Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium. In one aspect, a method includes receiving a text prompt describing a scene; processing the text prompt using a text encoder neural network to generate a contextual embedding of the text prompt; and processing the contextual embedding using a sequence of generative neural networks to generate a final video depicting the scene.

Abstract: In one embodiment, a method includes receiving a user query inputted on a head-mounted device from the head-mounted device, wherein the user query corresponds to multiple dialog-intents, executing multiple tasks corresponding to the multiple dialog-intents, generating a multi-perspective response by a stitching model based on two or more of execution results of the multiple tasks, wherein the stitching model combines the two or more of the execution results based on natural language processing, and wherein the multi-perspective response comprises a natural-language response combining the two or more execution results, and sending instructions to the head-mounted device for presenting the multi-perspective response on the head-mounted device.

Abstract: An apparatus and a method for vehicle profiling. The apparatus includes at least a sensor, wherein the at least a sensor is configured to capture vehicle data of a vehicle, at least a processor and a memory communicatively connected to the at least processor, the memory contains instructions configuring the at least processor to receive the vehicle data from the at least a sensor, generate a graphic model of the vehicle as a function of the vehicle data, determine a vehicle status as a function of the vehicle data, classify the vehicle status and the vehicle data into one or more vehicle status groups and generate a vehicle report as a function of the one or more vehicle status groups.

Abstract: A device may capture a plurality of preview frames of a document, and for each preview frame of the plurality of preview frames, process the preview frame to identify an object in the preview frame. Processing the preview frame may include converting the preview frame into a grayscale image, generating a blurred image based on the grayscale image, detecting a plurality of edges in the blurred image, defining at least one bounding rectangle based on the plurality of edges, and determining an outline of the object based on the at least one bounding rectangle. The device may determine whether a value of an image parameter, associated with the one or more preview frames, satisfies a threshold, and provide feedback to a user of the device, or automatically capture an image of the document, based on determining whether the value of the image parameter satisfies the threshold.

Abstract: Surveillance systems process video streams obtained by a surveillance device, such as a drone, to either obscure or highlight objects in a surveillance area based on tags associated with the objects. A method of providing obscurant data includes receiving image data including an image of a target and receiving a preference setting corresponding to the target. Obscurant data of at least a portion of the image data corresponding to the target are determined using the received preference setting. A method of providing surveillance image data includes capturing image data including an image of a target, querying a database to receive a preference setting corresponding to the target, determining the obscurant data of the portion of the image data, and selectively modifying the received image data according to the determined obscurant data to provide the surveillance image data.

Abstract: Methods, non-transitory computer-readable storage media, and computer or computer systems directed to detecting, analyzing, and tracking roads and grading activity using satellite or aerial imagery in combination with a machine learned model are described.

Abstract: Systems and methods for optimizing asset maintenance protocols by predicting vegetation-driven outages are disclosed. An example method includes determining, by one or more processors, a failure probability for each asset in a set of assets within a designated area. The example method further includes defining, by the one or more processors, an asset risk for each asset in the set of assets based on the failure probability, and clustering, by the one or more processors, vegetation within the designated area to determine a predicted vegetation-driven outage. The example method further includes optimizing, by the one or more processors, a set of asset maintenance protocols corresponding to the set of assets based on the asset risk and the predicted vegetation-driven outage.

Abstract: Methods, apparatus, systems, and computer-readable media are set forth for generating and/or utilizing image shortcuts that cause one or more corresponding computer actions to be performed in response to determining that one or more features are present in image(s) from a camera of a computing device of a user (e.g., present in a real-time image feed from the camera). An image shortcut can be generated in response to user interface input, such as a spoken command. For example, the user interface input can direct the automated assistant to perform one or more actions in response to object(s) having certain feature(s) being present in a field of view of the camera. Subsequently, when the user directs their camera at object(s) having such feature(s), the assistant application can cause the action(s) to be automatically performed. For example, the assistant application can cause data to be presented and/or can control a remote device in accordance with the image shortcut.

Abstract: Embodiments of this application disclose methods, systems, and devices for medical image analysis and medical video stream processing. In one aspect, a method comprises extracting video frames from a medical image video stream that includes at least two pathological-section-based video frames. The method also comprises identifying single-frame image features in the video frames, mapping the single-frame image features into single-frame diagnostic classification results, and performing a classification mapping based on a video stream feature sequence that comprises the single-frame image features. The classification mapping comprises performing a convolution operation on the video stream feature sequence through a preset convolutional layer, obtaining a convolution result in accordance with the convolution operation, and performing fully connected mapping on the convolution result through a preset fully connected layer.

Abstract: This invention provides a system and method that employs reduction in bandwidth and the amount of data stored, along with queuing of data, so that the significant and/or relevant shipboard visual information is detected and communicated continuously from the ship (or other remote vehicle/location) to shore without loss of useful/high-priority information and within the available bandwidth of that typical, available satellite link. The system and method supports remote configuration and management from shore to the ship over the same communications channel but in the reverse direction.

Abstract: The present invention relates to a device, a system and a method for tracking at least one object and player on a playing field/court and/or for monitoring the course of the game.

Abstract: A computing system identifies broadcast video data for a game. The computing system generates tracking data for the game from the broadcast video data using computer vision techniques. The tracking data includes coordinates of players during the game. The computing system generates optical character recognition data for the game from the broadcast video data by applying one or more optical character recognition techniques to each frame of the plurality of frames to extract score and time information from a scoreboard displayed in each frame. The computing system detects a plurality of events that occurred in the game by applying one or more machine learning techniques to the tracking data. The computing system receives play-by-play data for the game. The computing system generates enriched tracking data. The generating includes merging the play-by-play data with one or more of the tracking data, the optical character recognition data, and the plurality of events.

Abstract: An electronic device is disclosed. The electronic device comprises a memory for storing a content, and a processor for: acquiring a probability value for each of a plurality of objects included in each of a plurality of frames configuring the stored content; grouping the plurality of objects into at least one group according to a correlation value between the plurality of objects, the correlation value being obtained on the basis of the acquired probability value; counting, for each of a plurality of frames for each group, a case where the acquired probability value is equal to or greater than a preconfigured threshold value; and acquiring a summary content on the basis of a result of the counting.

Abstract: Methods, systems and apparatuses for a custom artificial neural network (ANN) architecture are disclosed.

Abstract: A modular tracking system is described comprising of the network of independent tracking units optionally accompanied by a LIDAR scanner and/or (one or more) elevated cameras. Tracking units are combining panoramic and zoomed cameras to imitate the working principle of the human eye. Markerless computer vision algorithms are executed directly on the units and provide feedback to motorized mirror placed in front of the zoomed camera to keep tracked objects/people in its field of view. Microphones are used to detect and localize sound events. Inference from different sensor is fused in real time to reconstruct high-level events and full skeleton representation for each participant.

Abstract: Systems, methods, and computer program products of intelligent image analysis using object detection models to identify objects and locate and detect features in an image are disclosed. The systems, methods, and computer program products include automated learning to identify the location of an object to enable continuous identification and location of an object in an image during periods when the object may be difficult to recognize or during low visibility conditions.

Abstract: In accordance with one disclosed method, a computing system may cause a first computing device to display a first notification of a first event detected at a monitored location, and may cause a second computing device to display a second notification of a second event detected at the monitored location. The computing system may additionally cause the second computing device to cease display of the second notification in response to a change of status of the first event.

Abstract: An information processing apparatus that detects an object from an image, includes a determination unit configured to determine a boundary at which a movement of the object between partial regions in the image is estimated, the partial regions each including a plurality of objects, and an estimation unit configured to estimate movement information indicating a number of objects that have passed the boundary determined by the determination unit.

Abstract: A computing system includes one or more processors and a memory storing instructions that, when executed by the one or more processors, cause the system to perform operations. The operations include obtaining sensor data from a sensor of a vehicle, the sensor data including point cloud frames at different positions, orientations, and times, the sensor data used to generate a map, determining a position and an orientation of the sensor corresponding to a capture of each of the point cloud frames according to a simultaneous localization and mapping (SLAM) algorithm, and depicting, on an interface, a graphical illustration of the determined positions at which the point cloud frames were captured.

Abstract: Image data captured from a traveling vehicle is considered, and it is not possible to reduce the transmission band of the image data. It is assumed that a radar 4 mounted in a traveling vehicle 10 detects a certain distant three-dimensional object at a time T in a direction of a distance d1 [m] and an angle ?1. Since the vehicle 10 travels at a vehicle speed Y [km/h], it is predicted that a camera 3 is capable of capturing the distant three-dimensional object at a time (T+?T) and an angle ?1 or at a distance d2 [m]. Therefore, if a control unit 2 outputs a request to the camera 3 in advance, so as to cut out an image of the angle ?1 or the distance d2 [m] at the time (T+?T), when the time (T+?T) comes, the camera 3 transfers a whole image and a high-resolution image being a cutout image of only a partial image, to the control unit 2.

Abstract: An apparatus includes at least one camera configured to capture a series of image frames for traffic lanes in front of an ego vehicle, where each of the series of image frames is captured at a different one of a plurality of times. A target object detection and tracking controller is configured to process each of the image frames using pixel measurements extracted from the respective image frame to determine, from the pixel measurements, a predicted time to line crossing for a target vehicle detected in the respective image frame at a time corresponding to capture of the respective image frame.

Abstract: An apparatus that supports driving of a vehicle may include a location check device that determines whether the vehicle is located within a circuit area, based on location information of the vehicle, a camera device that obtains an image around the vehicle, and a processor that recognizes a specific object from the obtained image based on whether the vehicle is located within the circuit area, sets a region of interest (ROI) with respect to the specific object, and performs image-processing on the ROI preferentially.

Abstract: For one embodiment of the present invention, a method of object detection and part segmentation is described. The method includes generating a three-dimensional (3D) point cloud including a plurality of points in a volume of a space of the point cloud based on receiving point cloud dat. The method further includes obtaining, with a global transformer, a set of global feature tokens and local feature points of the 3D point cloud, enforcing with an attention mechanism information flow across the set of global feature tokens and local feature point, and generating, with the global transformer, a set of new global feature tokens and new local feature points with modeled global context information.

Abstract: LiDAR (light detection and ranging) and RADAR (radio detection and ranging) systems are commonly used to generate point cloud data for 3D space around vehicles, for such functions as localization, mapping, and tracking. Improved techniques for processing the point cloud data that has been collected are provided. The improved techniques include mapping one or more point cloud data points into a depth map, the one or more point cloud data points being generated using one or more sensors; determining one or more mapped point cloud data points within a bounded area of the depth map, and detecting, using one or more processing units and for an environment surrounding a machine corresponding to the one or more sensors, a location of one or more entities based on the one or more mapped point cloud data points.

Abstract: Navigation systems can identify objects in an environment and generate representations of those objects. A representation of an articulated vehicle can include two segments rotated relative to each other about a pivot, with a first segment corresponding to a first portion of the articulated vehicle and the second segment corresponding to a second portion of the articulated vehicle. The articulated object can be tracked in the environment by generating estimated updated states of the articulated agent based on previous states and/or measured states of the object using differing motion model updates for the differing portions. The estimated updated states may be determined using one or more filtering algorithms, which may be constrained using pseudo-observables.

Abstract: The invention relates to a method for increasing the safety of driving functions in a partially automated or fully autonomous vehicle, including the following steps: capturing (S1) an environment image or a sequence of environment images by means of at least one surroundings detection sensor (2), detecting (S2) driving lane boundaries in the environment image or the sequence of environment images, determining (S3) a driving lane course based on the detected driving lane boundaries; retrieving (S4) a further driving lane course from a data source; checking the plausibility (S5) of the determined driving lane course by verifying a matching of the driving lane courses; identifying (S6) a degree of matching; setting (S7) a confidence value based on the degree of matching; deciding (S8) whether the determined driving lane course is provided to a driving function.

Abstract: Method for estimating a vertical offset of road segment for vehicle is disclosed. The method includes tracking road reference on road segment at first moment in time from first observation point in order to form first representation. Method includes obtaining vehicle movement data over time period extending from first moment in time to second moment in time, tracking road reference on road segment at second moment in time from second observation point in order to form second representation.

Abstract: A method for detecting road diseases by intelligent cruise via an unmanned aerial vehicle (UAV), the UAV and a detecting system therefor are provided. The method for detecting road diseases by intelligent cruise via UAV, wherein a road disease detection model and a road recognition model based on deep learning network are built in the UAV, wherein the method specifically comprises a step of: automatically flying the UAV on a predetermined route on the actual road determined by the road recognition model, and obtaining road disease test results by the road surface disease detection model. The present invention adopts the road recognition model and road disease detection model based on deep learning network, which can realize automatic cruise and automatic road disease detection, only need to set a predetermined route or area range, which is convenient and fast.

Abstract: A driver monitor system and method of optimizing use of cognitive resources of a user in a vehicle is disclosed. The system includes in-cabin video cameras, a display device, and processing circuitry. The method performed by the processing circuitry includes monitoring eye gaze direction using images from the in-cabin video cameras, gradually softening image regions and lines displayed in the display device that are outside a first vision area of the user of the vehicle, and using the eye gaze direction to predict whether the user is transitioning into a new focus vision area that had been outside the first vision area. When the user's eyes begin to move, begin sharpening softened image regions and lines such that by the time eyes have shifted into the new focus vision area, the image regions and lines in the new focus vision area reach full sharpness.

Abstract: Aspects of the present disclosure relate to a control system, system and method for controlling a transition between a first driving mode and a second driving mode of a vehicle. The present disclosure relates to receiving image data of an occupant of the vehicle; quantifying one or more characteristics of the occupant's gaze based at least in part on the received image data, the one or more characteristics being indicative of the occupant's attention to a non-driving task; determining a transition profile for transitioning between the first driving mode and the second driving mode based at least in part on the one or more characteristics of the occupant's gaze; and generating and outputting a control signal to transition from the first driving mode to the second driving mode, the control signal comprising instructions for controlling one or more vehicle systems in accordance with the determined transition profile.

Abstract: A method and apparatus for detecting an object in a three-dimensional (3D) point cloud are disclosed. The apparatus includes a three-dimensional (3D) model reconstruction unit configured to reconstruct a three-dimensional (3D) model using three-dimensional (3D) point cloud data, a two-dimensional (2D) object detection unit configured to detect an object in at least one two-dimensional (2D) image captured to generate the three-dimensional (3D) point cloud data, a three-dimensional (3D) position estimation unit configured to form a cylinder surrounding the object in the three-dimensional (3D) model based on a two-dimensional (2D) position of the detected object so as to estimate a three-dimensional (3D) position of the detected object using the cylinder, and an object region extraction unit configured to acquire three-dimensional (3D) point cloud data contained in the cylinder so as to extract an object region from the acquired three-dimensional (3D) point cloud data.

Abstract: Comprehensive 2D learning images are collected for learning subjects. Standardized 2D gallery images of many gallery subjects are collected, one per gallery subject. A 2D query image of a query subject is collected, of arbitrary viewing aspect, illumination, etc. 3D learning models, 3D gallery models, and a 3D query model are determined from the learning, gallery, and query images. A transform is determined for the selected learning model and each gallery model that yields or approximates the query image. The transform is at least partly 3D, such as 3D illumination transfer or 3D orientation alignment. The transform is applied to each gallery model so that the transformed gallery models more closely resemble the query model. 2D transformed gallery images are produced from the transformed gallery models, and are compared against the 2D query image to identify whether the query subject is also any of the gallery subjects.

Abstract: A matching position output system includes: a shooting means configured to shoot an image of an object; a generating means configured to generate information identifying the object from the shot image of the object; and an output means configured to, based on the generated information identifying the object, acquire information of a matching position of the object from a recording means configured to associate and record the information identifying the object and the information of the matching position of the object, and output the information of the matching position of the object.

Abstract: Systems and methods for extracting information from a sheet of material to facilitate organization of information from paper, and other physical writing surfaces are provided. An example system includes a sheet of material and a device for scanning the sheet with an optical sensor. The sheet of material includes an indication region. The indication region allows for indictors to be marked corresponding with at least one a corresponding subregion to be extracted. The sheet of material further includes at least one fiducial mark for identifying a boundary of the sheet. The device includes a processor operably coupled to the optical sensor for causing the optical sensor to scan the sheet and detect a boundary thereof using the fiducial marks and further identify a designated subregion of the sheet. Upon identification of the designated subregion, the processor is configured to extract information contained in the designated subregion for organization of information.

Abstract: An object management system includes an acquisition means for acquiring an image in which a surface of a registration target object, having a circle and a handwritten character drawn thereon, is captured, a generation means for detecting an ellipse corresponding to the circle from the image and generating a registration image in which the image is applied with projective transformation such that the ellipse becomes a circle, and a registration means for writing the registration image into a storage means as data for determining the sameness of the registration target object.

Abstract: An object is to improve character recognition accuracy of handwritten characters, originally a single continuous character string, described discontinuously. An image area corresponding to a handwritten character is separated from a document image obtained by scanning a document and a character block including characters having the same baseline is extracted. Then, in a case where a plurality of character blocks is extracted from the first image area, a single character block is generated by combining character blocks based on a position relationship of the plurality of character blocks.

Abstract: Disclosed herein is a method of identifying a line of a stave or a stem of a note in a digital image of a musical score comprising: converting the digital image into a matrix in which at least one cell of the matrix corresponds to a pixel of the digital image; setting the at least one cell of the matrix to a first value if the corresponding pixel of the digital image has a pixel intensity above a threshold intensity; identifying adjacent cells having the first value as linked cells, the adjacent cells corresponding to pixels being adjacent in one of a horizontal direction or a vertical direction of the digital image; identifying linked cells having a number of cells exceeding a threshold as a chain of cells; grouping adjacent chains of cells into a group of chains; determining a dimension of the group of chains; and comparing the dimension with a dimension threshold; wherein, if the dimension is above the dimension threshold, determining that pixels corresponding to linked cells of the group of chains corresp

Abstract: A document reading device includes a document conveyer, a first reader reading, in a first reading position, a first surface of a conveyed document such that a read area is larger than the conveyed document, a second reader reading, in a second reading position, a surface (second surface) opposite to the first surface such that a read area is larger than the conveyed document, a region detector executing a process of detecting a first document region that is a region of a document in first document image data and a process of detecting a second document region that is a region of the document in second document image data, and a cropping processor cropping a document portion on the first surface as first cropped image data and cropping a document portion on the second surface as second cropped image data, based on one of the document regions successfully detected.

Abstract: A data processing method includes detecting a first operation for a first page, determining a layout type of the first page, the first page including a first section and a second section, and in response to the layout type being same as a preset type, adjusting a relative position between the first section and the second section so that both of the first section and the second section are visible when moving along a first direction in a display area.

Abstract: The disclosure provides a method and a device for processing information, an electronic device, and a storage medium, belonging to a field of artificial intelligence including computer vision, deep learning, and natural language processing. In the method, the computing device recognizes multiple text items in the image. The computing device classifies multiple text items into a first set of name text items and a second set of content text items based on semantics of the text items. The computing device performs a matching operation between the first set and the second set based on a layout of the text items in the image, and determines matched name-content text items. The matched name-content text items include a name text item in the first set and a content text item matching the name text item and in the second set. The computing device outputs the matched name-content text items.

Abstract: This application relates to a system for building machine learning or deep learning data sets for automatically recognizing geographical area information comprising a plurality of geographical area components provided on items. The system may include a first image database configured to store a first plurality of sets of images of geographical area information of items, each first set including an image of an entirety of geographical area information of an item. The system may also include a second image database configured to store a second plurality of sets of images of the geographical area information of the items, each second set including images of individual geographical area components. The system may further include a controller configured to convert the first plurality of sets of images into the second plurality of sets of images.

Abstract: A method for determining the authenticity of an identity document is provided that includes capturing, by an electronic device, image data of an identity document. The identity document has a first side and a second side. The image data includes an image of the first side and an image of the second side. The first side includes data about a person associated with the identity document and second side includes a PDF417 barcode which has features. The method also includes determining a class of the identity document. The features of the PDF417 barcode adhere to criteria established for the determined class of the identity document. Moreover, the method includes analyzing the features of the PDF417 barcode for anomalies. In response to determining the features of the PDF417 barcode are free of anomalies, the identity document is determined to be authentic. Otherwise, the identity document is determined to be fraudulent.

Abstract: The present application relates to an occluded pedestrian re-identification method, including steps of obtaining global features and local features of occluded pedestrians, and recombining the local features into a local feature map; obtaining a heat map of key-points of pedestrian images and a group of key-point confidences, obtaining a group of features of the pedestrian key-points by using the local feature map and the heat map; obtaining a local feature group by using the global features to enhance each key-point feature in the group of features of pedestrian key-points according to Conv, and an adjacency matrix of key-points is obtained through the key-points, the local feature group and the adjacency matrix of key-points are used as the input of GCN to obtain the final features of pedestrian key-points.

Abstract: A display device includes a display panel; a metal plate and a filling member disposed on a same layer on a bottom surface of the display panel and that include different materials; a fingerprint sensor disposed on a bottom surface of the filling member and that overlaps the filling member; and a member-sensor bonding member disposed between the fingerprint sensor and the filling member and that bonds the fingerprint sensor to the filling member. At least a part of a side surface of the filling member and a side surface of the metal plate are spaced apart from each other with an air gap therebetween.

Abstract: Disclosed are a fingerprint identification structure, a driving method thereof and an electronic device. The fingerprint identification structure includes: a driving electrode layer; a piezoelectric material layer; a receiving electrode layer, which includes M receiving electrodes; an auxiliary driving electrode layer, which is located at the side of the piezoelectric material layer away from the receiving electrode layer, and is arranged in a layer different from the driving electrode layer; and a first insulating layer, the auxiliary driving electrode layer includes N auxiliary driving electrodes; the N driving electrodes and the N auxiliary driving electrodes are alternately arranged; and the orthographic projection, on the piezoelectric material layer, of an i-th auxiliary driving electrode overlaps with an interval between the orthographic projections, on the piezoelectric material layer, of an i-th driving electrode and an (i+1)-th driving electrode.

Abstract: In one example, a method comprises: receiving an indication that an object is within a distance from an array of ultrasonic transducers; based on the indication, configuring a subset of the array of ultrasonic transducers to perform an ultrasonic sensing operation on the object; obtaining, from the subset of the array of ultrasonic transducers, an output of the ultrasonic sensing operation; and performing an action based on the output of the ultrasonic sensing operation.

Abstract: A texture recognition device and a manufacturing method thereof are provided. The texture recognition device includes a backlight component and a photosensitive component. The photosensitive component is on a light-outputting side of the backlight component and configured to detect light emitted by the backlight component and reflected by a texture of a detection object to recognize an image of the texture of the detection object. The photosensitive component includes a plurality of photosensitive sensors and an antistatic layer on a side, away from the backlight component, of the plurality of photosensitive sensors, and orthographic projections of the plurality of photosensitive sensors on a plane where the antistatic layer is located are within the antistatic layer.

Abstract: In a method of obtaining a reference image for optical object recognition, first light sources among a plurality of light sources included in a display panel are driven where the first light sources are disposed to correspond to an object recognition window which is a partial region of the display panel, a plurality of images are contemporaneously obtained based on reflected light received by an object recognition sensor through the object recognition window while driving the first light sources, at least one of the plurality of images is obtained as a first reference image for optical object recognition associated with a first object, and the first reference image is stored.

Abstract: The present disclosure discloses a display panel and a display device. The display panel includes a display substrate, the display substrate includes a plurality of pixel units; and a touch layer and a fingerprint recognition layer disposed on the display substrate; wherein, the touch layer includes at least one touch unit; the fingerprint recognition layer includes at least one primary fingerprint recognition unit, each of the primary fingerprint recognition units includes NSFPS sub-fingerprint recognition units corresponding to Npixel pixel units, where NSFPS, Npixel are positive integers.

Abstract: According to one embodiment, a biometric authentication device includes a sensor and an illumination device. The sensor includes a first base material, a plurality of photoelectric conversion elements and a first control unit. The illumination device includes a second base material, a plurality of light emitting elements and a second control unit. Upon lighting the light emitting element at a predetermined position in the groups of the light emitting elements by the second control unit, the first control unit controls the operation of the photoelectric conversion elements so as to acquire a signal from the photoelectric conversion elements at predetermined positions of the groups of the photoelectric conversion elements.