Abstract: A method and system to receive information from at least one multivariable sensor, each multivariable sensor being deployed in an environment, having internet connectivity to communicate with at least one other device over the internet, and selectively determining at least one attribute of multiple events in its environment; receive an indication of a location of the multivariable sensor; receive an indication of a location of an augmented reality device; determine an alarm based on the received information from the at least one multivariable sensor; determine a location for the alarm and a location of a solution associated with the alarm; and present, in a field of view display on the augmented reality device, a visualization of the determined alarm and at least one of the determined location for the alarm and the determined location for the solution associated with the alarm.

Abstract: Methods and systems for automatically augmenting an endoscope video for a task based on status data that describes the status of a surgical robot system that was used to at least partially perform the task. The system comprises an event detector and an augmenter. The event detector is configured to: receive the status data; and identify one or more patterns in the status data that indicate an event occurred during the task. The augmenter is configured to receive an endoscope video captured during the task, the endoscope video being time synchronised with the status data; and in response to the event detector identifying a pattern in the status data that indicates an event occurred, automatically augment metadata that forms part of or is associated with the endoscope video with information identifying the event and the time the event occurred with respect to the endoscope video.

Abstract: Introduced here are techniques for incentivizing the consumption of foodstuffs through the use of augmented reality features. A parent may initiate an application on a mobile phone and scan a plate of one or more foodstuffs by initiating a live view captured by a camera. The parent can then select a given foodstuff that a child is unlikely to consume. When the child views the given foodstuff through the mobile phone, a portion of an augmented reality feature could be shown. For example, the child may be able to see the tail of a creature, a corner of a item, etc. As the child consumes the given foodstuff, additional portion(s) of the augmented reality feature are exposed, and the augmented reality feature becomes increasingly visible. Such a technique incentivizes the child to continue eating the given foodstuff so that the augmented reality feature can be fully seen.

Abstract: A method, apparatus, and computer program product are therefore provided for disambiguating non-deterministic spoken dialog commands as they relate to points-of-interest and information regarding points-of-interest within or near the field of view of a user. An apparatus may include a processor and a memory including program code instructions.

Abstract: The application relates to an appliance (1) for creating and monitoring a safety area (2) in a working space (3). The suggested appliance (1) comprises a recording device (4). The recording device (4) is configured to record images in a monitoring area (8) which encompasses the safety area (2). The appliance (1) further comprises an illumination device (5) which is configured to project lines or patterns which define the safety area (2), into the working space (3). Furthermore, the appliance (1) comprises an evaluation device (9) which is configured to evaluate the images which are recorded by the recording device (4), with respect to a violation of the safety area (2). The illumination device (5) is configured to project a temporally coded sequence (11, 11?, 11?, 11??, 11???) of the lines (6) or patterns into the working space (3).

Abstract: Methods and apparatus to measure brand exposure in media streams are disclosed. An example method to determine brand exposures included in media content disclosed herein comprises determining whether a scene detected from a media stream corresponding to the media content matches a reference scene, identifying an expected region of interest in the detected scene based on information describing the reference scene when the detected scene is determined to match the reference scene and the reference scene is not specified to be a scene of no interest, and determining whether a reference brand identifier associated with the reference scene is included in the expected region of interest identified in the detected scene.

Abstract: A method for highlight extraction capable of automatically extracting a highlight from a video including a sporting event is provided. The method for highlight extraction may include: identifying a video including a sporting event, log information that sequentially records events occurring in the sporting event, and a keyword related to the video; tagging the video with game information related to the video; extracting at least one piece of log information corresponding to the keyword and determining at least one frame that corresponds to the log information extracted from the tagged video; and creating a highlight video by combining the at least one determined frame.

Abstract: A video-related method, system and device are disclosed herein. The method, system and device, in an embodiment, involve processing geographic information associated with participants and processing rating data related to videos. The method, system and device also involve displaying a map interface that displays symbols representing the participants. The symbols vary based, at least in part, on differences in the rating data.

Abstract: According to various embodiments, a system for accessing video content is disclosed. The system includes one or processors on a video hosting platform for hosting the video content, where the processors are configured to generate an automated transcription of the video content and apply text clustering modules based on a trained neural network to segment the video content.

Abstract: A store management server configured to communicate with a first gate control device that controls a first gate installed in a store includes a network interface, and a processor configured to identify a customer entering the store, acquire customer information about the detected customer, detect that the customer is holding a commodity in the store, acquire sales restriction information about the commodity held by the customer, determine whether the customer is eligible to buy the commodity held by the customer based on the acquired customer information and sales restriction information, and control the network interface to transmit, when it is not determined that the customer is eligible to buy the commodity, a first command to the first gate control device. The first command causes the first gate control device to close the first gate.

Abstract: In some embodiments, a first homography, created from two images of a roadway, is decomposed to determine an ego-motion, and the ego-motion is used to adjust a previous estimate of a road plane. The adjusted previous estimate of the road plane is combined with the current estimate of the plane to create a second homography, and the second homography is used to determine residual motion and vertical deviation in the surface of the roadway. In some embodiments, multiple road profiles each corresponding to a common portion of a roadway are adjusted in slope and offset by optimizing a function having a data term, a smoothness term and a regularization term; and the adjusted road profiles are combined into a multi-frame road profile. In some embodiments, road profile information for a predetermined number of data points is transmitted in periodic data bursts, with more than one data point per data burst.

Abstract: Methods, systems, and computer program products for providing automatic responsive actions to biometrically detected events are provided. Aspects include receiving, from a biometric sensing device associated with an animal, biometric data associated with the animal and animal location data associated with the animal. Aspects also include determining an intensity level of an event associated with the animal based on the biometric data and approximating a location of the event based on the animal location data and the biometric data. In response to determining that the intensity level of the event exceeds a predetermined intensity threshold, aspects include initiating a responsive action proximate to the location of the event.

Abstract: The present technology relates to an image processing device, an image processing method, and a program capable of making it easier to recognize standing objects. Movement transformation of moving a subject position where a subject appears in a target image to be processed is performed, depending on a subject distance from a vanishing point in the target image to the subject position. The present technology can be applied to, for example, the image processing and the like of an image taken by a camera unit onboard a vehicle or other moving body.

Abstract: A system for providing traffic information to a driver of an ego vehicle includes at least one sensor that provides image data of detected objects, including other vehicles, in a surrounding area of the ego vehicle; a position detection device that detects a position of the ego vehicle; a display unit that renders visual information relating to an environmental model of the surrounding area of the ego vehicle, the position, and corresponding digital map data; and an image analysis unit configured to determine a vehicle characteristic of at least one other vehicle in the surrounding area of the ego vehicle from a vehicle image of the at least one other vehicle generated based on the image data from the at least one sensor and to display information on the display unit based on the determined vehicle characteristic.

Abstract: A method for detecting the presence of an emergency vehicle includes receiving a plurality of image frames over a period of time, determining an EV colour component for each image frame based on the ratio of a first colour relative to the total colour in each pixel and assigning to a pixel a first value if the EV colour component exceeds a predefined threshold value and a second value if the EV colour component does not, and determining for an EV colour value for the first colour based on the sum of all of the first values for each image frame. The method also includes generating a time domain representation, converting the time domain representation for the plurality of image frames to a frequency spectrum, and determining if any flashing light sources of the first colour associated with one or more types of emergency vehicles is present.

Abstract: A system for detecting the face of a vehicle driver comprises an instrument cluster for the driver including a carrier and an dashboard directed towards the driver and covered by a glass, said dashboard being embedded in the carrier; an image capture device comprising a camera configured to capture sequential images of the driver's gaze; and a controller configured to analyze the driver's gaze; at least one intrinsic element of the instrument cluster defines a reflecting surface configured to directly reflect the sequential images of the driver's gaze such as a glossy plastic, a chrome-plated element, a display screen or the dashboard glass; the camera comprises a field of view oriented towards the reflecting surface so as to indirectly capture the images of the driver's gaze.

Abstract: A method or system is able to refine Global Positioning System (“GPS”) information for guiding a vehicle via extracted metadata using a GPS refinement (“GR”) model managed by a virtuous cycle containing sensors, machine learning center (“MLC”), and a cloud based network (“CBN”). The GR system includes a set of outward facing cameras, a vehicle onboard computer (“VOC”), and GR model. The outward facing cameras mounted on a vehicle are capable of collecting external images representing a surrounding environment in which the vehicle operates. The VOC is configured to generate a positional vehicle location with respect to the surrounding environment in accordance with the external images and historical stored data obtained from CBN. The GR model is configured to generate a driving guidance based on combined information between the positional vehicle location and GPS data.

Abstract: A method for driver identification including recording a first image of a vehicle driver; extracting a set of values for a set of facial features of the vehicle driver from the first image; determining a filtering parameter; selecting a cluster of driver identifiers from a set of clusters, based on the filtering parameter; computing a probability that the set of values is associated with each driver identifier of the cluster; determining, at the vehicle sensor system, driving characterization data for the driving session; and in response to the computed probability exceeding a first threshold probability: determining that the new set of values corresponds to one driver identifier within the selected cluster, and associating the driving characterization data with the one driver identifier.

Abstract: Various example implementations of the present invention are directed towards systems and methods to quantify biometric acquisition and identification. A test facility evaluates the acquisition by a biometric station of biometric information. Evaluations can relate to biometric information acquisition speed, rates of acquisition failure, rates of biometric information extraction failure, rates of biometric match failure, calculating a true identification rate, and/or calculating other metrics related to quantifying biometric acquisition and identification. The test facility can calculate an efficiency metric and an effectiveness metric of the quantification determinations, and output such results.

Abstract: Various example implementations of the present invention are directed towards systems and methods to quantify biometric acquisition and identification. A test facility evaluates the acquisition by a biometric station of biometric information. Evaluations can relate to biometric information acquisition speed, rates of acquisition failure, rates of biometric information extraction failure, rates of biometric match failure, calculating a true identification rate, and/or calculating other metrics related to quantifying biometric acquisition and identification. The test facility can calculate an efficiency metric and an effectiveness metric of the quantification determinations, and output such results.

Abstract: An image search device includes an extractor configured to acquire an entire image including an area image that is an image of a target object and extract one or more combinations of the area image and attribute information from the entire image on the basis of color, an information storage unit configured to search for the area image combined with the designated attribute information from the extracted one or more combinations of the area image and the attribute information, and an interface configured to output a search result.

Abstract: A system for holding an image display apparatus (60) for displaying an image captured by means of an image capturing apparatus (20) comprises a movable holding apparatus (70) for an alterable hold of the image display apparatus (60), a controllable drive device (72) for moving the holding apparatus (70), comprising a control signal input (74) for receiving a control signal, and a controller (40) comprising a signal input (42) for receiving a signal that represents an orientation or a change in the orientation of the viewing direction (28) of the image capturing apparatus (20) in space or that facilitates a determination of the orientation or the change in the orientation of the viewing direction (28) of the image capturing apparatus (20), and comprising a control signal output (47), couplable to the control signal input (74) of the controllable drive device (72), for providing a control signal for controlling the controllable drive device (72).

Abstract: The present invention is a fiducial marker (1) comprising an external region (11), an inner region (12) positioned inside said external region (11) and which has a substantially contrast color with respect to the external region (11), and at least one pattern element (131) positioned inside said inner region (12) and which has a substantially contrast color with respect to the inner region (12); in order to be sensed by means of an image sensor (2). Said fiducial marker (1) is characterized in that the external region (11) has a polygonal periphery, and the inner region (12) has a substantially circular or elliptical periphery, and said pattern elements (131) have a circular or elliptical form.

Abstract: Various client devices include displays and one or more image capture devices configured to capture video data. Different users of an online system may authorize client devices to exchange information captured by their respective image capture devices. A user viewing video data captured by another user's client device identifies an object of interest in the video data to the other user's client device. The other user's client device modifies captured video data so a focal point of the captured video data is the object of interest and so the object of interest is magnified in the captured video data. Subsequently, the modified video data is transmitted to the client device of the user viewing the captured video data.

Abstract: Systems and methods for digitized document image data spillage recovery are provided. One or more memories may be coupled to one or more processors, the one or more memories including instructions operable to be executed by the one or more processors. The one or more processors may be configured to capture an image; process the image through at least a first pass to generate a first contour; remove a preprinted bounding region of the first contour to retain text; generate one or more pixel blobs by applying one or more filters to smudge the text; identify the one or more pixel blobs that straddle one or more boundaries of the first contour; resize the first contour to enclose spillage of the one or more pixel blobs; overlay the text from the image within the resized contour; and apply pixel masking to the resized contour.

Abstract: Method for the calculation resource-saving semantic segmentation of image data of an imaging sensor with the aid of an artificial neural network, in particular, of a convolutional neural network, the artificial neural network including an encoder path, a decoder path, the encoder path transitioning into the decoder path, the transition taking place via a discriminative path, the following steps taking place in the discriminative path: dividing an input tensor as a function of a division function into at least one first slice tensor and at least one second slice tensor, the input tensor originating from the encoder path; connecting the at least one first slice tensor to the at least one second slice tensor as a function of a connection function in order to obtain a class tensor; and outputting the class tensor to the decoder path of the neural network

Abstract: A method for the calculation resource-saving semantic segmentation of image data of an imaging sensor with an artificial neural network, in particular, of a convolutional neural network, the artificial neural network including an encoder path, a decoder path (and a skip component), including: initial connection (merge) of an input tensor to a skip tensor with an initial connection (merge) function/connection instruction to obtain a merged tensor, the input tensor and the skip tensor being dependent on the image data; application of a function of a neural network, in particular, of a convolution to the merged tensor to obtain a proof reader tensor; second connection (merge) of the proof reader tensor to the input tensor with a second connection (merge) function/connection instruction to obtain an output tensor; outputting the output tensor to the decoder path of the artificial neural network.

Abstract: A method for determining a spatial position of an object includes obtaining an image with at least one camera. The object is identified in the image. At least one of a pixel size of the object in the image and a pixel offset of the object from a center of an image plane of the at least one camera is determined. A distance between the object and the camera image plane is determined using either the pixel size and the pixel offset. The spatial position is determined using the distance and at least one known distance between the object and another element of the image.

Abstract: In an object determination device, an image recognition score calculated from an image acquired by a camera is associated with measured distance values acquired by a ranging sensor. The measured distance values are grouped and an object is generated, and the measured distance values and image recognition score of said object are stored in an object table. Then, the object is recognized as a person or as a thing on the basis of a measured distance likelihood, an image recognition likelihood and a match ratio calculated from the values of the object table. Therefore, the image recognition score of the image acquired from the camera and the measured distance values acquired from the ranging sensor are combined and the object is recognized as a person or as a thing, thus making it possible to perform said recognition processing with high speed and high accuracy.

Abstract: A method for reinforcement learning adopting a centering operation using a weight corresponding to a behavior probability is provided. The method includes steps of: a computing device instructing a reinforcement learning agent to (a) determine a k-th estimated parameter by referring to (i) a probability distribution for estimating the k-th behavior if k=1, and (ii) the probability distribution generated by selecting a (k?1)-th behavior if k>1; (b) select one of N behavior candidates as the k-th behavior by referring to (i) the k-th estimated parameter, and (ii) k-th contexts; (c) if k-th behavior probabilities are acquired, (i) generate k-th weighted expectation value of the k-th contexts, and (ii) generate k-th adjusted contexts by applying the centering operation to the k-th contexts; and (d) if a k-th reward is acquired, generate the probability distribution for estimating a (k+1)-th behavior using the k-th adjusted contexts, the k-th behavior probabilities, and the k-th reward.

Abstract: A method and an apparatus for generating vehicle damage information are provided. The method includes: acquiring a to-be-processed vehicle image; for a target detection model in at least one pre-trained target detection model: inputting the vehicle image to the target detection model to generate a suspected damage area detection result; and determining a location of a suspected damage area in the vehicle image based on the generated suspected damage area detection result. A mechanism for detecting a suspected damage area is provided based on the target detection model, improving the vehicle damage assessment efficiency.

Abstract: A method for detecting a point of interest (POI) change in a pair of inputted POI images. A first processor of the method: generates triplets of training POI images using a base of training POI images and trains a convolutional neural network (CNN) of three-stream Siamese type based on the triplets of training POI images. A second processor of the method: computes, for each image of the pair of inputted POI images, a descriptor of that image using a stream of the CNN of three-stream Siamese type, computes a similarity score based on the descriptors of the images of the pair of inputted POI images using a similarity score function, and selectively detects the POI change based on the similarity score. A third processor of the method generates the base of training POI images to include an initial set of POI images and a set of synthetic POI images.

Abstract: Disclosed herein is a system and method for performing on-demand supervised learning-based data analytics directly on blockchain without the need to extract data from the blockchain to an external intermediary platform. The system and method disclosed herein enable supervised learning-based analysis directly on the blockchain, for data records stored on the blockchain. Furthermore, the system and methods disclosed herein allows a training and validation of supervised learning-based algorithms using data records from the transaction blocks of the blockchain without the need of an intermediary platform external to the blockchain.

Abstract: An identification device has a processor configured to carry out plural identification processing by which an individual is identified based on plural acquired data different from each other indicating the individual and, when the identification of the individual by one or more identification processing of the plural identification processing fails and the identification of the individual by one or more other identification processing of the plural identification processing succeeds, learn the at least one identification processing by which the identification of the individual fails.

Abstract: Various embodiments are generally directed to techniques of adjusting the combination of the samples in a training batch or training set. Embodiments include techniques to determine an accuracy for each class of a classification model, for example. Based on the determined accuracies, the combination of the samples in the training batch may be adjusted or modified to improve the training of the classification model.

Abstract: The present invention relates to a parameter training method for a convolutional neural network, CNN, for detecting items of interest visible in images by data processing means (11a, 11b, 11c) of at least one server (1a, 1b, 1c), the method being characterized in that it is implemented based on a plurality of training image databases, wherein said items of interest are already annotated, the CNN being a CNN common to said plurality of training image databases and having a common core and a plurality of encoding layers, each one specific to one of said plurality of training image databases. The present invention also relates to a method for detecting items of interest visible in an image.

Abstract: The present invention relates to a method for association of items of interest visible in a video consisting of a sequence of K images, characterized in that it comprises the implementation by data processing means (21) of a terminal (2), of steps of: (a) Association of each item of interest from a first category visible in a kth image of said video with an item of interest from a second category, different from the first category, visible in said kth image; (b) Calculation of an association cost for a plurality of pairs of an item of interest from the first category visible in at least one image of the video with an item of interest from the second category visible in at least one image of the video, depending on at least the counters of association of pairs of one item of interest from the first category with one item of interest from the second category; (c) Use of a combinatorial optimization algorithm depending on the calculated association costs so as to reassociate each item of interest from the firs

Abstract: Computer vision systems and methods for detecting anomalous building models are provided. The systems and methods can detect anomalies in building models using one or more of an independent univariate Gaussian algorithm, a multivariate Gaussian algorithm, a combination of a multivariate Gaussian algorithm for continuous features and a frequency histogram algorithm for discrete features, and/or a bin frequency model. The system automatically processes computerized models to determine anomalies, and indicates whether the models are accurate and whether correction is required.

Abstract: An applicator device includes a camera, a print head, and a driver. The camera captures an image of an application surface of an application target. The print head applies a material to the application surface. The driver causes the print head to move between a first position where the print head does not block a field of view of the capturing, by the camera, of the image of the application surface and a second position where the print head blocks the field of view of the capturing, by the camera, of the image of the application surface. The print head applies the material to the application surface when the print head is at the second position, based on the image of the application surface captured by the camera when the print head is at the first position.

Abstract: Techniques are disclosed for providing information describing a fault code for an object. A display device having an electronic display detects a fault occurrence event in the object. The display device determines a fault code indicative of the fault occurrence event and one or more parameters associated with the fault code. The display device generates a machine-readable code representing the fault code and the one or more parameters.

Abstract: A collar comprising a top surface having a legend capable of being read by a reading device, a collar body and an inner wall configured to surround and secure the collar to an object placed within a slot defined by the inner wall.

Abstract: A transaction card may include a body having a first part connectable with a second part, and a magnetic stripe on a first main side of the body. A first section of the magnetic stripe is provided on the first part of the body, and a second section of the magnetic stripe is provided on the second part of the body. The transaction card may further include an engagement member extending from the first part of the body for retention within a slot of the second part of the body, the slot embedded within the body between the first main side and a second main side.

Abstract: A transaction card may include a body having a first part connectable with a second part, and a magnetic stripe on a first main side of the body. A first section of the magnetic stripe is provided on the first part of the body, and a second section of the magnetic stripe is provided on the second part of the body. The transaction card may further include an engagement member extending from the first part of the body for retention within a slot of the second part of the body, the slot embedded within the body between the first main side and a second main side.

Abstract: A tire includes: a carcass ply extending from one bead core to another bead core; side-wall rubber arranged at an outer side in the tire-width direction of the carcass ply; a belt arranged at an outer side in the tire-radial direction of the carcass ply in the tread; and a pad arranged between the carcass ply and an end region in the tire-width direction of the belt, in which the pad extends up to an outer side in the tire-width direction, the pad and side-wall rubber is laminated at part of the carcass ply, and an electronic component is embedded between the pad and side-wall rubber.

Abstract: An energy-efficient sequencer comprising inline multipliers and adders causes a read source that contains matching values to output an enable signal to enable a data item prior to using a multiplier to multiply the data item with a weight to obtain a product for use in a matrix-multiplication in hardware. A second enable signal causes the output to be written to the data item.

Abstract: The present invention relates to a method of secure classification of an input data by means of a convolutional neural network, CNN, the method comprising the implementation by data processing means (11a, 11b, 11c, 21) of at least one device (1a, 1b, 1c, 2), of steps of: (a) Determination, by application of said CNN to said input data, of a first classification vector of said input data associating with each of a plurality of potential classes a representative integer score of the probability of said input data belonging to the potential class, the first vector corresponding to one possible vector among a first finite and countable set of possible vectors, each possible vector of the first set associating with each of the plurality of potential classes an integer score such that said scores of the possible vector constitute a composition of a predefined whole total value; (b) Construction, from the first vector, of a second classification vector of said input data, such that the second vector also belongs to

Abstract: A hybrid deep-learning action prediction architecture system is described that predicts actions. The architecture includes a main path and an auxiliary path. The main path may contain multiple layers of convolutional neural networks for further aggregation to coarser time spans. The resultant data produced by the convolutional neural networks is passed to multiple layers of LSTMs. The outputs from LSTMs are then combined with the profile in the auxiliary path to predict an action label.

Abstract: A method for local approximation of a predictive model may include receiving unclassified data associated with a plurality of unclassified data items. The unclassified data may be classified based on a first predictive model to generate classified data. A first data item may be selected from the classified data. A plurality of generated data items associated with the first data item may be generated using a generative model. The plurality of generated data items may be classified based on the first predictive model to generate classified generated data. A second predictive model may be trained with the classified generated data. A system and computer program product are also disclosed.

Abstract: Aspects for forward propagation of a convolutional artificial neural network are described herein. The aspects may include a direct memory access unit configured to receive input data from a storage device and a master computation module configured to select one or more portions of the input data based on a predetermined convolution window. Further, the aspects may include one or more slave computation modules respectively configured to convolute a convolution kernel with one of the one or more portions of the input data to generate a slave output value. Further still, the aspects may include an interconnection unit configured to combine the one or more slave output values into one or more intermediate result vectors, wherein the master computation module is further configured to merge the one or more intermediate result vectors into a merged intermediate vector.

Abstract: Disclosed is a method, system, and computer readable medium to manage (and possibly replace) cycles in graphs for a computer device. The method includes detecting a compound operation including a first tensor, the compound operation resulting from source code represented in a first graph structure as part of a compilation process from source code to binary executable code. To address a detected cycle, an instance of a proxy class may be created to store a pointer to a proxy instance of the first tensor based on the detection. In some examples, using the instance of the proxy class facilitates implementation of a level of indirection to replace a cyclical portion of the graph structure with an acyclical portion such that the second graph structure indicates assignment of a result of the compound operation to the proxy instance of the first tensor. Optimization may reduce a total number of indirection replacements.