An apparatus and a method for segmenting a steel microstructure phase are provided. The apparatus includes a storage configured for storing a machine learning algorithm and a processing device that segments a microstructure phase using the machine learning algorithm. The processing device is configured to receive label data, to learn a machine learning model by use of the label data as learning data for the machine learning model, and to segment a phase of a steel microstructure image by use of the learned machine learning model.

Systems and methods for compacting and anonymizing telemetry data in a storage system. A controller of a storage device may generate telemetry data based on collected features indicative of the performance of the storage device. The controller may store the telemetry data in the telemetry memory of the storage device. The controller may then transform the telemetry data into transformed telemetry data based on a dimension reduction algorithm, and transmit the transformed telemetry data to the host device.

A covariance matrix computation unit 81 computes a pseudo-in-domain covariance matrix from one or both of a within class covariance matrix and a between class covariance matrix of an out-of-domain Probabilistic Linear Discriminant Analysis (PLDA) model. A simultaneous diagonalization unit 82 computes a generalized eigenvalue and an eigenvector for a pseudo-in-domain covariance matrix and the class covariance matrix of the out-of-domain PLDA model on the basis of simultaneous diagonalization. An adaptation unit 83 computes one or both of a within class covariance matrix and a between class covariance matrix of an in-domain PLDA model using the generalized eigenvalues and eigenvectors. The covariance matrix computation unit 81 computes the pseudo-in-domain covariance matrix based on the out-of-domain PLDA model and a covariance matrix of in-domain data.

The present invention provides a monocular unsupervised depth estimation method based on contextual attention mechanism, belonging to the technical field of image processing and computer vision. The invention adopts a depth estimation method based on a hybrid geometric enhancement loss function and a context attention mechanism, and adopts a depth estimation sub-network, an edge sub-network and a camera pose estimation sub-network based on convolutional neural network to obtain high-quality depth maps. The present invention uses convolutional neural network to obtain the corresponding high-quality depth map from the monocular image sequences in an end-to-end manner. The system is easy to construct, the program framework is easy to implement, and the algorithm runs fast; the method uses an unsupervised method to solve the depth information, avoiding the problem that ground-truth data is difficult to obtain in the supervised method.

A region specification apparatus specifies a region of an object which is included in an input image and which includes a plurality of subclass objects having different properties. The region specification apparatus includes a first discriminator that specifies an object candidate included in the input image. The first discriminator has a component configured to predict at least one of movement or transformation of a plurality of anchors according to the property of the subclass object and specify an object candidate region surrounding the object candidate.

The present disclosure discloses a method and apparatus for generating an image, a device, a storage medium and a program product, relates to the field of artificial intelligence, and particularly to computer vision and deep learning technologies, and may be applied in smart cloud and power grid inspection scenarios. A particular implementation of the method comprises: acquiring an original insulator image; performing an image transformation on the original insulator image to obtain a composite insulator image; and inputting the original insulator image and the composite insulator image into a pre-trained generative adversarial network to generate a target insulator image. According to the implementation, the image transformation is performed on the original insulator image, and then, massive target insulator images are generated through the generative adversarial network.

The invention belongs to the field of computer vision technology, and provides a fully automatic natural image matting method. For image matting of a single image, it is mainly composed of the extraction of high-level semantic features and low-level structural features, the filtering of pyramid features, the extraction of spatial structure information, and the late optimization of the discriminator network. The invention can generate accurate alpha matte without any auxiliary information, saving the time for scientific researchers to mark auxiliary information and the interaction time when users use it.

A method of causing one or more processors to execute: performing learning of a model that is an algorithm of a vector neural network type to reproduce correspondence between a plurality of first data elements included in a first data set and a pre-label corresponding to each of the plurality of first data elements, in which the model has one or more neuron layers, each of the one or more neuron layers has one or more neuron groups, each of the one or more neuron groups has one or more neurons, and each of the one or more neurons outputs first intermediate data based on at least one of a first vector and a first activation; and inputting the first data set into the learned model and acquiring the first intermediate data output by the one or more neurons by being associated with the neuron.

An illustrative method includes accessing, by a computing device, a model simulating light scattered by a simulated target, the model comprising a plurality of parameters. The method further includes generating, by the computing device, a set of possible histogram data using the model with a plurality of values for the parameters. The method further includes determining, by the computing device, a set of components that represent the set of possible histogram data, the set of components having a reduced dimensionality from the set of possible histogram data.

The present disclosure provides a system and method for removing noise from an ultrasonic signal using a generative adversarial network (GAN). The present disclosure provides three input formats for the neural network (NN) in order to feed one-dimensional (1D) input data to the network. The system is generalizable to multiple noise sources, as it learns from different motion functions and noise types. The end-to-end system of the present disclosure is trained on raw ultrasonic signals with very little pre-processing or feature extraction.