Abstract: A computer-implemented method for guiding a surgeon to acquire one or more registration points on a body of a patient. The method includes matching a non-patient-specific virtual mask to image data of a patient, and obtaining a plurality of image surface points defining a skin surface in the image data. The method further comprises determining, based on the matched virtual mask, for at least one of the plurality of image surface points, a priority value indicative of a priority for acquiring a registration point at a position, relative to the patient's body, that corresponds to a position of the at least one image surface point relative to the skin surface defined in the image data. The method also includes triggering display of a visualization of the priority value, or of a visualization of information derived from the priority value.

Abstract: One embodiment of the present invention sets forth a technique for dynamically cropping image data transmitted to an endpoint device. The technique includes computing a first visual interest score for a first visual interest region within a digital image based on content included in the first visual interest region, computing a second visual interest score for a second visual interest region within the digital image based on content included in the second visual interest region, and determining that the first visual interest region is preferred over the second visual interest region based on the first visual interest score and the second visual interest score. The technique further includes setting a location within the first visual interest region as a point of visual interest and transmitting the digital image and the location of the point of visual interest to a computing device for displaying a portion of the digital image that includes the point of visual interest.

Abstract: A method for processing video data, comprising: receiving raw video data, representative of a plurality of frames; detecting, using the raw video data, one or more regions of interest in a detection frame that belongs to the plurality of frames, for example using a region proposal network; performing a cropping process on a portion of the raw video data representative of the detection frame, based on the regions of interest, so as to generate cropped raw video data; performing image processing on the cropped raw video data, including demosaicing, so as to generate processed image data for the detection frame; and analyzing the processed image data, for example using an object detection process, to determine information relating to at least one of said one or more regions of interest.

Abstract: A system and method is disclosed for requesting an updated location, including a point of interest, from a user for display on an interactive map. On receiving from an interested user a request for a point of interest associated with a subscribed user's current location, a request is transmitted to the subscribed user to check-in to the point of interest. In response to the request, the subscribed user selects a selected point of interest from a selection list. A representation of the subscribed user and the selected point of interest is then provided for display on an interactive map associated with the interested user.

Abstract: Embodiments are directed to a smart camera device that analyzes independent video streams.

Abstract: Disclosed embodiments may include a method that includes setting an influence level for each index that a neural network can accept in one sample to a same level for a neural network, receiving a training corpus including training input samples and a corresponding correct training prediction samples, generating, using the neural network, prediction samples, identifying an accuracy for each index by comparing the prediction samples with the corresponding correct training prediction samples, adjusting the influence level for each index based on the accuracy for each index, identifying one or more poorly accurate indexes for the neural network, receiving a first input sample including one or more characters, generating one or more normalized first input samples by applying one or more buffers to the one or more poorly accurate indexes, and generating, using the neural network, a categorization of each character in the one or more normalized first input samples.

Abstract: A processor-implemented neural network method includes: determining a respective probability density function (PDF) of normalizing a statistical distribution of parameter values, for each channel of each of a plurality of feature maps of a pre-trained neural network; determining, for each channel, a corresponding first quantization range for performing quantization of corresponding parameter values, based on a quantization error and a quantization noise of the respective determined PDF; determining, for each channel, a corresponding second quantization range, based on a signal-to-quantization noise ratio (SQNR) of the respective determined PDF; correcting, for each channel, the corresponding first quantization range based on the corresponding second quantization range; and generating a quantized neural network, based on the corrected first quantization range corresponding for each channel.

Abstract: A computing system for a vehicle may include processors to execute instructions to receive, by a middleware software layer of the computing system, messages published by one or more of a plurality of autonomous processing modules, and determine, by the middleware software layer, and based on the published messages, whether a similarity comparison of a first vehicle parameter and a second vehicle parameter satisfy a similarity threshold value. The first vehicle parameter and the second vehicle parameter relate to the messages published by the one or more autonomous processing modules. In response to determining that the similarity comparison of the first vehicle parameter and the second vehicle parameter does not satisfy the similarity threshold value, provide, by the middleware software layer, an instruction to a control system to cause the vehicle to perform an action, and cause, by the control system, the vehicle to perform the action.

Abstract: An unsupervised machine learning method with self-supervision losses improves a slice-wise spatial resolution of 3D medical images with thick slices, and does not require high resolution images as the ground truth for training. The method utilizes information from high-resolution dimensions to increase a resolution of another desired dimension.

Abstract: Methods, systems, and articles of manufacture, including computer program products, are provided for image cleanup. In some embodiments, there is provide a method which may include subsampling a first image to a first level image of a multiscale transform; performing, based on a machine learning model, an identification of a foreground portion of the first level image and a background portion of the first level image; generating, based on the identification of the foreground portion and the background portion, a first mask; upscaling the first mask to a resolution corresponding to the first image depicting the foreground item; applying the upscaled first mask to the first image to form a second image depicting the foreground item; and providing the second image depicting the foreground item to a publication system. Related systems and articles of manufacture, including computer program products, are also provided.

Abstract: In some examples, an autonomous robotic welding system comprises a workspace including a part having a seam, a sensor configured to capture multiple images within the workspace, a robot configured to lay weld along the seam, and a controller. The controller is configured to identify the seam on the part in the workspace based on the multiple images, plan a path for the robot to follow when welding the seam, the path including multiple different configurations of the robot, and instruct the robot to weld the seam according to the planned path.

Abstract: Systems and methods for detecting the cracks in illuminated electronic device screens are disclosed. In one embodiment, the method includes receiving an image of an electronic device screen and retrieving a plurality of kernels, each having values corresponding to a line region and a non-line region, with the orientation of the line region and the non-line region differing for each kernel. At least some of the kernels are applied to the image to obtain, at various locations of the image, values corresponding to the line regions and the non-line regions. Based on the values corresponding to the line regions and the non-line regions, cracks are automatically identified in the electronic device screen.

Abstract: The present invention provides a method for determining a plane, a method for displaying Augmented Reality (AR) display information and corresponding devices. The method comprises the steps of: performing region segmentation and depth estimation on multimedia information; determining, according to the result of region segmentation and the result of depth estimation, 3D plane information of the multimedia information; and, displaying AR display information according to the 3D plane information corresponding to the multimedia information. With the method for determining a plane, the method for displaying AR display information and the corresponding devices provided by the present invention, virtual display information can be added into a 3D plane, the reality of the display effect of enhanced display can be improved, and the user experience can be improved.

Abstract: A computer-implemented method for 3D reconstruction including obtaining 2D images and, for each 2D image, camera parameters which define a perspective projection. The 2D images all represent a same real object. The real object is fixed. The method also includes obtaining, for each 2D image, a smooth map. The smooth map has pixel values, and each pixel value represents a measurement of contour presence. The method also includes determining a 3D modeled object that represents the real object. The determining iteratively optimizes energy. The energy rewards, for each smooth map, projections of silhouette vertices of the 3D modeled object having pixel values representing a high measurement of contour presence. This forms an improved solution for 3D reconstruction.

Abstract: A method for foreground and background segmentation related to a virtual three-dimensional (3D) video conference, the method may include segmenting each image of multiple images of a video stream, to segments, each segment has one or more properties that are constant; determining temporal properties of the segments; and classifying each segment as a background segment or a foreground segment, based at least in part, on the temporal properties of the segments.

Abstract: Method for cutting a three-dimensional model of a dental scene, or “scene model.” The method includes acquiring a view of the scene model, called the “analysis view.” The method includes analyzing the analysis view by a neural network in order to identify, in the analysis view, at least one elementary zone representing an element of the dental scene, and assigning a value to at least one attribute of the elementary zone. The method includes identifying a region of the scene model represented by the elementary zone on the analysis view, and assigning, in the region, a value to an attribute of the scene model in accordance with the value of the attribute of the elementary zone.

Abstract: Embodiments may: select a set of training images; extract a first set of features from each training image of the set of training images to generate a first feature tensor for each training image; extract a second set of features from each training image to generate a second feature tensor for each training image; reduce a dimensionality of each first feature tensor to generate a first modified feature tensor for each training image; reduce a dimensionality of each second feature tensor to generate a second modified feature tensor for each training image; construct a first generative model representing the first set of features and a second generative model representing the second set of features of the set of training images; identify a first candidate image; and apply a regression algorithm to the first candidate image and each of the first generative model and the second generative model to determine whether the first candidate image is similar to the set of training images.

Abstract: Systems and methods for roof condition assessment from digital images using machine learning are disclosed, including receiving an image of a structure having roof characteristic(s), first pixel values depicting the structure, second pixel values outside of the structure depicting a background surrounding the structure, and first geolocation data; generating a synthetic shape image of the structure from the image using machine learning, including pixel values forming a synthetic outline shape, and having second geolocation data; mapping the synthetic shape onto the image, based on the first and second geolocation data, and changing the second pixel values so as to not depict the background; assessing roof characteristic(s) based on the first pixel values with a second machine learning algorithm resulting in a plurality of probabilities, each for a respective roof condition classification category, and determining a composite probability based upon the plurality of probabilities so as to classify the roof char

Abstract: Examples are described of segmenting an image into image regions based on depicted categories of objects, and for refining the image regions semantically. For example, a system can determine that a first image region in an image depicts a first category of object. The system can generate a color distance map of the first image region that maps color distance values to each pixel in the first image region. A color distance value quantifies a difference between a color value of a pixel in the first image region and a color value of a sample pixel in a second image region in the image. The system can process the image based on a refined variant of the first image region that is refined based on the color distance map, for instance by removing pixels from the first image region whose color distances fall below a color distance threshold.

Abstract: A method for intelligent control of an elevator to give priority to a floor containing the total highest priority of passengers acquires first images taken by first cameras, identifies targets in the first images, and determines priority levels of the targets. In response to receiving a plurality of first service signals from a number of floors in a building, the method controls a driving unit to drive an elevator to move to service the passengers as targets with a highest total priority level according to individual priority levels of the targets within a group of passengers waiting on each floor of the building.

Abstract: A method of real-time image correction of a video stream from a surgical endoscope of a surgical robotic system. The video stream comprises a sequence of images. The method comprises: for an image in the sequence of images, identifying a plurality of regions, each region having a different range of values of at least one image property to another region; applying a mask to the image which applies a region-specific modification to each region, each region-specific modification modifying the at least one image property of that region; deriving data from the surgical robotic system; determining a relationship between features in the image from the derived data; modifying the mask for a predicted relationship between the features at the time of a subsequent image in the video stream; and applying the modified mask to the subsequent image in the video stream.

Abstract: A three-dimensional shape data processing apparatus includes a processor configured to divide a three-dimensional space including a three-dimensional shape constituted by one or more forming surfaces, which are flat surfaces or curved surfaces or include both flat surfaces and curved surfaces, into plural three-dimensional regions each having a predetermined size, specify, for each of the three-dimensional regions, one or more of the forming surfaces that interfere(s) with the three-dimensional region, and redivide each of the three-dimensional regions until a size of the three-dimensional region reaches a size necessary for reproduction of a property of a portion of the three-dimensional shape expressed by one or more of the forming surfaces that interfere(s) with the three-dimensional region.

Abstract: Aspects of the present disclosure include methods for processing and displaying multi-channel spectral histograms. Methods according to certain embodiments include obtaining a histogram of cytometric data for a sample, wherein the cytometric data comprises measurements from particles irradiated in the sample flowing in a flow stream, generating the representation of the histogram by encoding the histogram, the encoding comprising: assigning a color to each histogram value in the representation of the histogram, and duplicating each color corresponding to each histogram value in the representation of the histogram a predetermined number of times, and using the representation of the histogram for displaying the histogram. Systems for practicing the subject methods are also provided. Non-transitory computer readable storage mediums are also described.

Abstract: The present disclosure provides a copy area identification method and device, which relates to the technical field of image processing. A copy area identification method according to the present disclosure includes: extracting feature information of multiple layers from an image to be processed; encoding the feature of the multiple layers information respectively, and jointly decoding according to encoded information of the multiple layers to acquire a jointly decoded output; acquiring information of pixels according to the jointly decoded output, wherein the information of the pixels comprises rotation angle information of the copy area, and a distance between each of the pixels and a border of the copy area; and determining a position of a border of the copy area according to the information of the pixels.

Abstract: Systems and methods for color replacement are described. Embodiments of the disclosure include a color replacement system that adjusts an image based on a user-input source color and target color. For example, the source color may be replaced with the target color throughout the entire image. In some embodiments, a user provides a speech or text input that identifies a source color to be replaced. The user may then provide a speech or text input identifying the target color, replacing the source color. A color replacement system creates and embedding of the source color, segments the image based on the source color embedding, and then replaces the color of segmented portion of the image with the target color.

Abstract: Detecting multiple people in a room includes detecting a point cloud corresponding to at least one user moving in the room, forming a bounding box corresponding to coordinates of points of the point cloud, and determining if the point cloud corresponds to multiple people based on a size of the point cloud, a presence of separate clusters of points in the point cloud, and/or detecting one or more people entering or leaving the room. Detecting a point cloud may include using a tracking device. The size of the point cloud may be compared with a size of a point cloud corresponding to a single user of the room. Detecting one or more people entering or leaving the room may include previously detecting entrances of the room by tracking movement of people in the room and determining a location where movement results in more or less people in the room.

Abstract: Computer technology for use when one or more devices is receiving low resolution video typically due to low internet bandwidth. Multiple devices are chosen for a cluster, with the devices of the cluster being in device to device communication (see definition of “device to device communication” herein). One, or more, of the devices in the cluster is set up to do GAN image enhancement locally for the cluster, and the low resolution video stream is sent to the GAN enhancement device(s) and then enhanced to be sent to other device(s) in the cluster as a higher resolution, clearer video image by device to device communication.

Abstract: Systems and methods are disclosed for controlling image annotation. One method includes acquiring a digital representation of image data and generating a set of image annotations for the digital representation of the image data. The method also may include determining an association between members of the set of image annotations and generating one or more groups of members based on the association. A representative annotation from the one or more groups may also be determined, presented for selection, and the selection may be recorded in memory.

Abstract: A map segmentation method and device, a motion estimation method, and a device terminal. Obtaining all map point data and constructing a corresponding graph based on these map points, where the graph includes each node formed by map points and corresponding edges; measuring and estimating to-be-estimated state variables corresponding to the node to obtain a corresponding measured and estimated value, where the to-be-estimated state variables include homogeneous coordinates of the node and a corresponding sensor pose upon node obtaining; calculating an inter-point correlation between any two adjacent nodes in the graph based on the measured and estimated value to determine whether the inter-point correlation is less than a preset correlation threshold; and when the inter-point correlation is less than the preset correlation threshold, removing an edge corresponding to the corresponding adjacent nodes, to separate the static points and the dynamic points and complete map segmentation.

Abstract: A device obtains image data associated with a document. Using a first machine learning model, the device determines, for the document, a first classification of one of a plurality of document types and a first confidence score associated with the first classification, and a second classification of one of the plurality of document types and a second confidence score associated with the second classification based on the image data. The device determines a difference between the first confidence score and the second confidence score, compares the difference and a threshold value, and accept the first classification of the document when the difference satisfies the threshold value.

Abstract: The image processing device acquires feature quantities (maximum value, minimum value, average value, histogram, etc.) of the entire area (GA) of the image and feature quantities of each local area (LA) of the image from the input image, and calculates a plurality of modulation gain values (gamma correction curves) for GA and each LA. Furthermore, the image processing device determines the correction intensity for each LA from the feature quantity of the GA and the feature quantity of each LA, and creates the LA correction intensity map. Finally, the image processing device finally applies the result of combining a plurality of modulation gain values based on the LA correction intensity map to the input image.

Abstract: A system for disparity estimation includes one or more feature extractor modules configured to extract one or more feature maps from one or more input images; and one or more semantic information modules connected at one or more outputs of the one or more feature extractor modules, wherein the one or more semantic information modules are configured to generate one or more foreground semantic information to be provided to the one or more feature extractor modules for disparity estimation at a next training epoch.

Abstract: Methods, systems, and media for secure authentication of users using one or more biometric recognition systems are provided. In some embodiments, the method comprises: receiving an indication that a biometric identifier is to be used to authenticate a user to a service; receiving (i) the biometric identifier of the user from a capture device and (ii) knowledge-based secondary information associated with the user from an input device; determining a Voronoi cell identifier that corresponds to the biometric identifier; calculating a hash of the Voronoi cell identifier and the knowledge-based secondary information; transmitting the hash to a server device for verification; in response to transmitting the hash to the server device, receiving a response indicating whether the hash matches a previously stored hash that was stored in the server device; and determining whether to automatically authenticate the user to the service based on the response from the server device.

Abstract: An image processing apparatus includes an input unit that inputs an image, and a processor configured to read out a program stored in a memory, and executes the program. The processor is configured to detect an intended subject from the input image by a first detection method, set an intended subject region for the detected intended subject, detect the intended subject from the input image by a second detection method different from the first detection method, and update the set intended subject region by using a detection result of the second detection method.

Abstract: Method, computer readable medium, and apparatus of recognizing character zone in a digital document. In an embodiment, the method includes classifying a segment of the digital document as including text, calculating at least one parameter value associated with the classified segment of the digital document, determining, based on the calculated at least one parameter value, a zonal parameter value, classifying the segment of the digital document as a handwritten text zone or as a printed text zone based on the determined zonal parameter value and a threshold value, the threshold value being based on a selection of an intersection of a handwritten text distribution profile and a printed text distribution profile, each of the handwritten text distribution profile and the printed text distribution profile being associated with a zonal parameter corresponding to the determined zonal parameter value, and generating, based on the classifying, a modified version of the digital document.

Abstract: Disclosed are systems, methods, and non-transitory computer-readable media for automatic image selection for online product catalogs. An image selection system gathers feature data for images of an item included in listings posted to an online marketplace. The image selection system uses the feature data as input in a machine learning model to determine probability scores indicating an estimated probability that each image is suitable to represent the item. The machine learning model is trained based on a set of training images of the item that have been labeled to indicate whether they are suitable to represent the image. The image selection system compares the probability scores and selects an image to represent the item as a stock image based on the comparison.

Abstract: A system includes a probe configured to transmit ultrasound signals directed to a target blood vessel and receive echo information associated with the transmitted ultrasound signals. The system also includes at least one processing device configured to process the received echo information and generating a three-dimensional ultrasound image of the target blood vessel; obtain a three-dimensional vascular model corresponding to the target blood vessel; identify a best-fit of the three-dimensional vascular model onto the three-dimensional target image; store the best fit of the three-dimensional vascular model as a segmentation result; and calculate, based on the segmentation result, measurements for the target blood vessel.

Abstract: The present disclosure relates to a method for generating a structure of a PDF-document, wherein the PDF-document comprises elements. The method comprises detecting document cells of the PDF-document dependent on commands of a page description language for printing the elements of the PDF-document. The method comprises determining parts of the PDF-document dependent on the PDF-document by a machine learning module. The determining of the respective part comprises associating a respective portion of the elements of the PDF-document with the respective part. Furthermore, a respective label may be assigned to the respective part. The method may further comprise using a symbolic artificial intelligence module, wherein rules of the symbolic AI-module for reconciling the document cells with the parts may be applied. The elements of the structure of the PDF-document may be generated and labelled dependent on a result of the reconciling and dependent on the respective label to the respective part.

Abstract: Disclosed are systems, methods, and computer-readable storage media to perform an interactive image recolorization process. The method includes receiving user input including a stroke drawn on an image presented on a client device. The stroke comprises a user-specified color. The method further includes determining a region of interest in the image. The method further includes recolorizing the region of interest on the image based on the user-specified color and causing presentation of a result of the recolorization on the client device.

Abstract: A fluorescence image registration method includes obtaining at least one fluorescence image of a biochip. An interior local area. Sums of pixel values in the interior local area along a first direction and a second direction are obtained. A plurality of first template lines is selected to find a minimum total value of the sums of pixel values corresponding to the first template lines. Pixel-level correction is performed on a local area of the track line to obtain a pixel-level track cross. Other track crosses on the biochip is obtained, and the pixel-level correction is performed on the other track crosses. The position of the pixel-level track line is corrected by a center-of-gravity method to obtain the subpixel-level position of the track line. The subpixel-level positions of all sites uniformly distributed on the biochip is obtained.

Abstract: Automated systems and methods for training a multilayer neural network to classify cells and regions from a set of training images are presented. Automated systems and methods for using a trained multilayer neural network to classify cells within an unlabeled image are also presented.

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: A display system is disclosed. The display system includes a display, an eye-tracking sensor, and a processor communicatively coupled to the display and the eye-tracking sensor. The computing device receives data from the eye-tracking sensor, determines, based on the eye-tracking data, whether a user's eye is looking at a display in the display system, and based on whether the eye is determined to be looking at the display, controls a power consumption of the display.

Abstract: A method, system and computer program product for reducing learning time for a newly installed camera is disclosed. The method includes generating a new unusual activity model for the newly installed camera, based on portion(s) of existing and established unusual activity model(s) built for different older camera(s), where the existing and established unusual activity model(s) relate to at least one same static object appearing within Fields Of Views (FOVs) of the new and older cameras.

Abstract: Methods, systems, and apparatus, including medium-encoded computer program products, for computer aided design of physical structures using generative design processes. A method includes obtaining a design space for a modeled object, one or more design criteria for the modeled object, and one or more in-use load cases; iteratively modifying a generatively designed three dimensional shape of the modeled object in the design space in accordance with the one or more design criteria and the one or more in-use load cases for the physical structure, comprising: performing numerical simulation of the modeled object in accordance the one or more in-use load cases, computing shape change velocities for an implicit surface in a level-set representation of the three dimensional shape, changing the shape change velocities in accordance with a polynomial function, and updating the level-set representation using the shape change velocities to produce an updated version of the three dimensional shape.

Abstract: Disclosed are systems, devices and methods for imaging and image-based sorting of particles in a flow system, including cells in a flow cytometer. In some aspects, a system includes a particle flow device to flow particles through a channel, an imaging system to obtain image data of a particle during flow through the channel, a processing unit to determine a property associated with the particle and to produce a control command for sorting the particle based on sorting criteria associated with particle properties, and an actuator to direct the particle into one of a plurality of output paths of the particle flow device in real-time.

Abstract: A personality model is created for a population and used as an input to a text generation system. Alternative texts are created based upon the emotional effect of the generated text. Certain words or phrases are “pinned” in the output, reducing the variability of the generated text so as to preserve required information content, and a number of tests provide input to a discriminator network so that proposed outputs both match an outside objective regarding the information content, emotional affect, and grammatical acceptability. A feedback loop provides new “ground truth” data points for refining the personality model and associated generated text.

Abstract: The present application provides a method and a device for estimating background motion of infrared image sequences and a storage medium. The method comprises: determining regions to be matched according to the maximum displacement of two adjacent frames of images in background motion, dividing the region to be matched into T sub-regions, and generating M groups of sub-window sets in the sub-regions randomly, and computing the background motion displacement corresponding to these sub-window sets by synchronous mutual information matching, then calculating the background motion displacement between two adjacent frames of images by a potential function voting method according to the background motion displacement corresponding to these sub-window sets. The method in the present application is based on image block matching, and does not depend on feature point extraction, so it has high robustness for background motion estimation of infrared image sequences with few feature points and noise interference.

Abstract: A method for adjusting/monitoring an orthodontic appliance worn by a patient. Creating a numerical three-dimensional reference model of at least a part of a patient's arch. Creating a numerical three-dimensional reference model of the teeth of the at least a part of the patient's arch. Acquiring at least one updated image, under actual acquisition conditions, and searching, by deformation of the initial reference model, for an updated reference model corresponding to the positioning of the teeth during the acquisition of the updated image. Determining, depending a comparison model corresponding to an anticipated positioning of the teeth, a comparison model. Comparing the updated reference model with the comparison model. If comparison leads to an unsatisfactory result, creation of a replacement orthodontic appliance, by modification of the orthodontic appliance worn or by creation of a new orthodontic appliance, the replacement orthodontic appliance being configured accordingly.

Abstract: A crop generation system determines multiple types of saliency data and multiple crop candidates for an image. Multiple region of interest (“ROI”) ensembles are generated, indicating locations of the salient content of the image. For each crop candidate, the crop generation system calculates an evaluation score. A set of crop candidates is selected based on the evaluation scores.