Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating an output image. In one aspect, one of the methods includes generating the output image intensity value by intensity value according to a generation order of pixel—color channel pairs from the output image, comprising, for each particular generation order position in the generation order: generating a current output image representation of a current output image, processing the current output image representation using a decoder neural network to generate a probability distribution over possible intensity values for the pixel—color channel pair at the particular generation order position, wherein the decoder neural network includes one or more local masked self-attention sub-layers; and selecting an intensity value for the pixel—color channel pair at the particular generation order position using the probability distribution.

Abstract: Various implementations include systems for processing audio signals. In particular implementations, a process includes receiving an audio signal, wherein the audio signal includes a speech component of the user and a noise component; filtering the audio signal with a self-speech filter that utilizes an intrinsic user vector to filter out the speech component, wherein the intrinsic user vector is determined based on a voice input of the user; and outputting a filtered audio signal in which the speech component of the user has been substantially removed from the audio signal.

Abstract: A moving picture coding apparatus includes an intra-inter prediction unit which calculates a second motion vector by performing a scaling process on a first motion vector of a temporally neighboring corresponding block, when selectively adding, to a list, a motion vector of each of one or more corresponding blocks each of which is either a block included in a current picture to be coded and spatially neighboring a current block to be coded or a block included in a picture other than the current picture and temporally neighboring the current block, determines whether the second motion vector has a magnitude that is within a predetermined magnitude or not within the predetermined magnitude, and adds the second motion vector to the list when the intra-inter prediction unit determines that the second motion vector has a magnitude that is within the predetermined magnitude range.

Abstract: A self-healing video surveillance system is described. The self-healing video surveillance system includes one or more surveillance cameras that are configured to store configuration data of a network video recorder in operable communication with the surveillance cameras. The network video recorder includes configuration data, and a module configured to store the configuration data on the surveillance cameras and/or retrieve configuration data stored on the surveillance cameras. A new network video recorder introduced into the network retrieves the stored configuration data to self-configure with minimal or no human interaction. In embodiments, configuration data is distributed among a plurality of surveillance cameras, and may be stored in encrypted format.

Abstract: Methods, systems, and apparatus, including medium-encoded computer program products, for pre-processing image data before 3D object tracking includes, in at least one aspect, a method including: receiving, at a first computer, image frames from a camera; identifying, by the first computer, locations of interest in the image frames; finding sequences of the locations, wherein each of the sequences satisfies a motion criterion for locations identified in at least three image frames from the camera; and sending output data for the sequences of the locations to a second computer for processing the sequences in the output data by interpolating between specified 2D positions in specific image frames for the sequences, using timestamps of the specific image frames, to produce a virtual 2D position at a predetermined point in time, which is usable for constructing a 3D track of a ball in motion.

Abstract: An illumination device for illuminating a region monitored by at least one image sensor. The illumination device has at least one light source carrier with at least one light source arranged thereon. Furthermore, the illumination device has a control and interface unit and an adjusting unit, which can be actuated by the control and interface unit and with which the light source carrier is positionally adjustable.

Abstract: Disclosed are GPS-based target tracking system, method and dome camera. The system comprises: a first camera and a second camera. The first camera obtains GPS coordinate of a target during the tracking of the target and sends the GPS coordinates to the second camera. The second camera adjusts a shooting angle of the second camera according to the GPS coordinates, and tracks the target after detecting the target in a second monitoring image shot by the second camera. Therefore, in the scheme, if the target moves out of the monitoring range of the first camera, the second camera continues to track the target, namely the tracking range of the target is expanded through linkage of the plurality of cameras.

Abstract: An object-tracking system is disclosed. The tracking system is designed for environments where global positioning system (GPS), radio frequency (RF), and/or cellular communication signals are unavailable. The system is configured to use camera-captured images of the surrounding environment in conjunction with inertial measurements to perform visual and/or traditional odometry. An object detection algorithm and/or tracking scheme may be used to detect objects within the captured images, to help determine a user position relative to the objects. The detector architecture may be configured to allow for target (and/or object) agnostic camera detection and/or tracking that is easily configurable and/or reconfigurable depending on for the type of object to be detected and/or tracked.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating an output image. In one aspect, one of the methods includes generating the output image intensity value by intensity value according to a generation order of pixel—color channel pairs from the output image, comprising, for each particular generation order position in the generation order: generating a current output image representation of a current output image, processing the current output image representation using a decoder neural network to generate a probability distribution over possible intensity values for the pixel—color channel pair at the particular generation order position, wherein the decoder neural network includes one or more local masked self-attention sub-layers; and selecting an intensity value for the pixel—color channel pair at the particular generation order position using the probability distribution.

Abstract: A method and system are described for automatically centering in an XY plane a field of view of a patient's eye under high magnification during ophthalmic surgery. The method includes automatically moving the center of the field of view to the center of a circular image detected in a real-time video signal acquired from the field of view of the patient's eye under high magnification during ophthalmic surgery. The system configured to perform the method has a processor and a non-transitory computer-readable medium accessible to the processor containing instructions executable by the processor to perform the method.

Abstract: Systems and methods for high fidelity multi-modal out-of-band biometric authentication with cross-checking are disclosed. According to one embodiment, a method for integrated biometric authentication may include (1) receiving, from a user, biometric data; (2) at least one computer processor performing machine-based biometric matching on the biometric data; (3) the at least one computer processor determining that human identity confirmation is necessary; (4) the at least one computer processor processing the biometric data; (5) the at least one computer processor identifying at least one contact for human identity confirmation; (6) the at least one computer processor sending at least a portion of the processed biometric data for the user to the at least one contact; (7) receiving, from the at least one contact, human confirmation information; and (8) the at least one computer processor authenticating the user based on the machine-based biometric matching and the human confirmation information.

Abstract: Provided is a control device including control circuitry configured to control a position and an attitude of a microscope by driving an arm that supports the microscope on the basis of a captured image of a surgical operating site photographed by the microscope during a surgical operation so that a position and attitude condition is satisfied. The position and attitude condition is a condition that prescribes a position and an attitude of the microscope with respect to the surgical operating site to obtain a desired captured image corresponding to the position and attitude condition.

Abstract: Systems and methods for high fidelity multi-modal out-of-band biometric authentication with cross-checking are disclosed. According to one embodiment, a method for integrated biometric authentication may include (1) receiving, from a user, biometric data; (2) at least one computer processor performing machine-based biometric matching on the biometric data; (3) the at least one computer processor determining that human identity confirmation is necessary; (4) the at least one computer processor processing the biometric data; (5) the at least one computer processor identifying at least one contact for human identity confirmation; (6) the at least one computer processor sending at least a portion of the processed biometric data for the user to the at least one contact; (7) receiving, from the at least one contact, human confirmation information; and (8) the at least one computer processor authenticating the user based on the machine-based biometric matching and the human confirmation information.

Abstract: Techniques described herein relate to adjusting a quality level of media streaming. A user device may transmit a request for a first segment of a media file at a first streaming quality level. The user device may determine changes in network communication quality associated with a wireless communication in that the user device is engaged. In response to a determination, the user device may adjust quality levels of media streaming.

Abstract: Enhanced contrast between an object of interest and background surfaces visible in an image is provided using controlled lighting directed at the object. Exploiting the falloff of light intensity with distance, a light source (or multiple light sources), such as an infrared light source, can be positioned near one or more cameras to shine light onto the object while the camera(s) capture images. The captured images can be analyzed to distinguish object pixels from background pixels.

Abstract: Systems and methods for determining one or more dimensions of an object using a digital image are described.

Abstract: In accordance with various aspects of the disclosure, a system, a method, and computer readable medium having instructions for processing images is disclosed. For example, the method includes receiving, at an image processor, a set of images corresponding to a scene changing with time, decomposing, at the image processor, the set of images to detect static objects, leaner objects, and mover objects in the scene, the mover objects being objects that change spatial orientation in the scene with time, and compressing, using the image processor, the mover objects in the scene separately at a rate different from that of the static objects and the leaner objects for storage and/or transmission.

Abstract: A method for automatically detecting and tracking multiple targets in a multi-camera surveillance zone and system thereof. In each camera view of the system only a simple object detection algorithm is needed. The detection results from multiple cameras are fused into a posterior distribution, named TDP, based on the Bayesian rule. This TDP distribution represents a likelihood of presence of some moving targets on the ground plane. To properly handle the tracking of multiple moving targets with time, a sample-based framework which combines Markov Chain Monte Carlo (MCMC), Sequential Monte Carlo (SMC), and Mean-Shift Clustering, is provided. The detection and tracking accuracy is evaluated by both synthesized videos and real videos. The experimental results show that this method and system can accurately track a varying number of targets.

Abstract: A detection area is decided in a case where a target has gone out-of-frame. If a target is being imaged by a camera continuously, it is determined whether the target has gone out-of-frame. If the target has gone out-of-frame, then the magnitude and direction of motion of the camera are detected. If camera motion is large, it can be concluded that the camera user is imaging the target while tracking it. Accordingly, it can be concluded that the target will again be imaged at the center of the imaging zone. An area defined as a region in which the target will be detected is set at the center of the imaging zone. If camera motion is small in a case where the target goes out-of-frame, it can be concluded that the user is waiting for the target to re-enter the imaging zone and therefore the edge of the imaging zone is set as the detection area.

Abstract: An image processing apparatus includes: an object state detector detecting a state of an object in input images inputted in time series; an image-capturing controller controlling image-capturing of the input images in accordance with the detected state of an object; and an image combiner combining pixel values at corresponding pixel positions of the input images, and outputting the pixel values as an output image, the input images being subjected to image-capturing control by the image-capturing controller.

Abstract: Time-lapsing mirror methods and related systems.

Abstract: A video surveillance system includes at least one of a camera or a streamer. A data base server is coupled to the camera and can store metadata for a video clip from the camera or streamer. A media storage server is coupled to both the camera or, the streamer, and to the data base server to store the clip in the absence of any network video recorders.

Abstract: Segments of interest within video footage are identified automatically. Such segments of interest include portions of the footage which were taken using a hand held camera and which show a static subject, portions which track the motion of a moving subject, and portions which show an abrupt event. Methods are presented for identifying such segments, based on motion and/or colour analysis of the video footage or based on observations of discontinuities in the video footage following a period of stability.

Abstract: A method for detecting touch-point coordinates includes: scanning a plurality of signal points in sequence to generate a plurality of original data including a plurality of original touch point data; performing a grouping algorithm for the original touch point data to group the original touch point data into a plurality of group sets; and calculating a barycentric coordinate of each of the group sets and outputting the barycentric coordinate as a touch-point coordinate of each of the group sets.

Abstract: An image recognition device includes: an image sensor that captures an image via an image forming optical system and outputs image information repeatedly; a detector that detects differences between reference image information related to a reference image and a plurality of sets of partial information included in the image information; a recognition circuit that includes a microprocessor and recognizes a set of partial information, for which a smallest difference value among a plurality of differences detected in correspondence to the plurality of sets of partial information is calculated, as information matching the reference image information if the smallest value is smaller than a threshold value; and a controller that includes a microprocessor and sets the threshold value in correspondence to a color indicated in the reference image information.

Abstract: According to one embodiment, a display device includes an image data generating unit and a display unit. The image data generating unit generates image data on an image. The display unit presents a video including the image data to an image viewer on a moving body to superimpose the video on an actual outside scene. The image data includes a first display object and a second display object. The first display object includes a lower part object. The lower part object has a lower part, an upper right part and an upper left part. A length of the lower part along a normal direction is longer than lengths of the upper right and upper left parts. The second display object is disposed in the image with reference to a display position of the first display object. The second display object includes information on the moving body.

Abstract: A target tracking apparatus and method according to an exemplary embodiment of the present invention may quickly and accurately perform target detection and tracking in a photographed image given as consecutive frames by acquiring at least one target candidate image most similar to a photographed image of a previous frame among prepared reference target images, determining one of the target candidate images as a target confirmation message based on the photographed image, and calculating a homography between the determined target confirmation image and the photographed image, and searching the photographed image of the previous image for feature points according to the calculated homography, and tracking an inter-frame change from the previous frame of the found feature points to a current frame.

Abstract: A wireless video camera system including a single or dual HDTV digital camera; a system for the wireless transmission of uncompressed single or multiple SMPTE 292M, HDSDI data streams over an RF band; and automatic self-pointing transmitting and receiving antennas.

Abstract: An image signal processing unit partitions a frame image into a plurality of partitioned blocks and calculates for each partitioned block luminance information, color information, contrast information, and a motion vector quantity. A grouping unit groups the partitioned blocks into an image area of the same object based on the luminance information and the color information of each partitioned block. A moving amount calculating unit calculates a moving amount of an object contained in a grouped image area based on the motion vector quantity of each partitioned block, a detection value of a motion detection sensor, and a focal length calculated by a focal length calculating unit. A tracking object selecting unit selects an object in the image area of the highest contrast value as a tracking target object out of objects each having a moving amount exceeding a threshold value.

Abstract: Methods and systems for video retargeting and view selection using motion saliency are described. Salient features in multiple videos may be extracted. Each video may be retargeted by modifying the video to preserve the salient features. A crop path may be estimated and applied to a video to retarget each video and generate a modified video preserving the salient features. An action score may be assigned to portions or frames of each modified video to represent motion content in the modified video. Selecting a view from one of the given modified videos may be formulated as an optimization subject to constraints. An objective function for the optimization may include maximizing the action score. This optimization may also be subject to constraints to take into consideration optimal transitioning from a view from a given video to another view from another given video, for example.

Abstract: An imaging apparatus is equipped with: imaging means, for obtaining images of a subject; imaging control means, for controlling the imaging means to sequentially obtain a plurality of images; reference image selecting means, for selecting a single reference image from among the plurality of images; moving object region specifying means, for specifying moving object regions within corresponding images, which are the plurality of images other than the selected reference image; positioning means, for positioning the plurality of corresponding images with respect to the reference image in the case that the plurality of images have been imaged by flow shooting, the positioning being performed such that the amount of positioning is greater at the specified moving object regions than at non moving object regions; and combining means, for combining the plurality of corresponding images, which have been positioned by the positioning means, with the reference image.

Abstract: A system and method for implementing beyond field-of-view tracked object positional indicators for television event directors and camera operators. The present invention includes a camera having a field-of-view. The camera tracks an off-screen object. A coordinate manager blends an on-screen indication of distance that the object is away from said field-of-view. The camera is positioned to avoid the object in the field-of-view.

Abstract: An image processing apparatus includes: an object state detector detecting a state of an object in input images inputted in time series; an image-capturing controller controlling image-capturing of the input images in accordance with the detected state of an object; and an image combiner combining pixel values at corresponding pixel positions of the input images, and outputting the pixel values as an output image, the input images being subjected to image-capturing control by the image-capturing controller.

Abstract: A method for presentation interaction. The method includes, receiving by a computer system an indication of a manual selection of a region proximate to an audience member of an audience wherein the indication is received via an interaction with a displayed image of the audience. The method also includes utilizing a microphone array communicatively coupled with a beam-forming component of the computer system to focus audio pickup from the region proximate to the audience member in response to receiving the indication. The method also includes displaying an enhanced image of the region proximate to the audience member using the computer system in response to receiving the indication.

Abstract: A computerized method of video analysis that includes receiving several series of video frames generated by a number of image sensors. Each image sensor has its own field-of-view which can, but does not have to, overlap with another image sensor's field-of-view. The image sensors monitor a portion of a monitored environment. The computerized method also includes concurrently tracking, independent of calibration, multiple objects within the monitored environment as the objects move between fields-of-view, and multiple objects within one field-of-view. The tracking is based on the plurality of received series of video frames.

Abstract: A terminal has a visual axis by using which a user points to a target object. An information generating server acquires an internal image or an external image of the target object pointed by a user, and generates a display image which is displayed on the terminal.

Abstract: A multi-directional remote control system and method is adapted for use with an entertainment system of a type including a display such as a monitor or TV and having display functions employing a mouse type control. The remote controller may be conveniently held in one hand of a user and still provides full mouse type functionality. The remote control system and method images the controller to detect relative motion between the controller and screen. This position information is used for control of a cursor or other GUI interface.

Abstract: There are provided an automatic vision sensor placement apparatus and method which analyze static spatial information in various aspects, calculate priorities of spaces using an agent that models the movement pattern of a user, suggest a method of effectively covering a given space, provide a camera placement method to appropriately install cameras with various levels of performance in a specified space, and calculate the optimal number of cameras to be installed based on installation cost.

Abstract: The present invention discloses a method for image pickup and an image pickup device using the same. The image pickup device includes a lens, an image-processing program, and a user interface. The lens is utilized to obtain a main image by shooting an object. The image-processing program divides the main image into a plurality of sub-image blocks, computes the resolution of every sub-image block, and compares each resolution with a threshold. The user interface displays the compared result of each sub-image block to indicate if the lens and the object are parallel.

Abstract: Coherent motion regions extend in time as well as space, enforcing consistency in detected objects over long time periods and making the algorithm robust to noisy or short point tracks. As a result of enforcing the constraint that selected coherent motion regions contain disjoint sets of tracks defined in a three-dimensional space including a time dimension. An algorithm operates directly on raw, unconditioned low-level feature point tracks, and minimizes a global measure of the coherent motion regions. At least one discrete moving object is identified in a time series of video images based on the trajectory similarity factors, which is a measure of a maximum distance between a pair of feature point tracks.

Abstract: An image generating device for an optical, in particular medical tracking system, for determining the position of a recorded image point, includes a light scattering effect generating device, in particular a star and/or cross effect generating device. The invention further provides a tracking system including such an image generating device and to a method for determining the position of a recorded image point using an optical, in particular medical tracking system, in which: an image is generated using an image generating device; a light scattering effect, in particular a star and/or cross effect, is generated on the image for predetermined image points; and in which the position of an image point is ascertained on the basis of the light scattering pattern projected on the image, in particular on the basis of the star and/or cross effect lines and/or their intersection points.

Abstract: A system for automatically capturing an event of interest in a venue is disclosed, comprising a plurality of cameras for capturing video images of the event; and at least one processor communicatively connected to said plurality of cameras and con figured to execute a plurality of modules, said modules comprising a rectification module for combining the video images to form a wide-angle view; at least one of a motion analysis module configured for tracking salient blobs that are not part of a background of the wide-angle view, an activity analysis module configured for extracting temporal and spatial patterns from the wide-angle view, and a shape and appearance module configured for selecting one or more objects in the wide-angle view based on descriptors that are scale and rotational invariant; and a region of interest selector for selecting a viewpoint from the wide-angle view based on output from at least one of the motion analysis module, the activity analysis module, and the shape and appearance module,

Abstract: Segments of interest within video footage are identified automatically. Such segments of interest include portions of the footage which were taken using a hand held camera and which show a static subject, portions which track the motion of a moving subject, and portions which show an abrupt event. Methods are presented for identifying such segments, based on motion and/or color analysis of the video footage or based on observations of discontinuities in the video footage following a period of stability.

Abstract: Disclosed is an apparatus for photographing an image using a digital camera capable of providing a preview image, including: an image sensor for capturing an image of a subject; a first image signal processor for processing at least one captured image as a sequent image in order to display the captured image as a preview image on a display window and detecting a face area from the captured image; a buffer for storing the image including at least one detected face area; a capture button for capturing a still image during outputting the sequent image as the preview image on the display window; a second image signal processor for processing the captured still image using the capture button; a controller for controlling to store the captured still image and information related on the face area detected from the first image signal processor prior to the input of the capture button; and a memory for storing the processed still image together with the information related on the detected face area.

Abstract: In order to appropriately detect abnormalities for the purpose of surveillance with it being possible to prevent an unnecessary increase in data amount of the monitoring images, a monitoring system includes an image capturing section that captures a moving image of a monitored area, a variation reduced image generating section that generates a variation reduced image by reducing a temporal variation in an image, based on a plurality of moving-image making-up images included in the moving image captured by the image capturing section, a condition storing section that stores thereon a condition which is required to be satisfied by a variation reduced image which is generated by using a plurality of moving-image making-up images included in a moving image which is judged to show an abnormality, a satisfaction judging section that judges whether the variation reduced image generated by the variation reduced image generating section satisfies the condition, an output moving image generating section that, when the

Abstract: An image pick-up apparatus that enables to appropriately set a holding time in which there is a possibility of returning to a state that allows tracking when a subject cannot be tracked temporarily during the subject tracking to improve the ease-of-use for a user. A specifying unit specifies a subject included in a captured image. A display unit displays the captured image on a screen and displays identification information showing that the specified subject is tracked. A tracking unit tracks the subject. A setting unit sets a holding time in which the display of the identification information is held according to at least one of a focal length of the image pick-up apparatus and a subject distance. An elimination unit eliminates a display of the identification information when the holding time has passed after the tracking unit lost the subject.

Abstract: A portable navigation device with a TV function includes a TV module, a navigation module, a display module, a sound module, a first processing unit, a second processing unit and a TV command storage module. The TV module generates baseband TV signals, the navigation module generates position information, the display module displays image information, the sound module plays sounds, the first processing unit performs navigation programs, the second processing unit performs TV programs, and the TV command storage module stores TV commands from the first processing unit and the second processing unit. TV command transmission between the first processing unit and the second processing unit is achieved by sending interrupt signals or reset signals.

Abstract: An image recognition device includes; an image sensor that captures an image via an image forming optical system and outputs image information repeatedly; a detector that detects differences between reference image information related to a reference image and a plurality of sets of partial information included in the image information; a recognition circuit that includes a microprocessor and recognizes a set of partial information, for which a smallest difference value among a plurality of differences detected in correspondence to the plurality of sets of partial information is calculated, as information matching the reference image information if the smallest value is smaller than a threshold value; and a controller that includes a microprocessor and sets the threshold value in correspondence to a color indicated in the reference image information.

Abstract: An automatic tracking apparatus is provided, which is capable of solving a failure occurred in an automatic tracking operation in connection with a zooming operation, and capable of tracking an object in a stable manner, while a zooming-up operation, or a zooming-down operation is carried out in a high speed.

Abstract: A microscope image pickup system is one having a microscope apparatus enabled to change an observation state by driving one or more optical members, which comprises: an image pickup unit for picking up an image of an observation object; an image process unit for applying an image process to an image picked up by the image pickup unit; an input unit for inputting a process factor of the image process unit, wherein an observation state of the microscope apparatus and/or an image pickup condition of the image pickup unit are set up and an image is picked up so as to pick up an image of an image quality being equal to, or better than, that of an image to which an image process is applied on the basis of the process factor input from the input unit.