REDES NEURONALES

VolverVolver

Resultados 165 resultados LastUpdate Última actualización 20/10/2021 [05:45:00] pdf PDF xls XLS

Solicitudes publicadas en los últimos 30 días / Applications published in the last 30 days



Página1 de 7 nextPage   por página


VISION-BASED CELL STRUCTURE RECOGNITION USING HIERARCHICAL NEURAL NETWORKS

NºPublicación: US2021319217A1 14/10/2021

Solicitante:

IBM [US]

Resumen de: US2021319217A1

Methods, systems, and computer program products for vision-based cell structure recognition using hierarchical neural networks and cell boundaries to structure clustering are provided herein. A computer-implemented method includes detecting a style of the given table using at least one style classification model; selecting, based at least in part on the detected style, a cell detection model appropriate for the detected style; detecting cells within the given table using the selected cell detection model; and outputting, to at least one user, information pertaining to the detected cells comprising image coordinates of one or more bounding boxes associated with the detected cells.

traducir

Method for segmentation of underground drainage pipeline defects based on full convolutional neural network

NºPublicación: US2021319265A1 14/10/2021

Solicitante:

UNIV ZHENGZHOU [CN]
BESTDR INFRASTRUCTURE HOSPITAL PINGYU [CN]

CN_112258496_A

Resumen de: US2021319265A1

A method for segmentation of underground drainage pipeline defects based on full convolutional neural network includes steps of: collecting a data set of the underground drainage pipeline defects; processing the data set of the underground drainage pipeline defects; optimizing with a semantic segmentation algorithm; adjusting model hyperparameters; training a model; verifying the model; and testing the model. The method adopts a deep learning algorithm, optimizes the FCN full convolutional neural network, develops a semantic segmentation method suitable for complex and similar defect characteristics of underground drainage pipelines, and adopts real underground drainage pipeline defect detection big data, thereby realizing pixel-level segmentation of the underground drainage pipeline defects and providing better robustness and generality. The detection accuracy and efficiency of the underground drainage pipeline defects are effectively improved.

traducir

SYSTEM AND METHOD FOR DETECTING ADVERSARIAL ATTACKS

NºPublicación: WO2021205746A1 14/10/2021

Solicitante:

MITSUBISHI ELECTRIC CORP [JP]

US_2021319784_A1

Resumen de: WO2021205746A1

A linguistic system for transcribing an input, where the linguistic system comprises a processor configured to execute a neural network multiple times while varying weights of at least some nodes of the neural network to produce multiple transcriptions of the input. Further, determine a distribution of pairwise distances of the multiple transcriptions; determine a legitimacy of the input based on the distribution; and transcribe the input using stored weights of the nodes of the neural network when the input is determined as legitimate to produce a final transcription of the input.

traducir

SYSTEM, APPARATUS AND METHOD FOR EVALUATING NEURAL NETWORKS

NºPublicación: WO2021204440A1 14/10/2021

Solicitante:

SIEMENS AG [DE]

EP_3893160_A1

Resumen de: WO2021204440A1

System, Apparatus and Method for Evaluating Neural Networks System, apparatus and method for evaluating neural networks associable with an industrial environment (140) is disclosed. The neural networks are configurable to enable autonomous operations in the industrial environment (140). The method comprising generating explanation (134) of the neural networks, wherein the neural networks generate predictions based on the operations in the industrial environment (140); wherein the explanation (134) comprises at least one of sensitivities, heatmaps and highlights of the predictions generated by the neural networks; generating objective coefficients for the neural networks based on predefined performance indicators of the explanation (134); evaluating the neural networks based on comparison of the objective coefficients; and determining a follow-up operation for the industrial environment (140) based on the evaluation of the neural networks.

traducir

PROJECTING IMAGES CAPTURED USING FISHEYE LENSES FOR FEATURE DETECTION IN AUTONOMOUS MACHINE APPLICATIONS

NºPublicación: WO2021206896A1 14/10/2021

Solicitante:

NVIDIA CORP [US]

US_2021312203_A1

Resumen de: WO2021206896A1

In examples, live perception from wide-view sensors may be leveraged to detect features in an environment of a vehicle. Sensor data generated by the sensors may be adjusted to represent a virtual field of view different from an actual field of view of the sensor, and the sensor data - with or without virtual adjustment - may be applied to a stereographic projection algorithm to generate a projected image. The projected image may then be applied to a machine learning model - such as a deep neural network (DNN) - to detect and/or classify features or objects represented therein. The machine learning model may be pre-trained on training sensor data generated by a sensor having a field of view less than the wide-view sensor such that the virtual adjustment and/or projection algorithm may update the sensor data to be suitable for accurate processing by the pre-trained machine learning model.

traducir

Asymmetric Functionality Activation for Improved Stability in Neural Networks

NºPublicación: US2021319320A1 14/10/2021

Solicitante:

GOOGLE LLC [US]

Resumen de: US2021319320A1

Thus, aspects of the present disclosure address model “blow up” by changing the functionality of the activation, thereby providing “dead” or “dying” neurons with the ability to recover from this situation. As one example, for activation functions that have an input region in which the neuron is turned off by a 0 or close to 0 gradient, a training computing system can keep the neuron turned off when the gradient pushes the unit farther into the region (e.g., by applying an update with zero or reduced magnitude). However, if the gradient for the current training example (or batch) attempts to push the unit towards a region in which the neuron is active again, the system can allow for a non-zero gradient (e.g., by applying an update with standard or increased magnitude).

traducir

META-LEARNING NEURAL ARCHITECTURE SEARCH VIA GRAPH NETWORKS ON SEARCH SPACE LATTICES

NºPublicación: US2021319272A1 14/10/2021

Solicitante:

TOYOTA RES INST INC [US]

Resumen de: US2021319272A1

One or more embodiments of the disclosure include systems and methods that use meta-learning to learn how to optimally find a new neural network architecture for a task using past architectures that were optimized for other tasks, including for example tasks associated with autonomous, semi-autonomous, assisted, or other driving applications. A computer implemented method of the disclosure includes configuring a search space lattice comprising nodes representing operator choices, edges, and a maximum depth. The method includes defining an objective function. The method further includes configuring a graph network over the search space lattice to predict edge weights over the search space lattice. The method also includes alternating optimization between (1) weights of the graph network, to optimize the objective function over a validation set, and (2) weights corresponding to nodes of the search space lattice that are randomly initialized or configured using previously trained paths in the search space lattice.

traducir

TOKEN-POSITION HANDLING FOR SEQUENCE BASED NEURAL NETWORKS

NºPublicación: US2021319288A1 14/10/2021

Solicitante:

MICROSOFT TECHNOLOGY LICENSING LLC [US]

Resumen de: US2021319288A1

Embodiments of the present disclosure include a method for token-position handling comprising: processing a first sequence of tokens to produce a second sequence of tokens, wherein the second sequence of tokens has a smaller number of tokens than the first sequence of tokens; masking at least some tokens in the second sequence to produce masked tokens; moving the masked tokens to the beginning of the second sequence to produce a third sequence; encoding tokens in the third sequence into a set of numeric vectors in a first array; and processing the first array in a transformer neural network to determine correlations among the third sequence, the processing the first array producing a second array.

traducir

TRAINING A DATA CODING SYSTEM FOR USE WITH MACHINES

NºPublicación: WO2021205066A1 14/10/2021

Solicitante:

NOKIA TECHNOLOGIES OY [FI]

Resumen de: WO2021205066A1

Example embodiments relate to training (1110) encoder and/or decoder neural networks (1102, 1104). An apparatus may obtain a primary model (1002) configured to perform a task. The apparatus may further obtain an auxiliary model (1004) and train (1008) the auxiliary model to perform the same task, for example based on a first loss function (1006) comprising an output of the primary model (1002) and an output of the auxiliary model (1004). The apparatus may further determine gradients of a second loss function (1108) with respect to input of the auxiliary model (1004). The second loss function may comprise the output of the auxiliary model. An effective depth of the auxiliary model may be lower than an effective depth of the primary model in order to obtain gradients with higher magnitude. The apparatus may then train (1110) the encoder and/or decoder neural networks (1102, 1104) based on the gradients of the second loss function. Apparatuses, methods, and computer programs are disclosed.

traducir

TRAINING A DATA CODING SYSTEM COMPRISING A FEATURE EXTRACTOR NEURAL NETWORK

NºPublicación: WO2021205065A1 14/10/2021

Solicitante:

NOKIA TECHNOLOGIES OY [FI]

Resumen de: WO2021205065A1

Example embodiments provide a system for training a data coding pipeline comprising a feature extractor neural network (1012), an encoder neural network (1014), and a decoder neural network (1024) configured to reconstruct input data based on encoded features. A plurality of losses corresponding to different tasks (1241) may be determined for the coding pipeline. Tasks may be performed based on an output of the coding pipeline. A weight update may be determined for at least a subset of the coding pipeline based on the plurality of losses. The weight update may be configured to reduce a number of iterations for fine-tuning the coding pipeline for one of the tasks. This enables faster adaptation of the coding pipeline for one of the tasks after deployment of the coding pipeline. Apparatuses, methods, and computer programs are disclosed. Apparatuses, methods, and computer programs are disclosed.

traducir

END-TO-END CAMERA CALIBRATION FOR BROADCAST VIDEO

NºPublicación: WO2021207569A1 14/10/2021

Solicitante:

STATS LLC [US]

US_2021319587_A1

Resumen de: WO2021207569A1

A system and method of calibrating a broadcast video feed are disclosed herein. A computing system retrieves a plurality of broadcast video feeds that include a plurality of video frames. The computing system generates a trained neural network, by generating a plurality of training data sets based on the broadcast video feed and learning, by the neural network, to generate a homography matrix for each frame of the plurality of frames. The computing system receives a target broadcast video feed for a target sporting event. The computing system partitions the target broadcast video feed into a plurality of target frames. The computing system generates for each target frame in the plurality of target frames, via the neural network, a target homography matrix. The computing system calibrates the target broadcast video feed by warping each target frame by a respective target homography matrix.

traducir

METHOD FOR CONSTRUCTING PEST DETECTION MODEL

NºPublicación: WO2021203505A1 14/10/2021

Solicitante:

FJDYNAMICS TECH NANJING LTD [CN]

CN_112464971_A

Resumen de: WO2021203505A1

A method for constructing a pest detection model, the method comprising the following steps: constructing a training set, a validation set, and a test set on the basis of pictures of a pest situation assessing lamp; enhancing data of the training set and oversampling a small target sample; constructing a convolutional attention network in a mode in which a VGG16 convolutional model and an attention mechanism of a neural network are combined; training a convolution attention grid model by means of a first-order momentum stochastic gradient descent algorithm, and finely adjusting parameters of the convolution attention grid model by utilizing the verification set; and testing the convolution attention model by using the test set to construct a pest detection model, and as such, the pest detection model has high detection precision and strong robustness.

traducir

IMAGE-BASED DEFECTS IDENTIFICATION AND SEMI-SUPERVISED LOCALIZATION

NºPublicación: US2021319546A1 14/10/2021

Solicitante:

SAMSUNG DISPLAY CO LTD [KR]

Resumen de: US2021319546A1

A system for manufacturing defect classification is presented. The system includes a first neural network receiving a first data as input and generating a first output, a second neural network receiving a second data as input and generating a second output, wherein first neural network and the second neural network are trained independently from each other, and a fusion neural network receiving the first output and the second output and generating a classification. The first data and the second data do not have to be aligned. Hence, the system and method of this disclosure allows various type of data that are collected during manufacturing to be used in defect classification.

traducir

IMAGE RECOGNITION METHOD AND APPARATUS

NºPublicación: US2021319249A1 14/10/2021

Solicitante:

CANAAN BRIGHT SIGHT CO LTD [CN]

CN_110874632_A

Resumen de: US2021319249A1

An image recognition method and apparatus. The method comprises: obtaining original image data, convolutional neural network configuration parameters, and convolutional neural network operation parameters from a data transfer bus, the original image data comprising M pieces of pixel data, and M being a positive integer (101); and performing convolutional neural network operation on the original image data by a convolutional neural network operation module according to the convolutional neural network configuration parameters and the convolutional neural network operation parameters (102), wherein the convolutional neural network operation module comprises a convolution operation unit, a batch processing operation unit, and an activation operation unit connected in sequence. The method improves the real timeliness of image recognition.

traducir

Dense and Discriminative Neural Network Architectures for Improved Object Detection and Instance Segmentation

NºPublicación: US2021319242A1 14/10/2021

Solicitante:

INCEPTION INST OF ARTIFICIAL INTELLIGENCE LTD [AE]

Resumen de: US2021319242A1

This disclosure relates to improved techniques for performing computer vision functions, including common object detection and instance segmentation. The techniques described herein utilize neural network architectures to perform these functions in various types of images, such as natural images, UAV images, satellite images, and other images. The neural network architecture can include a dense location regression network that performs object localization and segmentation functions, at least in part, by generating offset information for multiple sub-regions of candidate object proposals, and utilizing this dense offset information to derive final predictions for locations of target objects. The neural network architecture also can include a discriminative region-of-interest (Rol) pooling network that performs classification of the localized objects, at least in part, by sampling various sub-regions of candidate proposals and performing adaptive weighting to obtain discriminative features.

traducir

GENERATOR EXPLOITATION FOR DEEPFAKE DETECTION

NºPublicación: US2021319240A1 14/10/2021

Solicitante:

INTEL CORP [US]

Resumen de: US2021319240A1

An apparatus to facilitate generator exploitation for deepfake detection is disclosed. The apparatus includes one or more processors to: alter a generative neural network of a deepfake generator with one or more modifications for deepfake detection; train the generative neural network having the one or more modifications and a discriminative neural network of the deepfake generator, wherein training the generative neural network and the discriminative neural network to facilitate the generative neural network to generate deepfake content comprising the one or more modifications; and communicate identification of the one or more modifications to a deepfake detector to cause the deepfake detector to identify deepfake content generated by the deepfake generator that comprises at least one of the one or more modifications.

traducir

TEMPORALLY DISTRIBUTED NEURAL NETWORKS FOR VIDEO SEMANTIC SEGMENTATION

NºPublicación: US2021319232A1 14/10/2021

Solicitante:

ADOBE INC [US]

Resumen de: US2021319232A1

A Video Semantic Segmentation System (VSSS) is disclosed that performs accurate and fast semantic segmentation of videos using a set of temporally distributed neural networks. The VSSS receives as input a video signal comprising a contiguous sequence of temporally-related video frames. The VSSS extracts features from the video frames in the contiguous sequence and based upon the extracted features, selects, from a set of labels, a label to be associated with each pixel of each video frame in the video signal. In certain embodiments, a set of multiple neural networks are used to extract the features to be used for video segmentation and the extraction of features is distributed among the multiple neural networks in the set. A strong feature representation representing the entirety of the features is produced for each video frame in the sequence of video frames by aggregating the output features extracted by the multiple neural networks.

traducir

Convex Representation of Objects Using Neural Network

NºPublicación: US2021319209A1 14/10/2021

Solicitante:

GOOGLE LLC [US]

Resumen de: US2021319209A1

Methods, systems, and apparatus including computer programs encoded on a computer storage medium, for generating convex decomposition of objects using neural network models. One of the methods includes receiving an input that depicts an object. The input is processed using a neural network to generate an output that defines a convex representation of the object. The output includes, for each of a plurality of convex elements, respective parameters that define a position of the convex element in the convex representation of the object.

traducir

Transition Detector Neural Network

NºPublicación: US2021321150A1 14/10/2021

Solicitante:

GRACENOTE INC [US]

WO_2021207426_A1

Resumen de: US2021321150A1

In one aspect, an example method includes (i) extracting a sequence of audio features from a portion of a sequence of media content; (ii) extracting a sequence of video features from the portion of the sequence of media content; (iii) providing the sequence of audio features and the sequence of video features as an input to a transition detector neural network that is configured to classify whether or not a given input includes a transition between different content segments; (iv) obtaining from the transition detector neural network classification data corresponding to the input; (v) determining that the classification data is indicative of a transition between different content segments; and (vi) based on determining that the classification data is indicative of a transition between different content segments, outputting transition data indicating that the portion of the sequence of media content includes a transition between different content segments.

traducir

DEEP CONVOLUTIONAL NEURAL NETWORK WITH SELF-TRANSFER LEARNING

NºPublicación: US2021319559A1 14/10/2021

Solicitante:

GEN ELECTRIC [US]

US_2020013165_A1

Resumen de: US2021319559A1

Systems and techniques for facilitating a deep convolutional neural network with self-transfer learning are presented. In one example, a system includes a machine learning component, a medical imaging diagnosis component and a visualization component. The machine learning component generates learned medical imaging output regarding an anatomical region based on a convolutional neural network that receives medical imaging data. The machine learning component also performs a plurality of sequential downsampling and upsampling of the medical imaging data associated with convolutional layers of the convolutional neural network. The medical imaging diagnosis component determines a classification and an associated localization for a portion of the anatomical region based on the learned medical imaging output associated with the convolutional neural network. The visualization component generates a multi-dimensional visualization associated with the classification and the localization for the portion of the anatomical region.

traducir

System on a Chip with Deep Learning Accelerator and Random Access Memory

NºPublicación: US2021319305A1 14/10/2021

Solicitante:

MICRON TECHNOLOGY INC [US]

WO_2021207236_A1

Resumen de: US2021319305A1

Systems, devices, and methods related to a Deep Learning Accelerator and memory are described. An integrated circuit may be configured with: a Central Processing Unit, a Deep Learning Accelerator configured to execute instructions with matrix operands; random access memory configured to store first instructions of an Artificial Neural Network executable by the Deep Learning Accelerator and second instructions of an application executable by the Central Processing Unit; one or connections among the random access memory, the Deep Learning Accelerator and the Central Processing Unit; and an input/output interface to an external peripheral bus. While the Deep Learning Accelerator is executing the first instructions to convert sensor data according to the Artificial Neural Network to inference results, the Central Processing Unit may execute the application that uses inference results from the Artificial Neural Network.

traducir

METHODS AND APPARATUS TO IMPLEMENT PARALLEL ARCHITECTURES FOR NEURAL NETWORK CLASSIFIERS

NºPublicación: US2021319319A1 14/10/2021

Solicitante:

INTEL CORP [US]

Resumen de: US2021319319A1

Methods, apparatus, systems, and articles of manufacture are disclosed to implement parallel architectures for neural network classifiers. An example non-transitory computer readable medium comprises instructions that, when executed, cause a machine to at least: process a first stream using first neural network blocks, the first stream based on an input image; process a second stream using second neural network blocks, the second stream based on the input image; fuse a result of the first neural network blocks and the second neural network blocks; perform average pooling on the fused result; process a fully connected layer based on the result of the average pooling; and classify the image based on the output of the fully connected layer.

traducir

LEARNING AND DETECTION METHOD OF NEURAL NETWORK MODEL FOR FLAME DETERMINATION, AND APPARATUS PERFORMING THE SAME

NºPublicación: US2021319318A1 14/10/2021

Solicitante:

GYNETWORKS CO LTD [KR]

JP_2021018788_A

Resumen de: US2021319318A1

Disclosed herein is a learning method of a neural network model for flame determination. The learning method of a neural network includes generating a learning image including a fake image generated by combining a real fire image and an arbitrary flame image with a background image; inputting the learning image to a first neural network model and outputting a determination result for whether a flame is present; and updating a weight in a layer extracting features of the learning image from the first neural network model using the determination result. According to the present invention, data of various fire situations may be secured, a performance of the neural network model that determines an occurrence of the fire through the secured data may be increased, and a quality of data for learning may be increased to allow the neural network model itself to predict various situations of fires.

traducir

System and Method for Detecting Adversarial Attacks

NºPublicación: US2021319784A1 14/10/2021

Solicitante:

MITSUBISHI ELECTRIC RES LABORATORIES INC [US]

WO_2021205746_A1

Resumen de: US2021319784A1

A linguistic system for transcribing an input, where the linguistic system comprises a processor configured to execute a neural network multiple times while varying weights of at least some nodes of the neural network to produce multiple transcriptions of the input. Further, determine a distribution of pairwise distances of the multiple transcriptions; determine a legitimacy of the input based on the distribution; and transcribe the input using stored weights of the nodes of the neural network when the input is determined as legitimate to produce a final transcription of the input.

traducir

SYSTEM AND METHOD FOR AUGMENTING FEW-SHOT OBJECT CLASSIFICATION WITH SEMANTIC INFORMATION FROM MULTIPLE SOURCES

Nº publicación: US2021319263A1 14/10/2021

Solicitante:

IBM [US]

Resumen de: US2021319263A1

Embodiments may provide learning and recognition of classifications using only one or a few examples of items. For example, in an embodiment, a method of computer vision processing may be implemented in a computer comprising a processor, memory accessible by the processor, and computer program instructions stored in the memory and executable by the processor, the method may comprise training a neural network system implemented in the computer system to classify images into a plurality of classes using one or a few training images for each class and a plurality of associated semantic information, wherein the plurality of associated semantic information is from a plurality of sources and comprises at least some of class/object labels, textual description, or attributes, and wherein the neural network is trained by modulating the training images by sequentially applying the plurality of associated semantic information and classifying query images using the trained neural network system.

traducir

Página1 de 7 nextPage por página

punteroimgVolver