A feature extractor and novel training objective are provided for content-based image retrieval. For example, a computer-implemented method includes applying a query image and a search image to a neural network of a feature extraction network of a computing device, the query image indicating an object to be searched for in the search image. The feature extraction network includes the neural network, a spatial feature neural network receiving a first output of the neural network pertaining to the search image, and an embedding network receiving a second output of the neural network pertaining to the query image. The method includes generating spatial search features from the spatial feature neural network, generating a query feature from the embedding network, applying the query feature to an artificial neural network (ANN) index, and determining an optimal matching result of an object in the search image based on an operation using the ANN index.

A method comprises identifying a representation of first statistical information in a document, identifying descriptive text that describes the first statistical information, determining whether the descriptive text accurately describes the first statistical information, and upon determination that the descriptive text does not accurately describe the first statistical information, generating alternative descriptive text that accurately describes the first statistical information.

A method of detecting unusual motion is provided, including: determining features occurring during a fixed time period; grouping the features into first and second subsets of the fixed time period; grouping the features in each of the first and second subsets into at least one pattern interval; and determining when an unusual event has occurred using at least one of the pattern intervals.

A method for use with a computing device. The method may include receiving a data set including a plurality of univariate data points and determining a target kernel bandwidth for a kernel density estimator (KDE). Determining the target kernel bandwidth may include computing a plurality of sample KDEs and selecting the target kernel bandwidth based on the sample KDEs. The method may further include computing the KDE for the data set using the target kernel bandwidth. For one or more tail regions of the data set, the method may further include computing one or more respective tail extensions. The method may further include computing and outputting a renormalized piecewise density estimator that, in each tail region, equals a renormalization of the respective tail extension for that tail region, and, outside the one or more tail regions, equals a renormalization of the KDE.

Techniques are described for selective extraction of color attributes from digital images that overcome the challenges experienced in conventional systems for color extraction. In an implementation, a user applies a region selector to a source image to select a portion of the source image for color attribute extraction. A graphics editing system identifies a selected region of the source image as well as visual objects of the source image included as part of the selected region. The graphics editing system iterates through the selected visual objects and extracts color attributes from the visual objects, such as color values, patterns, gradients, gradient stops, opacity, color area, and so forth. The graphics editing system then generates a color palette that includes the extracted color attributes, and the color palette is able to be utilized for various image editing tasks, such as digital image creation and transformation.

A distance measurement accuracy is improved without increasing power consumption of an image processing apparatus that performs distance-measuring processing. In one embodiment, an image processing apparatus for calculating distance information on an image has a reliability calculation unit 113 configured to calculate reliability in accordance with contrast for each pixel of the image and a distance calculation unit 116 configured to calculate distance information on each of the pixels based on reliability of each of the pixels. The distance calculation unit 116 calculates the distance information about a second pixel group whose reliability is lower than that of a first pixel group by using a collation area whose size is larger than a predetermined size in a range in which an amount of calculation in a case where a collation area of the predetermined size is used for all the pixels of the image is not exceeded.

A method of producing a model to detect changes in forest cover is disclosed. The method includes obtaining forest-cover classification data of a land area. The land area includes one or more subregions having unchanged forest-cover classifications between a first time and a second time. The method further includes obtaining image data of the subregions at multiple times. For at least one forest-cover classification, the method includes applying a statistical analysis to the image data to determine one or more threshold values representing measurement variations. The method further includes comparing subsequently obtained image data to the one or more threshold values and classifying the one or more subregions as changed or unchanged based on the comparison of subsequently obtained image data to the one or more threshold values.

We propose a neural network-based base caller that detects and accounts for stationary, kinetic, and mechanistic properties of the sequencing process, mapping what is observed at each sequence cycle in the assay data to the underlying sequence of nucleotides. The neural network-based base caller combines the tasks of feature engineering, dimension reduction, discretization, and kinetic modelling into a single end-to-end learning framework. In particular, the neural network-based base caller uses a combination of 3D convolutions, 1D convolutions, and pointwise convolutions to detect and account for assay biases such as phasing and prephasing effect, spatial crosstalk, emission overlap, and fading.

An object recognition method of a refrigerator is disclosed. The disclosed object recognition method of a refrigerator comprises the steps of: obtaining a captured image of a storage compartment of a refrigerator; checking the change in the imaging direction of an image capturing device which has captured the image of the storage compartment, when a change in the captured image is confirmed compared to a previously stored image; and performing an object recognition operation of the captured image when the imaging direction is maintained.

An image processing system may include an imaging device for capturing an image and an image processing apparatus for processing the image. The imaging device may include an imaging unit for capturing the image, a first recording unit for recording information relating to the image, the information being associated with the image, and a first transmission control unit for controlling transmission of the image to the image processing apparatus. The image processing apparatus may include a reception control unit for controlling reception of the image transmitted from the imaging device, a feature extracting unit for extracting a feature of the received image, a second recording unit for recording the feature, extracted from the image, the feature being associated with the image, and a second transmission control unit for controlling transmission of the feature to the imaging device.