Abstract: An image processing apparatus acquires a first pixel size and a second pixel size that are pixel sizes in a predetermined axis direction of a first image and a second image captured at different points in time, respectively, determines whether the first pixel size and second pixel size differ from each other, and decides, if the first pixel size differs from the second pixel size, a size in a predetermined axis direction of a comparison area based on a larger one of the first pixel size and the second pixel size. The comparison area includes a plurality of gray levels, and is compared to a gray level of a position of interest in one of the first and second image, and the comparison area existing in the other of the first and second image, different from the one image.

Abstract: A machine learning method of a machine learning device including a machine learning classifier includes receiving, at the machine learning device, an image and first class information associated with the image, generating, at the machine learning device, second class information associated with the image by performing classification on the image by using the machine learning classifier, and as the second class information is generated, updating, at the machine learning device, the machine learning classifier by performing a first learning operation when a guide map is received together with the image and performing, at the machine learning device, a second learning operation different from the first learning operation when the guide map is not received together with the image.

Abstract: The subject matter of this specification can be implemented in, among other things, methods, systems, computer-readable storage medium. A method can include receiving, by a processing device, image data having one or more image frames indicative of a state of a meal preparation area. The method may further include, determining, based on the image data, a first feature characterization of a first meal preparation item associated with the state of the meal preparation area. The method may further include determining that the first feature characterization does not meet object classification criteria for a set of object classifications. The method may further include causing a notification indicating the first meal preparation item and one of an object classification or a classification status corresponding to the first meal preparation item on a graphical user interface (GUI).

Abstract: The present embodiments relate generally to systems and methods for performing a measurement on an ultrasound image displayed on a touchscreen device. The method may include: receiving, via the touchscreen device, first input coordinates corresponding to a point on the ultrasound image; using the first input coordinates as a seed for performing a contour identification process on the ultrasound image, wherein the contour identification process performs contour evolution using morphological operators to iteratively dilate from the first input coordinates; upon identification of a contour from the contour identification process, placing measurement calipers on the identified contour; and storing a value identified by the measurement calipers as the measurement.

Abstract: A computer produces predictions throughout a raster field in response to point data, by obtaining a partially empty matrix of point data, filling a matrix of extrapolated raster data by dilating the point data in a first convolutional neural network, and generating a matrix of aggregate raster data by combining the extrapolated raster data with organic raster data in a second convolutional neural network.

Abstract: A method including providing a sensor device including one or several sensors. The sensor device is arranged to perform at least one high-power type measurement and at least one low-power type measurement and includes at least one image sensor arranged to depict a person by a measurement of said high-power type. Each of the low-power type measurements over time requires less electric power for operation as compared to each of the high-power type measurements. The method includes detecting a potential presence of the person using at least one of said low-power type measurements. The method includes producing, using one of the high-power type measurements, an image depicting a person and detecting a presence of the person based on im-age analysis of the image. The method includes detecting, using at least one of the low-power type measurements, a maintained presence of the person. The method includes failing to detect a maintained presence of the person.

Abstract: A system and method to adjust an augmented reality (AR) image involve a front-facing camera in a vehicle. The camera obtains an image that includes a road surface ahead of the vehicle. The system also includes a processor to detect lane lines of the road surface in three dimensions, to perform interpolation using the lane lines to determine a ground plane of the road surface, and to adjust the AR image to obtain an adjusted AR image such that all points of the adjusted AR image correspond with points on the ground plane.

Abstract: A system for inspecting the reliability of an image. The system may include a processor in communication with a client device; and a storage medium. The storage medium may store instructions that, when executed, configure the processor to perform operations including: obtaining a plurality of images; categorizing the images into a plurality of image classes; calculating a plurality of probability outcomes; determining whether highest predicted probabilities of the images are less than a first threshold and whether an entropy of a predicted density of the probability outcomes exceeds a second threshold; indicating whether the image is associated with the image classes; ranking, the image amongst the plurality of images; filtering, a plurality of low reliability images according to a third threshold; providing, a likelihood of whether a user scanned a vehicle object associated with the image; and identifying a percentage of user scan failures.

Abstract: Methods and systems for autonomous cardiac mapping are disclosed. An example system for autonomous cardiac mapping of a heart chamber includes a processor being configured to acquire a representative geometric shell of the heart chamber, control a robotic device to autonomously navigate a mapping probe to a plurality of locations within the heart chamber based at least in part on the representative geometric shell, and generate a three-dimensional electroanatomical map of the heart chamber based on electrical data collected by the probe at the plurality of locations.

Abstract: Example methods and apparatus for associating media devices with a demographic composition of a geographic area are disclosed. Disclosed example apparatus are to obtain a plurality of Internet Protocol addresses assigned to a media device associated with a panel member. Disclosed example apparatus are also to determine a most used Internet Protocol address from the plurality of Internet Protocol addresses, determine a geographic location corresponding to the most used Internet Protocol address, associate a geographic area with the media device in response to a determination that the geographic location corresponds to a location of an internet service provider, determine a demographic profile associated with the geographic area, and associate the demographic profile with the media device.

Abstract: A method is provided for identifying improperly redacted information in documents. The documents are analyzed to detect redacted areas and text elements and to identify an intersection between a redacted area and a text element. When an area of the intersection is greater than an intersection threshold, the document is identified as containing improperly redacted information.

Abstract: Systems, methods, and computer programmable products are described herein for generating a robotic process automation bot script from a handwritten sketch. A digitally encoded image having a handwritten diagram with objects is received. The handwritten diagram is converted into a digital flowchart having a plurality of nodes. A confidence score is determined for the nodes based on (i) a textual similarity of contents of the node and a plurality of executable bot activities or (ii) a contextual similarity between a set of activities which measures how often they coappear in a plurality of reference bots. A robotic process automation bot script is generated by mapping the plurality of nodes to an executable bot activity of the plurality of executable bot activities based on the confidence score. The robotic process automation bot is then provided to a robotic process automation system for further modeling.

Abstract: Techniques are described for passive three-dimensional (3D) face imaging based on macro-structure and micro-structure image sizing, such as for biometric facial recognition. A set of images of a user's face is processed to extract authentication deterministic macro-structure (DMAS) measurements. A database includes profile DMAS measurements, profile location definitions for deterministic micro-structure (DMIS) feature regions, and profile DMIS signatures computed for the DMIS feature regions. A first-level authentication determination can be based on comparing the authentication DMAS measurements with the profile DMAS measurements. Authentication DMIS signatures can be computed from sub-images obtained for the DMIS feature regions at the profile location definitions. A second-level authentication determination can be based on comparing the authentication DMIS signatures with the profile DMIS signatures.

Abstract: The present disclosure relates to a communication method and system for converging a 5th-Generation (5G) communication system for supporting higher data rates beyond a 4th-Generation (4G) system with a technology for Internet of Things (IoT). The present disclosure may be applied to intelligent services based on the 5G communication technology and the IoT-related technology, such as smart home, smart building, smart city, smart car, connected car, health care, digital education, smart retail, security and safety services.

Abstract: A behavior recognition method includes extracting, by a computer, skeleton information including a plurality of joint positions for each frame of an image, calculating a first set of motion feature amounts from the skeleton information, calculating a plot position by plotting the first set of motion feature amounts on a feature amount space defined by a second set of motion feature amounts, the plot position being a position where the first set of motion feature amounts is plotted on the feature amount space, the feature amount space having a plurality of mapping areas in which respective ranges corresponding to predetermined higher-level behaviors to be recognized are mapped, and expanding, when a degree of divergence from a minimum distance to other distances among distances between the plot position and each of the plurality of mapping areas satisfies a predetermined criterion, a mapping area at the minimum distance from the plot position.

Abstract: Another aspect of the present disclosure relates to a method for identification verification. The method may include receiving, from a computing device, a request for identification verification of an individual at a web server. The web server may include an orientation verification module and a facial recognition module. The method may include receiving at a database an identification card image and a picture image. The method may include verifying an orientation of the identification card image is oriented relative to an accepted orientation using the orientation verification module. The method may include, if the identification card image is not oriented relative to the accepted orientation, using the web server, iteratively rotating the identification card image in one or more increments and assessing if the identification card image is oriented at the accepted orientation until the identification card image is oriented at the accepted orientation.

Abstract: A vehicle recognition unit of an information presenting device specifies a position of each of a plurality of vehicles, based on images captured by a plurality of imaging devices installed at a construction site. A vehicle designating unit receives designation of a designated vehicle, among the plurality of specified vehicles. A display control unit outputs a signal for displaying in an overlapping manner a proximity plot representing a position of another vehicle whose distance from the designated vehicle is less than a threshold on an overhead image of the construction site.

Abstract: A system and method for performing distributed facial recognition divides processing steps between a user engagement device/robot, having lower processing power, and a remotely located server, having significantly more processing power. Images captured by the user engagement device/robot are processed at the device/robot by applying a first set of image processing steps that includes applying a first face detection. First processed images having at least one detected face is transmitted to the server, whereat a second set of image processing steps are applied to determine a stored user facial image matching the detected face of the first processed image. At least one user property associated to the given matching user facial image is then transmitted to the user engagement device/robot. An interactive action personalized to the user can further be performed at the user engagement device/robot.

Abstract: An information processing device, an information processing method, and a program are provided. The information processing device includes a virtual object control unit that generates a control parameter that controls display of a first virtual object corresponding to a first real object and a second virtual object corresponding to a second real object on the basis of first posture information indicating a posture of the first real object and second posture information indicating a posture of the second real object, and a determination unit that makes a determination related to contact between the first virtual object and the second virtual object. The virtual object control unit generates the control parameter on the further basis of a result of the determination.

Abstract: Drift regularization is provided to counteract variation in drift coefficients in analog neural networks. In various embodiments, a method of training an artificial neural network is illustrated. A plurality of weights is randomly initialized. Each of the plurality of weights corresponds to a synapse of an artificial neural network. At least one array of inputs is inputted to the artificial neural network. At least one array of outputs is determined by the artificial neural network based on the at least one array of inputs and the plurality of weights. The at least one array of outputs is compared to ground truth data to determine a first loss. A second loss is determined by adding a drift regularization to the first loss. The drift regularization is positively correlated to variance of the at least one array of outputs. The plurality of weights is updated based on the second loss by backpropagation.