

HITACHI LTD [JP]

Resumen de: US2020160118A1

A computer constructs a neural network for executing image processing, the neural network being constituted of layers, each of which includes at least one node. The neural network includes a detection layer that realizes a process for detecting an object in an image. The computer is configured to execute: a first process of obtaining setting information for constructing the neural network including setting values relating to characteristics of a boundary of the object and a shape of the object, the setting values being values for calculating hyperparameters of the detection layer; a second process of constructing the neural network on the basis of the setting information. The second process includes a process of calculating the hyperparameters of the detection layer on the basis of the setting values.

traducir

DISTRIBUTED BATCH NORMALIZATION USING ESTIMATES AND ROLLBACK

NºPublicación: US2020160123A1 21/05/2020

Solicitante:

NVIDIA CORP [US]

US_2020160112_A1

Resumen de: US2020160123A1

A technique utilizing speculative execution and rollback for performing data parallel training of a neural network model is disclosed. Activations for a layer of the neural network model are normalized during a speculative normalization operation using estimated normalization parameters associated with a partial population of a set of training data allocated to a particular processor. Normalization parameters associated with the total population of the set of training data are generated by a distributed reduce operation in parallel with the speculative normalization operation. An optional rollback operation can revert the activations to a pre-normalization state if the estimated normalization parameters for the partial population are subsequently determined to be inaccurate compared to the normalization parameters for the population of the set of training data distributed across a plurality of processors.

traducir

LEARNING COPY SPACE USING REGRESSION AND SEGMENTATION NEURAL NETWORKS

NºPublicación: US2020160111A1 21/05/2020

Solicitante:

ADOBE INC [US]

Resumen de: US2020160111A1

Techniques are disclosed for characterizing and defining the location of a copy space in an image. A methodology implementing the techniques according to an embodiment includes applying a regression convolutional neural network (CNN) to an image. The regression CNN is configured to predict properties of the copy space such as size and type (natural or manufactured). The prediction is conditioned on a determination of the presence of the copy space in the image. The method further includes applying a segmentation CNN to the image. The segmentation CNN is configured to generate one or more pixel-level masks to define the location of copy spaces in the image, whether natural or manufactured, or to define the location of a background region of the image. The segmentation CNN may include a first stage comprising convolutional layers and a second stage comprising pairs of boundary refinement layers and bilinear up-sampling layers.

traducir

DISTRIBUTED BATCH NORMALIZATION USING PARTIAL POPULATIONS

NºPublicación: US2020160112A1 21/05/2020

Solicitante:

NVIDIA CORP [US]

US_2020160123_A1

Resumen de: US2020160112A1

A technique for performing data parallel training of a neural network model is disclosed that incorporates batch normalization techniques using partial populations to generate normalization parameters. The technique involves processing, by each processor of a plurality of processors in parallel, a first portion of a sub-batch of training samples allocated to the processor to generate activations for the first portion of the sub-batch. Each processor analyzes the activations and transmits statistical measures for the first portion to an additional processor that reduces the statistical measures from multiple processors to generate normalization parameters for a partial population of the training samples that includes the first portion from each of the plurality of processors. The normalization parameters are then transmitted back to each of the processors to normalize the activations for both the first portion and a second portion of the sub-batch of training samples allocated to each processor.

traducir

METHOD AND APPARATUS FOR ALIGNING 3D MODEL

NºPublicación: US2020160616A1 21/05/2020

Solicitante:

SAMSUNG ELECTRONICS CO LTD [KR]

Resumen de: US2020160616A1

Provided is a method and apparatus for aligning a three-dimensional (3D) model. The 3D model alignment method includes acquiring, by a processor, at least one two-dimensional (2D) image including an object, detecting, by the processor, a feature point of the object in the at least one 2D input image using a neural network, estimating, by the processor, a 3D pose of the object in the at least one 2D input image using the neural network, retrieving, by the processor, a target 3D model based on the estimated 3D pose, and aligning, by the processor, the target 3D model and the object based on the feature point.

traducir

Deep Learning Based Reservoir Modeling

NºPublicación: US2020160173A1 21/05/2020

Solicitante:

LANDMARK GRAPHICS CORP [US]

CA_3067013_A1

Resumen de: US2020160173A1

Embodiments of the subject technology for deep learning based reservoir modelling provides for receiving input data comprising information associated with one or more well logs in a region of interest. The subject technology determines, based at least in part on the input data, an input feature associated with a first deep neural network (DNN) for predicting a value of a property at a location within the region of interest. Further, the subject technology trains, using the input data and based at least in part on the input feature, the first DNN. The subject technology predicts, using the first DNN, the value of the property at the location in the region of interest. The subject technology utilizes a second DNN that classifies facies based on the predicted property in the region of interest.

traducir

SYSTEM AND METHOD FOR DYNAMIC SCHEDULING OF DISTRIBUTED DEEP LEARNING TRAINING JOBS

NºPublicación: US2020159589A1 21/05/2020

Solicitante:

SAMSUNG ELECTRONICS CO LTD [KR]

Resumen de: US2020159589A1

A scheduling algorithm for scheduling training of deep neural network (DNN) weights on processing units identifies a next job to provisionally assign a processing unit (PU) based on a doubling heuristic. The doubling heuristic makes use of an estimated number of training sets needed to complete training of weights for a given job and/or a training speed function which indicates how fast the weights are converging. The scheduling algorithm solves a problem of efficiently assigning PUs when multiple DNN weight data structures must be trained efficiently. In some embodiments, the training of the weights uses a ring-based message passing architecture. In some embodiments, performance using a nested loop approach or nested loop fashion is provided. In inner iterations of the nested loop, PUs are scheduled and jobs are launched or re-started. In outer iterations of the nested loop, jobs are stopped, parameters are updated and the inner iteration is re-entered.

traducir

SYSTEM AND METHOD FOR INCREMENTAL LEARNING THROUGH STATE-BASED REAL-TIME ADAPTATIONS IN NEURAL NETWORKS

NºPublicación: US2020160170A1 21/05/2020

Solicitante:

BANK OF AMERICA [US]

Resumen de: US2020160170A1

An artificial intelligence system and method for state-based learning using one or more adaptive response states of the artificial intelligence system are provided. A controller for modifying a neural network engine is configured to monitor a data stream having a data pattern by comparing the data pattern to a trained data pattern; identify a change in the data pattern of the data stream; determine a response state of the neural network learning engine, the state defining one or more neural network parameters for monitoring the data stream with the neural network learning engine; identify a predetermined policy for reconfiguring the neural network learning engine based on the data pattern and the response state; and in response to identifying the change in the data pattern and determining the response state, reconfigure the one or more neural network parameters according to the predetermined policy.

traducir

TRAINING A NEURAL NETWORK BASED ON TEMPORAL CHANGES IN ANSWERS TO FACTOID QUESTIONS

NºPublicación: US2020160166A1 21/05/2020

Solicitante:

IBM [US]

Resumen de: US2020160166A1

A method trains a neural network to identify an event based on discrepancies in answers to factoid questions at different times. One or more processors identify answers to a series of factoid questions. The processor(s) compare the answers from the series of factoid questions in order to determine discrepancies in the answers at different times, and then train a neural network to identify an event based on the discrepancies in the answers at the different times.

traducir

TECHNIQUES FOR IDENTIFYING SYNCHRONIZATION ERRORS IN MEDIA TITLES

NºPublicación: US2020160889A1 21/05/2020

Solicitante:

NETFLIX INC [US]

Resumen de: US2020160889A1

A neural network system that is trained to identify one or more portions of a media title where synchronization errors are likely to be present. The neural network system is trained based on a first set of media titles where synchronization errors are present and a second set of media titles where synchronization errors are absent. The second set of media titles can be generated by introducing synchronization errors into a set of media titles that otherwise lack synchronization errors. Via training, the neural network system learns to identify specific visual features included in one or more video frames and corresponding audio features that should be played back in synchrony with the associated visual features. Accordingly, when presented with a media title that includes synchronization errors, the neural network can indicate the specific frames where synchronization errors are likely to be present.

traducir

SYSTEMS AND METHODS FOR EMPLOYING PREDICATION IN COMPUTATIONAL MODELS

NºPublicación: US2020160848A1 21/05/2020

Solicitante:

FACEBOOK INC [US]

US_2019206390_PA

Resumen de: US2020160848A1

The disclosed method may include (1) determining whether a next operation of a plurality of operations of an artificial neural network (ANN) is dependent upon a Boolean predication value based on a representative value for a weight or an input of a node of the ANN, (2) based on the next operation not being dependent on the Boolean predication value, allowing the next operation to update a state of the ANN, and (3) based on the next operation being dependent on the Boolean predication value, performing at least one of (a) allowing, based on the Boolean predication value being a first value, the next operation to update the state of the ANN, and (b) preventing, based on the Boolean predication value being a second value different from the first value, the next operation from updating the state of the ANN. Various other methods and systems are also disclosed.

traducir

IMAGE PROCESSING APPARATUS AND CONTROL METHOD THEREOF

NºPublicación: US2020160096A1 21/05/2020

Solicitante:

SAMSUNG ELECTRONICS CO LTD [KR]

EP_3654240_A1

Resumen de: US2020160096A1

Provided are an image processing apparatus and a control method thereof. The image processing apparatus includes: a communication circuitry configured to communicate with an external device; a storage configured to store data; an image processor configured to perform image processing; and a controller configured to perform an operation, through a neural network, on an image frame contained in an image received by the communication circuitry, to determine a type of the image based on information according to the operation through the neural network, and to control the image processor based on the determined type of the image.

traducir

LEARNING TO GENERATE SYNTHETIC DATASETS FOR TRANING NEURAL NETWORKS

NºPublicación: US2020160178A1 21/05/2020

Solicitante:

NVIDIA CORP [US]

Resumen de: US2020160178A1

In various examples, a generative model is used to synthesize datasets for use in training a downstream machine learning model to perform an associated task. The synthesized datasets may be generated by sampling a scene graph from a scene grammar—such as a probabilistic grammar—and applying the scene graph to the generative model to compute updated scene graphs more representative of object attribute distributions of real-world datasets. The downstream machine learning model may be validated against a real-world validation dataset, and the performance of the model on the real-world validation dataset may be used as an additional factor in further training or fine-tuning the generative model for generating the synthesized datasets specific to the task of the downstream machine learning model.

traducir

SYSTEM AND METHOD FOR A CONVOLUTIONAL NEURAL NETWORK FOR MULTI-LABEL CLASSIFICATION WITH PARTIAL ANNOTATIONS

NºPublicación: US2020160177A1 21/05/2020

Solicitante:

ROYAL BANK OF CANADA [CA]

Resumen de: US2020160177A1

Effectively training machine learning systems with incomplete/partial labels is a practical, technical problem that solutions described herein attempt to overcome. In particular, an approach to modify loss functions on a proportionality basis is noted in some embodiments. In other embodiments, a graph neural network is provided to help identify correlations/causations as between categories. In another set of embodiments, a prediction approach is described to, based on originally provided labels, predict labels for unlabelled training samples such that the proportion of labelled labels relative to all labels is increased.

traducir

Road Surface Characterization Using Pose Observations Of Adjacent Vehicles

NºPublicación: US2020160070A1 21/05/2020

Solicitante:

FORD GLOBAL TECH LLC [US]

Resumen de: US2020160070A1

A computing system can crop an image based on a width, height and location of a first vehicle in the image. The computing system can estimate a pose of the first vehicle based on inputting the cropped image and the width, height and location of the first vehicle into a deep neural network. The computing system can then operate a second vehicle based on the estimated pose. The computing system may train a model to identify a type and a location of a hazard according to the estimated pose, the hazard being such things as ice, mud, pothole, or other hazard. The model may be used by an autonomous vehicle to identify and avoid hazards or to provide drive assistance alerts.

traducir

SYSTEMS AND METHOD FOR ACTION RECOGNITION USING MICRO-DOPPLER SIGNATURES AND RECURRENT NEURAL NETWORKS

NºPublicación: US2020160046A1 21/05/2020

Solicitante:

UNIV JOHNS HOPKINS [US]

WO_2019006473_PA

Resumen de: US2020160046A1

The present disclosure may be embodied as systems and methods for action recognition developed using a multimodal dataset that incorporates both visual data, which facilitates the accurate tracking of movement, and active acoustic data, which captures the micro-Doppler modulations induced by the motion. The dataset includes twenty-one actions and focuses on examples of orientational symmetry that a single active ultrasound sensor should have the most difficulty discriminating. The combined results from three independent ultrasound sensors are encouraging, and provide a foundation to explore the use of data from multiple viewpoints to resolve the orientational ambiguity in action recognition. In various embodiments, recurrent neural networks using long short-term memory (LSTM) or hidden Markov models (HMMs) are disclosed for use in action recognition, for example, human action recognition, from micro-Doppler signatures.

traducir

METHOD FOR TRAINING A CONVOLUTIONAL RECURRENT NEURAL NETWORK AND FOR SEMANTIC SEGMENTATION OF INPUTTED VIDEO USING THE TRAINED CONVOLUTIONAL RECURRENT NEURAL NETWORK

NºPublicación: US2020160065A1 21/05/2020

Solicitante:

NAVER CORP [KR]

EP_3608844_PA

Resumen de: US2020160065A1

A method for training a convolutional recurrent neural network for semantic segmentation in videos, includes (a) training, using a set of semantically segmented training images, a first convolutional neural network;(b) training, using a set of semantically segmented training videos, a convolutional recurrent neural network, corresponding to the first convolutional neural network, wherein a convolutional layer has been replaced by a recurrent module having a hidden state. The training of the convolutional recurrent neural network, for each pair of successive frames (t−1, t ∈ 1; T2) of a video of the set of semantically segmented training videos includes warping an internal state of a recurrent layer according to an estimated optical flow between the frames of the pair of successive frames, so as to adapt the internal state to the motion of pixels between the frames of the pair and learning parameters of at least the recurrent module.

traducir

AUTOMATIC SHIP TRACKING METHOD AND SYSTEM BASED ON DEEP LEARNING NETWORK AND MEAN SHIFT

Nº publicación: US2020160061A1 21/05/2020

Solicitante:

ZHUHAI DA HENGQIN TECH DEVELOPMENT CO LTD [CN]

KR_20200006167_A

Resumen de: US2020160061A1

An automatic ship tracking method and system based on deep learning network and mean shift, wherein the method includes: collecting surveillance video data which includes collecting coastal region surveillance video data under visible light and extracting each frame of image; performing preprocessing to extract a positive sample and a negative sample of a ship target; inputting the samples of the ship target in the video into a neural network to train a model by a region-based convolutional neural network method; extracting initial frame data of the video and performing ship detection and probability density calculation on initial moment data according to the trained model; and determining a ship tracking result at the current moment by a calculation result of a previous moment.

traducir

Página1 de 11 nextPage por página

punteroimgVolver