Abstract: A method of filtering a target pixel in an image forms, for a kernel of pixels comprising the target pixel and its neighbouring pixels, a data model to model pixel values within the kernel; calculates a weight for each pixel of the kernel comprising: (i) a geometric term dependent on a difference in position between that pixel and the target pixel; and (ii) a data term dependent on a difference between a pixel value of that pixel and its predicted pixel value according to the data model; and uses the calculated weights to form a filtered pixel value for the target pixel, e.g. by updating the data model with a weighted regression analysis technique using the calculated weights for the pixels of the kernel; and evaluating the updated data model at the target pixel position so as to form the filtered pixel value for the target pixel.

Abstract: Computer-implemented methods are provided for generating a data structure representing tabular information in a scanned image. Such a method can include storing image data representing a scanned image of a table, processing the image data to identify positions of characters and lines in the image, and mapping locations in the image of information cells, each containing a set of the characters, in dependence on said positions. The method can also include, for each cell, determining cell attribute values, dependent on the cell locations, for a predefined set of cell attributes, and supplying the attribute values as inputs to a machine-learning model trained to pre-classify cells as header cells or data cells in dependence on cell attribute values.

Abstract: System and method of facilitating generation of 3D training content is disclosed. In one embodiment, the method may include classifying a 3D model to determine the one or more objects associated with the 3D model by implementing a machine learning model. The classification of the 3D model includes analysis of the 3D model and extraction of feature set for the one or more objects in the 3D model in order to determine a class and a tag for the one or more objects in the 3D model. The method may further include assigning the class and the tag to the one or more objects in the 3D model. The method may further include assigning one or more functionalities to the objects based on the at least one tag, and creating a 3D training content by using the objects assigned with the at least one tag and the one or more functionalities.

Abstract: Methods and systems for training machine vision models (MVMs) with “noisy” training datasets are described. A noisy set of images is received, where labels for some of the images are “noisy” and/or incorrect. A progressively-sequenced learning curriculum is designed for the noisy dataset, where the images that are easiest to learn machine-vision knowledge from are sequenced near the beginning of the curriculum and images that are harder to learn machine-vision knowledge from are sequenced later in the curriculum. An MVM is trained via providing the sequenced curriculum to a supervised learning method, so that the MVM learns from the easiest examples first and the harder training examples later, i.e., the MVM progressively accumulates knowledge from simplest to most complex. To sequence the curriculum, the training images are embedded in a feature space and the “complexity” of each image is determined via density distributions and clusters in the feature space.

Abstract: An information processing device includes: a memory; and a processor coupled to the memory and configured to: accept specification of a feature region that includes a characteristic portion in a learning image; create first data that indicates a degree of overlap between the feature region and each of divided regions obtained by dividing the learning image; perform image conversion on the learning image by an image classification program in which element filters are combined; calculate an image feature value of each of the divided regions from an image obtained; create second data that indicates the image feature value corresponding to each of the divided regions; calculate a similarity between the first data and the second data; and use a result of evaluating the image classification program on the basis of the similarity, for execution of a genetic programming.

Abstract: The present disclosure relates to an object selection system that automatically detects and selects objects in a digital image utilizing a large-scale object detector. For instance, in response to receiving a request to automatically select a query object with an unknown object class in a digital image, the object selection system can utilize a large-scale object detector to detect potential objects in the image, filter out one or more potential objects, and label the remaining potential objects in the image to detect the query object. In some implementations, the large-scale object detector utilizes a region proposal model, a concept mask model, and an auto tagging model to automatically detect objects in the digital image.

Abstract: Apparatus, systems, and methods to generate an edge aware brush for navigation and segmentation of images via a user interface are disclosed. An example processor is to at least: construct a brush for segmentation of image data; provide an interactive representation of the brush with respect to the image data via a user interface, the interactive representation to be displayed and made available for interaction in each of a plurality of viewports provided for display of views of the image data in the user interface; enable update of the viewports based on manipulation of the representation; facilitate display of a preview of a segmentation of the image data corresponding to a location of the representation; and, when the segmentation is confirmed, facilitate generation of an output based on the segmentation.

Abstract: A method includes: capturing one or more images of an unorganized collection of items inside a first machine; determining one or more item types of the unorganized collection of items from the one or more images, comprising: dividing a respective image in the one or more images into a respective plurality of sub-regions; performing feature detection on the respective plurality of sub-regions to obtain a respective plurality of regional feature vectors, wherein a regional feature vector for a sub-region indicates characteristics for a plurality of predefined local item features for the sub-region; generating an integrated feature vector by combining the respective plurality of regional feature vectors; and applying a plurality of binary classifiers to the integrated feature vector; and selecting a machine setting for the first machine based on the determined one or more clothes type in the unorganized collection of items.

Abstract: A radiographic image processing apparatus includes a specifying unit that specifies an irradiation field region in a radiographic image, a determination unit that determines, based on the irradiation field region, a rotation angle of the radiographic image, a rotation unit that rotates, based on the determined rotation angle, the radiographic image, and a display control unit that performs control to display the rotated radiographic image on a display unit.

Abstract: Disclosed are a method, terminal and computer readable storage medium for image classification. The method includes: determining an image feature vector of an image based on a convolutional neural network, where the image comprises textual information; determining a text feature vector based on the textual information and an embedded network; determining an image-text feature vector by joining the image feature vector with the text feature vector; and determining a category of the image based on a result of a deep neural network, where the result is determined based on the image feature vector, the text feature vector and the image-text feature vector.

Abstract: An embodiment of the present application provides an image analysis method and device, and the method includes: receiving target information sent by a target image acquisition device; wherein the target information includes at least: a first image acquired by the target image acquisition device, and the identification information of the target image acquisition device; obtaining target area information corresponding to the identification information of the target image acquisition device according to a correspondence between pre-stored identification information of each image acquisition device and respective area information; in the first image, determining a second image corresponding to the target area information, and analyzing the second image.

Abstract: A method for modelling a composition of an object of interest comprises segmenting object of interest image data provided by computer tomography image data resulting in a plurality of image segments. A determined Hounsfield density value is then extracted from the object of interest image data for each image segment. A component ratio of at least two component classes is defined for the object of interest, the at least two component classes having different component Hounsfield density values. At least one component class is assigned to each image segment based on the corresponding determined Hounsfield density value resulting in simulated image segments comprising the component Hounsfield density values. The simulated image segments define simulated image data of the object of interest, where a ratio of the assigned component classes corresponds to the component ratio. A deviation between the simulated image data and the object of interest image data is then determined.

Abstract: An electronic apparatus is provided. The electronic apparatus includes a camera configured to obtain a user image by capturing an image of a user, a memory configured to store one or more instructions, and a processor configured to execute the one or more instructions stored in the memory, wherein the processor is further configured to, by executing the one or more instructions, recognize the user from a face region of the user image by using a first recognition model learned based on face information of a plurality of users, extract additional feature information regarding the recognized user from the user image, allow the first recognition model to additionally learn based on the extracted additional feature information, recognize the user from a person region of the user image by using an additionally learned second recognition model, and output a recognition result of the second recognition model.

Abstract: A method for world objects tracking and prediction by an autonomous vehicle includes receiving, from sensors of the AV, a first observation data; associating the first observation data with a first world object; determining hypotheses for the first world object; determining a respective hypothesis likelihood of each of the hypotheses indicating a likelihood that the first world object follows the intention; determining, for at least one hypothesis of the hypotheses, a respective state; and in response to a query, providing a hypothesis of the hypotheses based on the respective hypothesis likelihood of each of the hypotheses. A hypothesis corresponds to an intention of the first world object and the respective state includes predicted positions of the first world object.

Abstract: A method for producing an identification article is described. The method includes associating one or more data items with one or more line patterns, where the line patterns include one or more line segments. A particular data item is associated with at least a subset of line segments and the particular data item corresponds to an attribute of a user of the identification article. The method further includes integrating the line patterns with graphical features of the identification article. The graphical features being discernable by a human eye and the subset of line segments is indiscernible by the human eye. The method includes identifying a region of the identification article that includes the subset of line segments associated with the particular data item and associating the particular line pattern with an identifying feature of the user.

Abstract: A computing system includes a memory configured to store instructions, and one or more processors configured to execute the instructions to receive data relating to an image or a user, determine a feature from the data, identify a user preference from a user profile, obtain a model, and segment the image based on the feature, the user preference, and the model. The model is generated by determining a historical feature from historical data as an input, determining a desired output, obtaining a preliminary model based on the input and the desired output, determining an actual output of the preliminary model, determining error criteria between the actual output and the desired output, and generating the model by updating the preliminary model based on the error criteria.

Abstract: A medical image processing device includes a motion amount calculating unit configured to compare a first image obtained by capturing a subject image taken by an endoscope with a second image obtained by capturing the subject image at a chronologically different timing from the first image and calculate a motion amount for the second image every two or more areas in the first image; and an area specifying unit configured to specify a mask area other than the subject image included in the first image from an inside of a mask candidate area in which the motion amount is equal to or less than a specific first threshold value in the first image.

Abstract: Examples are disclosed that relate to computing devices and methods for synthesizing a novel pose of an object. One example provides a method comprising receiving a reference image of an object corresponding to an original viewpoint. The reference image of the object is translated into a depth map of the object, and a new depth map of the object is synthesized to correspond to a new viewpoint. A new image of the object is generated from the new viewpoint based on the new depth map of the object and the reference image of the object.

Abstract: An image processing apparatus that determines a correction parameter for correcting distortion or a position of an image projected on a screen by a projection unit, the apparatus comprises a holding unit which holds a reference pattern constituted by a background region and a plurality of significant pixels scattered in the background region and having a pixel value different from a pixel value of the background region, a generating unit which generates a pattern image to be projected by changing a value of a pixel near the significant pixel in the reference pattern, an acquisition unit which acquires a captured image obtained by an image capturing unit by capturing the pattern image generated by the generating unit so as to include a projection image projected by the projection unit, and a determining unit which determines the correction parameter from the captured image.

Abstract: A solution for determination of a three-dimensional difference image dataset of an examination volume. Here two-dimensional real image datasets relating to the examination volume are received via an interface, each of the two-dimensional real image datasets including a two-dimensional x-ray projection of the examination volume in relation to a projection direction. Furthermore, the first difference image dataset is determined based on the two-dimensional real image datasets and based on a first trained function via a processing unit. Here the first difference image dataset is at least two-dimensional, in particular at least three-dimensional, in particular the first difference image dataset is three-dimensional or four-dimensional. The determination of the first difference image dataset based on the two-dimensional real image datasets and based on a trained function enables mask recordings of the examination volume to be dispensed with, and thus the x-ray load of the examination volume to be reduced.