Abstract: Techniques for gesture-based device connections are described. For example, a method may comprise receiving video data corresponding to motion of a first computing device, receiving sensor data corresponding to motion of the first computing device, comparing, by a processor, the video data and the sensor data to one or more gesture models, and initiating establishment of a wireless connection between the first computing device and a second computing device if the video data and sensor data correspond to gesture models for the same gesture. Other embodiments are described and claimed.

Abstract: Techniques for gesture-based device connections are described. For example, a method may comprise receiving video data corresponding to motion of a first computing device, receiving sensor data corresponding to motion of the first computing device, comparing, by a processor, the video data and the sensor data to one or more gesture models, and initiating establishment of a wireless connection between the first computing device and a second computing device if the video data and sensor data correspond to gesture models for the same gesture. Other embodiments are described and claimed.

Abstract: A method includes obtaining, by a device and from a dynamic vision sensor (DVS), a set of images of an object that identifies that the object has moved. The method includes determining, by the device, that the object is associated with a predetermined posture based on the set of images. The method includes determining, by the device, a group to which the object belongs based on an attribute of the object and an attribute of the group. The method includes determining, by the device, whether the object is associated with the dangerous situation based on identifying that the object is associated with the predetermined posture and based on setting information associated with the group to which the object belongs.

Abstract: Methods, systems, and techniques for monitoring an object-of-interest within a region involve receiving at least data from two sources monitoring a region and correlating that data to determine that an object-of-interest depicted or represented in data from one of the sources is the same object-of-interest depicted or represented in data from the other source. Metadata identifying that the object-of-interest from the two sources is the same object-of-interest is then stored for later use in, for example, object tracking.

Abstract: A gimbal rotation method includes controlling a driving assembly based on an angle command indicating a target angle to drive a gimbal to rotate to the target angle in a first time period having a pre-set length, determining whether a new angle command is received within the first time period, and, if not, estimating an estimated target angle based on gimbal rotation angles indicated by a plurality of previously-received angle commands and controlling the driving assembly to drive the gimbal to rotate from the target angle to the estimated target angle in a second time period having the pre-set length.

Abstract: An object tracking apparatus for a sequence of images, wherein a plurality of tracks have been obtained for the sequence of images, and each of the plurality of tracks is obtained by detecting an object in several images included in the sequence of images. The apparatus comprises matching track pair determining unit configured to determine a matching track pair from the plurality of tracks, wherein the matching track pair comprise a previous track and a subsequent track which correspond to the same object and are discontinuous, and combining unit configured to combine the previous track and the subsequent track included in the matching track pair.

Abstract: An image processing system has a memory storing a video depicting a multi-entity event, a trained reinforcement learning policy and a plurality of domain specific language functions. A graph formation module computes a representation of the video as a graph of nodes connected by edges. A trained machine learning system recognizes entities depicted in the video and recognizes attributes of the entities. Labels are added to the nodes of the graph according to the recognized entities and attributes. The trained machine learning system computes a predicted multi-entity event depicted in the video. For individual ones of the edges of the graph, select a domain specific language function from the plurality of domain specific language functions and assign it to the edge, the selection being made at least according to the reinforcement learning policy. An explanation is formed from the domain specific language functions.

Abstract: Various embodiments of the present invention provide systems and method for monitoring of physical movement in relation to regions where movement is either unconditionally or conditionally unauthorized.

Abstract: Provided is an information provision system capable of reducing secondary damage in the case of a disaster. The information provision system includes: a vehicle; and an information provision apparatus communicatively connected to the vehicle and configured to provide information to the vehicle. The information provision apparatus includes an acquirer that acquires fire-related information indicating information about fire, a fire detector that detects an occurrence of fire based on the fire-related information, and a fire information generator that generates, in response to the detection of the occurrence of fire by the fire detector, fire avoidance information for avoiding the fire, and transmits the fire avoidance information to the vehicle. The vehicle includes an information receiver that receives the fire avoidance information, and a display controller that causes a display to display the fire avoidance information.

Abstract: A method for controlling a movable object to track a target object includes determining a change in one or more features between a first image frame and a second image frame, and adjusting a movement of the movable object based on the change in the one or more features between the first image frame and the second image frame. The one or more features are associated with the target object, and the first image frame and the second image frame are captured at different points in time using an imaging device on the movable object.

Abstract: A device has a camera that is moveable by one or more actuators. During operation the camera moves. For example, the camera may move to follow a user as they move within the physical space. Mechanical limitations result in the camera movement exhibiting discontinuities, such as small jerks or steps from one orientation to another while panning. If the camera is acquiring video data while moving, the resulting video data may appear jittery and be unpleasant for a user to view. An offset is determined between an intended orientation of the camera at a specified time and an actual orientation of the camera at that time. A portion of raw image data acquired at the specified time is cropped using the offset to produce cropped image data that is free from jitter due to the movement discontinuities.

Abstract: A visual tracker can be configured to obtain profile data associated with a pose of a living entity. In response to detecting a person, a camera can be selected from cameras. Additionally, in response to selecting the camera, the system can receive video data from the camera representative of a stance of the person. Consequently, the stance of the person can be estimated, resulting in an estimated stance.

Abstract: Provided is a method for blending of agricultural product utilizing hyperspectral imaging. At least one region along a sample of agricultural product is scanned using at least one light source of different wavelengths. Hyperspectral images are generated from the at least one region. A spectral fingerprint for the sample of agricultural product is formed from the hyperspectral images. A plurality of samples of agricultural product is blended based on the spectral fingerprints of the samples according to parameters determined by executing a blending algorithm.

Abstract: Sample traceability device and method for medical research and/or diagnosis. The invention relates to a sample traceability device for medical research and/or diagnosis, comprising a one-dimensional or two-dimensional optical code, the device comprising a system for reading one-dimensional or two-dimensional optical codes and a sample tracing control device comprising a sample tracing database manager, and a user interface screen. Said traceability device also comprises: an area for depositing at least two samples, a system for illuminating the deposit area, at least one digital camera oriented towards said deposit area, and a device for processing the image, said image processing device comprising: a module for locating said optical codes, and a module for reading the located optical codes, the sample tracing control device automatically receiving the information generated by the module for reading the located optical codes.

Abstract: Compensating for delay in a Pan-Tilt-Zoom (PTZ) camera system is disclosed. Client-side view transformation is carried out to emulate a future Field Of View (FOV) of the camera so that the impact of latency is reduced.

Abstract: A camera device according to an aspect of the invention includes an imaging unit, an imaging direction adjustment unit, a direction control unit that controls the imaging direction adjustment unit, a camera-side tracking processing unit that analyzes captured image data to acquire first target information indicating the position of a tracking target and outputs the first target information, a camera-side communication unit that receives second target information from a terminal device, and a camera-side target information correction unit that corrects the first target information on the basis of the second target information.

Abstract: A method of controlling a wearable device, and which includes obtaining, by a camera of the wearable device, an image including at least a partial field of vision of a user; identifying, by a controller of the wearable device, an object appearing in the image obtained by the camera at a first time point; storing information about the identified object in a memory; detecting, via a microphone of the wearable device, a speech of a user using the wearable device at a second time point; identifying, via the controller, a word included in the detected speech of the user corresponds to the object identified in the image; retrieving, via the controller, the information about the object from the memory based on the identified word included in the detected speech; and outputting the retrieved information about the object.

Abstract: An inventory cache that contains a rack inventory state is created, with a hardware management system (HMS), for each of a plurality of components on the software-defined data center (SDDC) rack. The rack inventor state is stored at a database. During a restart of the HMS the rack inventory state at the database is accessed by the HMS and used to create the rack inventory state instead of performing a SDDC rack discovery by the HMS.

Abstract: A method for controlling falling of a virtual object is performed at an electronic device, the method including: displaying, in a virtual scene, a follow button in an area corresponding to a first virtual object, the follow button being used for controlling a second virtual object to follow the first virtual object to fall; detecting a touch operation on the follow button of the first virtual object; obtaining falling dynamic information of the first virtual object; and controlling the second virtual object to fall in the virtual scene according to the falling dynamic information.

Abstract: Provided is an electronic device including a communication unit configured to receive data from a server, a display unit configured to display a first video obtained by photographing a plurality of objects having cameras mounted thereon, or second videos obtained by the objects, an input unit configured to receive a user input, and a controller configured to display the first video on the display unit, receive a user input for selecting at least one of the plurality of objects included in the first video, through the input unit, and control the display unit to display a second video obtained by the selected object, on a part or entirety of the display unit based on the user input.

Abstract: An apparatus comprises a receiver (401) receiving a first image and associated first depth data captured by a first depth-sensing camera. A detector (405) detects an image position property for a fiducial marker in the first image, the fiducial marker representing a placement of a second depth sensing image camera. A placement processor (407) determines a relative placement vector indicative of a placement of the second depth sensing image camera relative to the first depth-sensing camera in response to the image position property and depth data of the first depth data for an image position of the fiducial marker. A second receiver (403) receives a second image and second first depth data captured by the second depth sensing image camera. A generator (409) generates the representation of at least part the scene in response to a combination of at least the first image and the second image based on the relative placement vector.

Abstract: A variety of applications involve transmission of video data in real time. However, bandwidth limitations may result in degradation of the quality of the transmitted video. To address such limitations, a video stream may be limited to ‘regions of interest’ within each frame of the video stream. This can be accomplished by setting areas outside of the regions of interest to an a priori known color or ‘blanking’ color, and the resulting ‘blanked’ video stream can be compressed and transmitted without extensive modifications to existing transmission apparatus. The original video can be reconstituted by superimposing the received region-of-interest video data onto a background image of the environment that contains the regions-of-interest. The background image can be satellite imagery or other stored imagery that matches the perspective of the video stream. Geographical tag data can be used to superimpose the region-of-interest video data onto the background image.

Abstract: A semiconductor device for an apparatus having a movement drive unit and an image unit, includes an image detection unit, an image recognition unit and control unit. The image detection unit detects an object in a captured image and cuts out an image area including the object from the captured image as an object detection area image. The image recognition unit performs an image recognition processing for the object detection image area and output a recognition probability of the object. The control unit controls at least one of a moving speed of the movement drive unit and an imaging interval of the image unit based on the recognition probability.

Abstract: In one implementation, a virtual-world simulator includes a computing platform having a hardware processor and a memory storing a software code, a tracking system communicatively coupled to the computing platform, and a projection device communicatively coupled to the computing platform. The hardware processor is configured to execute the software code to obtain a map of a geometry of a real-world venue including the virtual-world simulator, to identify one or more virtual effects for display in the real-world venue, and to use the tracking system to track a moving perspective of one of a user in the real-world venue or a camera in the real-world venue. The hardware processor is further configured to execute the software code to control the projection device to simulate a virtual-world by conforming the identified one or more virtual effects to the geometry of the real-world venue from a present vantage point of the tracked moving perspective.

Abstract: An electronic apparatus and method for image-based mask frame interpolation is provided. The electronic apparatus receives a sequence of image frames, including a group of image frames and an intermediate image frame. The electronic apparatus generates foreground (FG) masks based on application of a trained Neural Network on the group of image frames and selects a Region-of-Interest (RoI) from a first image frame of the group of image frames. The electronic apparatus extracts first feature vectors for pixels within the selected RoI and second feature vector for a first pixel of the intermediate image frame. The electronic apparatus estimates a first mask value for the first pixel based on application of a statistical function on the first feature vectors and the second feature vector and generates an intermediate FG mask based on the estimated first mask value and re-estimation of mask values for remaining pixels of the intermediate image frame.

Abstract: The present invention provides methods and systems of tracking a target with an imaging device onboard an unmanned aerial vehicle. The method may comprise identifying one or more targets in a first image captured by the imaging device, receiving a user selection of a target to be tracked from the one or more targets in the one or more images, generating target information for the target to be tracked based on the user selection of the target, detecting a deviation of a detected position and/or a detected size of the target in a second image captured by the imaging device from the selected position and/or the selected size of the target in the first image and adjusting at least one of the UAV and the carrier, to reduce the determined deviation to maintain the target substantially within the field of view of the image device, based on the deviation.

Abstract: Disclosed herein are system, method, and computer program product embodiments for camera surveillance planning and tracking using, for example, infrared technologies. An embodiment operates by receiving a camera specification input and placement input, calculating a surveillance coverage based on the camera specification and the placement input, displaying the surveillance coverage on a map, and identifying at least one stereo pair.

Abstract: Provided is an electronic device including a communication unit configured to receive data from a server, a display unit configured to display a first video obtained by photographing a plurality of objects having cameras mounted thereon, or second videos obtained by the objects, an input unit configured to receive a user input, and a controller configured to display the first video on the display unit, receive a user input for selecting at least one of the plurality of objects included in the first video, through the input unit, and control the display unit to display a second video obtained by the selected object, on a part or entirety of the display unit based on the user input.

Abstract: A method for object tracking. The method includes capturing a sequence of images of a scene, detecting, by a hardware processor, based on a pattern of local light change across the sequence of images, a reflective light source in the scene, comparing, by the hardware processor in response to detecting the reflective light source, a location of the reflective light source in at least one image of the sequence of images and a target position within the at least one image to generate a result, and generating, by the hardware processor based on the result, a control signal for changing a field-of-view of a camera device such that the reflective light source substantially aligns with the target position within the field-of-view, wherein the reflective light source emits an object-reflected light.

Abstract: For point-of-interest determination and display, a processor detects an image event during a videoconference. The processor determines a point-of-interest for the video image of the videoconference based on the image event. The video image is at least a 180-degree image and the point-of-interest is a portion of the video image. The processor displays the point-of-interest from the video image on a display.

Abstract: Disclosed is an electronic device comprising a communication circuit, a memory, and a processor. The processor may be set to obtain an omnidirectional image, identify at least one face in the omnidirectional image, obtain at least one gaze direction from the at least one face, and generate at least one region of interest of the omnidirectional image on the basis of the at least one gaze direction. Also, various other embodiments specified throughout the present specification may also be possible.

Abstract: An information processing device according to one aspect of the present invention includes at least one memory storing instructions; and at least one processor coupled to the memory and configured to execute the instructions to: extract a candidate point which is a point contributing to a signal at a target point, based on a position of the target point in a three-dimensional space and a shape of an observed object, the target point being a point specified in an intensity map of the signal from the observed object, the intensity map being acquired by a radar; and generate an image indicating a position of the candidate point in a spatial image capturing the observed object.

Abstract: Provided is an analysis apparatus (10) including a person extraction unit (11) that analyzes video data to extract a person, a time calculation unit (12) that calculates a continuous appearance time period for which the extracted person has been continuously present in a predetermined area and a reappearance time interval until the extracted person reappears in the predetermined area for each extracted person, and an inference unit (13) that infers a characteristic of the extracted person on the basis of the continuous appearance time period and the reappearance time interval.

Abstract: System, methods, and other embodiments described herein relate to self-supervised training of a depth model for monocular depth estimation. In one embodiment, a method includes processing a first image of a pair according to the depth model to generate a depth map. The method includes processing the first image and a second image of the pair according to a pose model to generate a transformation that defines a relationship between the pair. The pair of images are separate frames depicting a scene of a monocular video. The method includes generating a monocular loss and a pose loss, the pose loss including at least a velocity component that accounts for motion of a camera between the training images. The method includes updating the pose model according to the pose loss and the depth model according to the monocular loss to improve scale awareness of the depth model in producing depth estimates.

Abstract: A method and an image processing system for detecting an object in an image are described. A set of line segments are detected in the image. A subset of the line segments is identified based on a projection space orientation that defines a projection space. Each one of the line segments of the subset of line segments is projected into the projection space to obtain a set of projected line segments, where each projected line segment of the set of projected line segments is represented by a respective set of projection parameters. A determination is performed, in the projection space, based on the sets of projection parameters and a shape criterion that characterizes the object, of whether the image includes an instance of the object. In response to determining that the image includes the instance of the object, the instance of the object is output.

Abstract: An object detection device calculates, for each of images acquired in time-series, a confidence score of each of a plurality of types of an object to be detected for each of a plurality of regions on the image, detects the type of the object corresponding to the confidence score in a region in which the confidence score is equal to or higher than a confidence threshold of the type for each of the images, tracks the detected object, counts a frequency of occurrence for each type of the detected object in a period in which the detected object is tracked, and updates the confidence threshold in such a way that the confidence threshold of a type having a higher frequency of occurrence is lower than the confidence threshold of a type having a lower frequency of occurrence.

Abstract: A method of authenticating physical activity in a monitored space, comprising: obtaining monitored space activity data from at least one activity detector deployed in the monitored space to capture simultaneously physical activity of a plurality of moving objects in said monitored space; analyzing said monitored space activity data to identify a plurality of spatiotemporal movement patterns in said monitored space; obtaining a plurality of inertial data patterns from each one of a plurality of applications executed in a plurality of mobile devices, each one of said inertial data patterns is recorded using an inertial sensing device installed in one of said plurality of mobile devices, each one of said plurality of mobile devices is associated with one of a plurality of persons; identifying at least one correlation between at least one of said plurality of inertial data patterns and at least one of said plurality of spatiotemporal movement patterns; authenticating said physical activity according to said at le

Abstract: Disclosed are a method and apparatus for composing a plurality of images of the celestial bodies. The method and apparatus can compose a plurality of images using various pieces of information associated with the plurality of images and a time interval during the plurality of images are taken, such that objects are accurately matched with no afterimage.

Abstract: A surveillance system of at least one site including videosurveillance cameras, the surveillance system including at least one surveillance camera management module and an artificial neural network for predicting the location of a target in a zone supervised by a camera, connected to the surveillance camera management module, the predicting artificial neural network including a target information acquisition input including data for prediction including data for identifying the camera in which a target has been detected, for positioning the target in the output state. The predicting artificial neural network includes an output of at least one identification of a probable camera the target of which will be probably identified and the camera management module includes an output for transmitting the at least one identification of probable camera to a member including a screen.

Abstract: A method and a controller for controlling a video processing unit to facilitate detection of newcomers in a first environment. The method comprises: capturing a thermal image of a human object in the first environment, the first environment being associated with a first climate; calculating, based on the thermal image, a thermal signature of a portion of the human object; determining that the human object has entered the first environment from a second environment when the thermal signature of the portion of the human object deviates from a predetermined thermal signature associated with the first environment, wherein the second environment is associated with a second, different, climate; and controlling the video processing unit to prioritize the human object over other human objects when processing video frames depicting the human object together with the other human objects.

Abstract: The present embodiments relate to improvements to audio/video (A/V) recording and communication devices, including improved approaches to using a neighborhood alert mode for triggering multi-device recording, to a multi-camera motion tracking process, and to a multi-camera event stitching process to create a series of “storyboard” images for activity taking place across the fields of view of multiple cameras, within a predetermined time period, for the A/V recording and communication devices.

Abstract: There is provided an image processing apparatus. A determination unit determines a movement direction of an object. A setting unit sets, within a shooting range, a plurality of processing areas that are arranged in a different direction from the movement direction. A selection unit selects a tracking point in each processing area of a predetermined shot image. A tracking unit tracks, inside each processing area, the tracking point across one or more shot images that are shot after the predetermined shot image.

Abstract: The systems and methods generally relate to blood speckle imaging. The systems and methods transmit a plurality of transmit (Tx) beams from an ultrasound probe to a region of interest (ROI). The plurality of Tx beams are transmitted successively with respect to each other. At least two of the plurality of Tx beams overlap at a common position. The systems and methods estimate velocity fields from the speckle tracking, and weight the velocity fields based on weighting signals to form weighted velocity fields. The weighting signal being centered at transmit beam axes of the plurality of Tx beams. The systems and methods interpolate the weighted velocity fields that overlap at the common position to adjust a portion of the weighted velocity fields, and generate an image on a display of a combined velocity field by combining the weighted velocity fields.

Abstract: In a three-dimensional measurement apparatus, a plurality of cameras and an arithmetic device are connected via a network, and includes: a plurality of photographing units; an image acquisition unit; a local time acquisition unit; a moving object detection unit to detect a moving object from the image; an image holding unit to hold an image in which the moving object is detected, in association with the local time at which the image has been photographed; an image selection unit; a three-dimensional position/synchronization deviation estimation unit to estimate a three-dimensional position of the moving object and a synchronization deviation between cameras, from an image selected by the image selection unit and a local time associated with the image; a synchronization deviation holding unit to hold the synchronization deviation; and an output unit to output an estimation result of the three-dimensional position/synchronization deviation estimation unit.

Abstract: Embodiments of the present specification provide abnormality detection methods, apparatuses, and devices for unmanned checkout. In an embodiment, a method comprising: obtaining, by a depth camera in a detection area of a photographing area for unmanned checkout, image data that comprises a depth image and a corresponding RGB image; detecting, based on the image data, one or more customers that enter the detection area for unmanned checkout using one or more of headcount detection to detect a quantity of the one or more customers or gesture detection to detect one or more gestures of the one or more customers; and in response to determining that the quantity of the one or more customers is greater than a predetermined threshold or the one or more gestures comprise an abnormal gesture, notifying a central control system to stop the unmanned checkout and trigger an alarm.

Abstract: The present invention reduces the amount of data outputted while maintaining the accuracy of an analysis process with a small delay amount, by an image analysis apparatus provided with: a deduction unit that deduces a second quality concerning an object in a second image, which is different from a first image associated with object data relating to an object to be inputted, on the basis of a first quality concerning the object relating to the object data and on the basis of the state of the object in the second image, the state being obtained by using a state model for deducing the position and the size of the object from the object data, while using a quality model for deducing the second quality concerning the object; and a determination unit that determines whether to use the object data for analysis on the basis of the deduced second quality.

Abstract: A method and system can generate video content from a video. The method and system can include a coordinator, an image detector, and an object recognizer. The coordinator can be communicatively coupled to a splitter and/or to a plurality of demultiplexer nodes. The splitter can be configured to segment the video. The demultiplexer nodes can be configured to extract audio files from the video and/or to extract still frame images from the video. The image detector can be configured to detect images of objects in the still frame images. The object recognizer can be configured to compare an image of an object to a fractal. The recognizer can be further configured to update the fractal with the image. The coordinator can be configured to embed metadata about the object into the video.

Abstract: A method for training a person recognition model using images from a camera 100, wherein the method has at least a reading-in step in which a detection signal 135 representing a detected person within a monitoring area of at least the camera 100 in a camera network is read in. The method also has at least a collecting step in which a plurality of image signals 140 from the camera 100 are collected using the detection signal 135 which has been read in, wherein the collected image signals 140 represent a recorded image section from each image from the camera 100. Finally, the method has at least an adapting step in which the person recognition model is adapted using the collected image signals 140 in order to recognize the detected person in an image from the camera 100 or from an at least second camera in the camera network.

Abstract: A method of depth estimation utilizing heterogeneous cameras, comprising, homogenizing a first camera image and a second camera image based on a first camera calibration dataset and a second camera calibration dataset respectively, wherein the first camera image and second camera image are distortion corrected and are zoom compensated, determining an initial image pair rectification transformation matrix of the homogenized first camera image and the homogenized second camera image, determining a delta image pair rectification transformation matrix based on the initial image pair rectification transformation matrix, determining a final image pair rectification transformation matrix based on the initial image pair rectification transformation matrix and the delta image pair rectification transformation matrix resulting in a final rectified image pair and disparity mapping the final rectified image pair based on a depth net regression.

Abstract: A method and apparatus for controlling an operation of a camera that allows a user to conveniently control a camera apparatus according to a gesture of a subject input through a lens of a camera, and the camera apparatus are provided. The method includes receiving an image input through a camera lens; generating an image frame; detecting a motion of a subject included in the image frame by comparing the image frame with at least one previous frame stored before the image frame is generated; determining whether the motion of the subject is a User Interface (UI) gesture; and performing, if the motion is the UI gesture, an operation corresponding to the UI gesture.