Abstract: In an information display system, a wearable display displays information within a field of view of the user. Monitoring target information related to a monitoring target is acquired. Field-of-view information for identifying the field of view of a user is acquired. The monitoring target present within the field of view of the user is identified based on the acquired field-of-view information. A display control unit displays, within the field of view of the user, a symbol display that indicates that the monitoring target information is present when the monitoring target information regarding the monitoring target captured within the field of view of the user is present, and displays, within the field of view of the user, the monitoring target information corresponding to the symbol display when an aim display set within the field of view of the user overlaps the symbol display within the field of view of the user.

Abstract: An augmented reality (AR) computing device for displaying computer-generated elements in augmented reality supported service activities is provided. The AR computing device is configured to scan a first location, using an AR display device, to capture location data including consumer data corresponding to one or more consumers and facial data associated with the one or more consumers. The AR computing device is also configured to select, using the AR display device, a first consumer of the one or more consumers, store the consumer data corresponding to the first consumer in a database, and receive AR data associated with the location data. The AR computing device is further configured to automatically display, using the AR display device, the AR data, receive an update for the AR data, automatically update the AR data in response to the received update, and display, using the AR display device, the updated AR data.

Abstract: Systems and methods for estimating a layout of a room are disclosed. The room layout can comprise the location of a floor, one or more walls, and a ceiling. In one aspect, a neural network can analyze an image of a portion of a room to determine the room layout. The neural network can comprise a convolutional neural network having an encoder sub-network, a decoder sub-network, and a side sub-network. The neural network can determine a three-dimensional room layout using two-dimensional ordered keypoints associated with a room type. The room layout can be used in applications such as augmented or mixed reality, robotics, autonomous indoor navigation, etc.

Abstract: Systems and methods for overlaying video segments of actions of audience members with video segments of an event performer are described. A computer implemented method includes: identifying, by a computer device, an event performer in video content; identifying, by the computer device, an audience member in the video content that has a social network relationship to the event performer; correlating, by the computer device, an action of the event performer in the video content to an action of the audience member in the video content; and generating, by the computer device, a composite image comprising an image of the action of the event performer and an image of the action of the audience member.

Abstract: An action recognition system is provided that includes a device configured to capture a video sequence formed from a set of unlabeled testing video frames. The system further includes a processor configured to pre-train a recognition engine formed from a reference set of CNNs on a still image domain that includes labeled training still image frames. The processor adapts the recognition engine to a video domain to form an adapted engine, by applying non-reference CNNs to domains that include the still image and video domains and a degraded image domain that includes labeled synthetically degraded versions of the frames in the still image domain. The video domain includes random unlabeled training video frames. The processor recognizes, using the adapted engine, an action performed by at least one object in the sequence, and controls a device to perform a response action in response to an action type of the action.

Abstract: Systems and methods for creating a story board with forensic video analysis on a video repository are provided. Some methods can include storing a plurality of video data streams in a data repository, storing asynchronous streams of metadata of each of the plurality of video data streams in the data repository, identifying a first object captured by at least one of the plurality of video data streams, using the asynchronous streams of metadata to identify correlations or interactions between the first object and a plurality of other objects over time, and replicating a story of the first object.

Abstract: An information processing device includes: an information acquisition unit that acquires information on a target included in an image; and a transmission unit that determines a transmission destination of the image or information on the image according to the information on the target and transmits the image or the information on the image to the determined transmission destination.

Abstract: According to one embodiment, a processing apparatus includes a memory and a processor. The processor is electrically coupled to the memory and is configured to calculate a size of an object based on a distance map. The distance map is acquired together with an image at one image capture by a single imaging optical system. Information indicative of a distance to the object included in the image is mapped in the distance map.

Abstract: Systems and methods use cameras to provide autonomous navigation features. In one implementation, a driver-assist system is provided for a vehicle. The system may include one or more image capture devices configured to acquire images of an area forward of the vehicle. The system may also include at least one processing device configured to receive, via one or more data interfaces, the images. The at least one processing device may be further configured to analyze the images acquired by the one or more image capture devices and cause at least one navigational response in the vehicle based on monocular and/or stereo image analysis of the images.

Abstract: An image recording system includes: a memory in which an image picked up by a camera and an index value are stored in association with each other, the camera being equipped in a vehicle, the index value indicating a possibility of collision between the vehicle and an obstacle at a time when the image is picked up; and at least one processing circuit configured to perform a quality changing process of decreasing a quality of the image, based on the index value.

Abstract: An obstacle detection device includes a detector, a generator, and an estimator. The detector detects a person region where a shape of a person is present, from a captured image. The generator generates an optical flow based on a non-detection image from which the person region is not detected and a detection image from which the person region is detected before the time of the non-detection image in a case where the person region is not detected by the detector. The estimator estimates the person region in the non-detection image based on the optical flow generated by the generator.

Abstract: An image processing apparatus includes an extracting unit configured to obtain a range image having pixel values corresponding to a distance of an object in a plurality of captured images obtained by capturing the object on a road surface so as to extract pixel values from the range image, a first generating unit configured to generate vertical direction distribution data indicating a frequency distribution of distance values in a vertical direction of the range image based on the extracted pixel values, a calculating unit configured to calculate a height of the road surface in the range image before generation of the vertical direction distribution data is completed, and a second generating unit configured to generate horizontal direction distribution data indicating a frequency distribution of distance values in a horizontal direction of the range image based on the extracted pixel values and the calculated height of the road surface.

Abstract: A vehicle display system includes a display device, a determiner, and a display controller. The display device is configured to provide a display on a window of a vehicle. The determiner is configured to determine awakening levels of a driver and an occupant of the vehicle. The display controller is configured to control displaying of the display device, and change the display provided by the display device in accordance with the awakening levels of the driver and the occupant determined by the determiner.

Abstract: Systems and methods for generating authentic digital memorabilia are described. A signor may be provided a digital photograph. The signor's signature, written message, or voice message may be received. Biometric authentication or verification may be performed on the signor's handwriting or voice sample through comparison with stored samples. If the verification signifies a high likelihood signor's handwriting or voice sample is authentic, creation of digital memorabilia is performed by embedding signor's signature or written message in a digital photograph and linking the signor's voice message with the photograph. The digital memorabilia is accompanied by a certificate of authenticity and distributed to a customer or displayed on a website.

Abstract: Provided are an electronic device and operation method thereof. The electronic device may include: a display having a biometric sensing region; a biometric sensor disposed in the biometric sensing region; and a processor.

Abstract: A method of host-directed illumination for verifying the validity of biometric data of a user is provided that includes capturing biometric data from a user with an authentication device during authentication and directing illumination of the biometric data from a host authentication system during the capturing operation. Moreover, the method includes comparing illumination characteristics of the captured biometric data against illumination characteristics expected to result from the directing operation, and determining that the user is a live user when the illumination characteristics of the captured biometric data match the illumination characteristics expected to result from the directing operation.

Abstract: A deep learning object detection and recognition system contains a number of cellular neural networks (CNN) based integrated circuits (ICs) operatively coupling together via the network bus. The system is configured for detecting and then recognizing one or more objects out of a two-dimensional (2-D) imagery data. The 2-D imagery data is divided into N set of distinct sub-regions in accordance with respective N partition schemes. CNN based ICs are dynamically allocated for extracting features out of each sub-region for detecting and then recognizing an object potentially contained therein. Any two of the N sets of sub-regions overlap each other. N is a positive integer. Object detection is achieved with a two-category classification using a deep learning model based on approximated fully-connected layers, while object recognition is performed using a local database storing feature vectors of known objects.

Abstract: A preferred image-recognition processor comprises a plurality of storage-processing units (SPU), with each SPU comprising at least a three-dimensional memory (3D-M) array vertically stacked above a pattern-processing circuit. The 3D-M array stores at least a portion of an image model from an image model database. The image data from the input are sent to all SPUs, which perform pattern recognition simultaneously.

Abstract: A data collection device includes a camera and a processing arrangement configured to capture at least two images of a medicament dose indicator of a medicament delivery device in each of the at least two images the processing arrangement identifies a displayed dosage using a first character identification technique, in each of the at least two images, determines whether the displayed dosages in the images are different and, if not, identifies at least one character in the images using a second character identification technique and determines the medicament dosage amount indicated by the medicament dose indicator based thereon. The first character recognition technique may be, for example, an optical pattern recognition or correlation, may be less computationally intensive than the second character recognition technique, for example, an optical character recognition (OCR). In another embodiment, the first and second character recognition techniques are different OCR processes.

Abstract: A method for determining at least one property related to at least part of a real environment comprises receiving a first image of a first part of a real environment captured by a first camera, wherein the first camera is a thermal camera and the first image is a thermal image and the first part of the real environment is a first environment part, providing at least one description related to at least one class of real objects, wherein the at least one description includes at least one thermal property related to the at least one class of real objects, receiving a second image of the first environment part and of a second part of the real environment captured by a second camera, wherein the second part of the real environment is a second environment part, providing an image alignment between the first image and the second image, determining, for at least one second image region contained in the second image, at least one second probability according to the image alignment, pixel information of the first image

Abstract: A system and methods are disclosed for enhancing license plate recognition (LPR) and vehicle feature recognition processes in automatic vehicle access control, parking management, automatic toll collection and security applications. The system uses optical character recognition (OCR) to read license plates, while utilizing image feature recognition to verify plate reading results, and correct any OCR read errors, thereby increasing system accuracy. The system automatically controls the actuation of one or a plurality of gates/barriers to allow entry and exit of authorized vehicles to or from a premises, a parking lot or a toll station.

Abstract: A method for collecting and processing remotely sensed imagery in order to achieve precise spatial co-registration (e.g., matched alignment) between multi-temporal image sets is presented. Such precise alignment or spatial co-registration of imagery can be used for change detection, image fusion, and temporal analysis/modeling. Further, images collected in this manner may be further processed in such a way that image frames or line arrays from corresponding photo stations are matched, co-aligned and if desired merged into a single image and/or subjected to the same processing sequence. A second methodology for automated detection of moving objects within a scene using a time series of remotely sensed imagery is also presented. Specialized image collection and preprocessing procedures are utilized to obtain precise spatial co-registration (image registration) between multitemporal image frame sets.

Abstract: Disclosed is a method and system for generating redacted video of a scene captured using a video camera and augmenting a manual review process of the redacted video of the scene. According to an exemplary embodiment, a video of the scene is redacted via probability-based analysis to detect and obscure privacy sensitive objects included in the captured video. A manual review process of the generated redacted video is augmented to use the object detection probability to enhance the reviewing video frame rate to expedite the manual review process.

Abstract: A considerable amount of time was required for the completion of an objectization process on image data. A time predictor in a processing device predicts the time required for each of a local processing device and remote processing devices to convert image data to objectized data. A device selector selects, based on the predicted time, the local processing device or one of the remote processing devices as a processing device that acquires the objectized data derived from the image data within a short period of time, and instructs the selected processing device to perform the objectization process. A data manager manages the image data that is to be transmitted from a data transceiver to the selected processing device, and the objectized data that is to be received by the data transceiver from the selected processing device.

Abstract: An image content commenting method and a related image management system can be utilized to establish a commenting record about partial content of an image stream. The image content commenting method includes reading a coordinate value of a cursor within a first image frame of the image stream, selecting a pattern object inside the first image frame according to the coordinate value, establishing connection between a first bookmark and the pattern object, and identifying the pattern object from a second image frame of the image stream and then connecting the first bookmark with the pattern object inside the second image frame.

Abstract: Examples disclosed herein relate, among other things, to wirelessly detecting an identifier of a network device connected to a network, accessing the network to obtain a network address associated with the network device based on the identifier, and communicating with the network device through the network using the network address.

Abstract: An apparatus for contextual execution comprises a processor, and a memory containing instructions, which when executed by the processor, cause the apparatus to receive, from a user terminal, a control input associated with an intent, obtain location data associated with a location of the user terminal, and determine a scored set of execution options associated with the control input. Further, the instructions, when executed by the processor cause the apparatus to obtain a contextual label associated with the location data, the label determined based on the application of one or more adapted pretrained deep learning models to the location data.

Abstract: A camera system including: an image sensor that is controlled by one or more camera parameters; and a pre-processing circuit adapted to associate weights respectively with cells of a grid, wherein the weights differ from the one or more camera parameters, wherein the cells respectively include a plurality of contiguous picture elements of the image sensor.

Abstract: A polarized image acquisition unit 20 acquires polarized images in a plurality of polarization directions. A reflection information generation unit 30 generates reflection information indicating reflection components from the polarized images in the plurality of polarization directions acquired by the polarized image acquisition unit 20. A reflection information using unit 40 uses the reflection information generated by the reflection information generation unit 30 to acquire an image of a viewed object appearing in the polarized images. A depth estimation unit estimates a depth value of a reflective surface area and acquires a position of the viewed object on the basis of an image of the viewed object appearing in the reflective surface area and the estimated depth value. Therefore, the viewed object positioned in, for example, an area of a blind spot can be easily checked.

Abstract: A system and method for identifying real time change in a scene of an unknown environment using an unmanned vehicle. In the context of unmanned vehicle navigation, it is critical to calculate the saliency map in real time and employ them in scene understanding. This will reduce the search space and ensure that the process is quicker. A domain specific ontology is created and a saliency model is developed. The saliency model detects key domain specific regions of interest in two consecutive images. The regions of interest is used for registration and change detection. The system is detecting the change by using visual saliency as an abstract feature that is developed in the environment. Probability of change is derived using the salient maps of the two images.

Abstract: There is provided an image processing device including a motion vector detection portion that performs comparison of a substantially spherical photographic subject such that, among a plurality of captured images including the photographic subject, an image as a processing target and another image as a comparison target are compared using each of the plurality of captured images as the processing target, and which detects a motion vector of a whole three-dimensional spherical model with respect to the processing target, a motion compensation portion that performs motion compensation on the processing target, based on the motion vector of each of the plurality of captured images that is detected by the motion vector detection portion, and a synthesis portion that synthesizes each of the captured images that are obtained as a result of the motion compensation performed by the motion compensation portion.

Abstract: A system that compares the images submitted with a preprocessed database containing pictures, drawings, and patent drawings, among other media. The images are interrelated by comparing the content of the patent images, the narrative in the patents with the other visual media which may or may not be pre-tagged.

Abstract: In one respect, there is provided a system that may include a processor and a memory. The memory may be configured to store instructions that results in operations when executed by the processor. The operations may include: generating a concatenated feature map set by at least combining a first feature map set and a second feature map set, the first feature map set and the second feature map set each indicating one or more occurrences of a feature within an image, the first feature map set having a different scale than the second feature map set; and classifying, by a convolutional neural network, an image, the classifying of the image being based at least on the concatenated feature map set. Related methods and articles of manufacture are also disclosed.

Abstract: The application discloses a method and an apparatus for recognizing RGB-D objects based on adaptive similarity measure of dense matching item, wherein the method can include at least the following steps: convolution neural network features of a to-be-queried object and a reference object are extracted; dense matching is carried out on the reference object and the to-be-queried object on the basis of the convolution neural network features fused with RGB and depth information; similarity between the reference object and the to-be-queried object is measured according to a dense matching result; and the to-be-queried object is classified based on the similarity between the reference object and the to-be-queried object. With the embodiments of the present application, at least in part, the technical problem of how to improve the robustness of object recognition is solved.

Abstract: To allow identifying or estimating a vehicle model of a vehicle even in a case where it is impossible to identify the vehicle model on the basis of a feature amount of at least a portion of a vehicle body of the vehicle, a vehicle model identification device (2) of the present disclosure includes a storage unit (14) that stores a first database in which a feature amount of at least a portion of a vehicle body of each of a plurality of vehicle models, for each of the vehicle models is stored, and a second database in which a feature amount of a component which is included in at least the portion of the vehicle body of each of the vehicle models is stored, and a processor (15) that performs a first vehicle model identification process of identify the vehicle model of the vehicle with reference to the first database on the basis of the feature amount of at least the portion of the vehicle body which is extracted from a captured image, and a second vehicle model identification process of identifying the vehicle m

Abstract: Systems and methods for identifying semantically and/or visually related information among a set of content items, such content items that include similar concepts or that have similar visual aspects, are disclosed. The disclosed techniques provide tools for identifying related information among various content items, such as text pages and documents, presentation slides and slide decks, etc. The disclosed techniques provide improved methods for searching among content items, organizing content items into categories, and pruning redundant content. Furthermore, the disclosed techniques provide improvements to computation of various metrics, including usage, performance, and impact metrics.

Abstract: A method for segmenting a projected pattern in an image recorded by a camera includes recording, by a camera in a learning phase, a multiplicity of images produced by virtue of a light source projecting the pattern from a plurality of different angles onto a projection surface in a clean room, wherein the projection surface has a plurality of respectively different distances from the light source for each angle; transforming the multiplicity of images into a frequency domain representation; obtaining a value range of occurring frequencies from the frequency domain representation of the multiplicity of images; and masking, in an application phase, frequencies other than the frequencies lying in the value range in a frequency domain representation of the image recorded by the camera, wherein a difference image produced in this manner is transformed back from the frequency domain representation.

Abstract: In one respect, there is provided a system that may include a processor and a memory. The memory may be configured to store instructions that results in operations when executed by the processor. The operations may include: training a machine learning model by at least processing a training set with the machine learning model, the training set including at least one synthetic image that is generated by applying one or more modifications to a non-synthetic image; determining, based at least on a result of the processing of the mixed training set, that the machine learning model is unable to classify images having a specific modification; and training the machine learning model with additional training data that includes one or more additional synthetic images having the specific modification.. Related methods and articles of manufacture are also disclosed.

Abstract: A keypoint detection system includes: a camera system including at least one camera; and a processor and memory, the processor and memory being configured to: receive an image captured by the camera system; compute a plurality of keypoints in the image using a convolutional neural network including: a first layer implementing a first convolutional kernel; a second layer implementing a second convolutional kernel; an output layer; and a plurality of connections between the first layer and the second layer and between the second layer and the output layer, each of the connections having a corresponding weight stored in the memory; and output the plurality of keypoints of the image computed by the convolutional neural network.

Abstract: A method is provided of forming a neural network for detecting surface defects in aircraft engine components. The method includes: providing (i) a pre-trained deep learning network and (ii) a learning machine network; providing a set of pixelated training images of aircraft engine components exhibiting examples of different classes of surface defect; training the trainable weights of the learning machine network on the set of training images.

Abstract: Techniques are described for using artificial intelligence (i.e., machine learning) to identify a system problem. Multiple machine learning stages, or models, are used to better categorize inputs and more quickly derive a reliable solution. Each subsequent model identifies a more specific category than a previous model, i.e., each successive model operates on a more granular level. The described techniques include staged machine learning models and user interface elements that are used to train a system and support an application development process. These techniques can be used to more easily create logic that is directed to solving a problem in a particular system.

Abstract: The present embodiments relate generally to ultrasound imaging methods, systems, and apparatus that identify an imaged needle in an ultrasound image. The embodiments involve performing edge detection on the ultrasound image to generate an edge-detected data set corresponding to the ultrasound image; performing a straight line detection operation on the edge-detected data set to detect one or more straight lines; based on the one or more straight lines detected in the edge-detected data set, identifying a region of interest (ROI) on the ultrasound image; transforming the ROI of the ultrasound image from a spatial domain to an analysis domain, to generate a transformed ROI; analyzing the transformed ROI to determine whether the transformed ROI, in the analysis domain, corresponds to a needle signature; and if the transformed ROI corresponds to the needle signature, identifying the detected one or more straight lines as the imaged needle.

Abstract: An automatic classifying system in a dining environment includes a picture uploading component implemented in an electronic device for transmitting a set of pictures via the Internet, and a server for directly or indirectly receiving the set of pictures. The server has a picture analysis component for classifying one of the pictures according to at least two classifications and generating an analysis result to a web-platform system so as to display the picture and the analysis result.

Abstract: A method and apparatus for processing image data is provided. The method includes the steps of employing a main processing network for classifying one or more features of the image data, employing a monitor processing network for determining one or more confusing classifications of the image data, and spawning a specialist processing network to process image data associated with the one or more confusing classifications.

Abstract: An information processing device include: a first data acquirer that acquires first observational data; a second data acquirer that acquires second observational data; and a processor that performs machine learning by semi-supervised learning to output an emotion of a person as a latent factor using training data of the first observational data and the second observational data, and that infers and outputs the emotion from at least one of the first observational data and the second observational data using a trained model.

Abstract: A non-transitory recording medium stores a program causing a computer to execute a process including determining a reference value on a decision surface classifying a group of image data represented by a plurality of types of feature quantities into two classes in a feature quantity space, based on a feature quantity of given image data that is closest to the decision surface and is classified into one of the classes of the group of image data in the feature quantity space, plotting and displaying pieces of image data of the group of image data along at least one axis indicating a degree of separation of each of the pieces of image data relative to the reference value, modifying, upon receiving a single piece of image data identified from the plotted and displayed pieces of image data, the decision surface based on the identified single piece of image data.

Abstract: A technique that includes predicting data acquired by a network of sensors based at least in part on a graphical model of the network, where the graphical model includes true value nodes, observed value nodes and edge factors based at least in part on historical pairwise dependencies for the observed value nodes. The technique includes detecting anomalous sensor data based at least in part on the predicted data.

Abstract: An object recognition system is provided that includes a device configured to capture a video sequence formed from unlabeled testing video frames. The system includes a processor configured to pre-train a recognition engine formed from a reference set of CNNs on a still image domain that includes labeled training still image frames. The processor adapts the recognition engine to a video domain to form an adapted recognition engine, by applying a non-reference set of CNNs to a set of domains that include the still image and video domains and a degraded image domain. The degraded image domain includes labeled synthetically degraded versions of the labeled training still image frames included in the still image domain. The video domain includes random unlabeled training video frames. The processor recognizes, using the adapted engine, a set of objects in the video sequence. A display device displays the set of recognized objects.

Abstract: A surveillance system is provided that includes a device configured to capture a video sequence, formed from a set of unlabeled testing video frames, of a target area. The surveillance system further includes a processor configured to pre-train a recognition engine formed from a reference set of CNNs on a still image domain that includes labeled training still image frames. The processor adapts the recognition engine to a video domain to form an adapted recognition engine, by applying a non-reference set of CNNs to domains including the still image and video domains and a degraded image domain. The degraded image domain includes labeled synthetically degraded versions of the frames included in the still image domain. The video domain includes random unlabeled training video frames. The processor recognizes, using the adapted engine, at least one object in the target area. A display device displays the recognized objects.

Abstract: This disclosure is directed to improving the precision of color reproducibility in a whole image region of any product provided with a base layer(s) regardless of any form of the base layer(s). The RIP (Raster Image Processor) device includes a color conversion processor that applies a color conversion process adapted for presence or absence of a base layer and for a base color to content data representing part or all of details of an image layer at a respective one of positions in an image region indicated by the image layer, and an image layer data generator that generates image layer data representing ejection-related information of ink droplets of a plurality of colors using the content data that has been subjected to the color conversion process.