Abstract: Provided is an information processing apparatus including: an output control unit that controls display of information regarding an erroneous recognition target different from a predetermined recognition target, the erroneous recognition target being specified as having a possibility of erroneous recognition on the basis of a result of recognition processing on at least one piece of specifying data by a recognizer generated for recognizing the predetermined recognition target and a result of clustering processing on a target recognized by the recognition processing. The recognizer is generated by learning based on at least one piece of learning data. The at least one piece of learning data includes the predetermined recognition target and is data in substantially the same context. The specifying data is data in substantially the same context as the context of the at least one piece of learning data.

Abstract: Certain aspects of the disclosure provide systems and methods for training an information extraction transformer model architecture directed to pre-training a first multimodal transformer model on an unlabeled dataset, training a second multimodal transformer model on a first labeled dataset to perform a key information extraction task processing the unlabeled dataset with the second multimodal transformer model to generate pseudo-labels for the unlabeled dataset, training the first multimodal transformer model based on a second labeled dataset comprising one or more labels, the pseudo-labels generated, or combinations thereof to generate a third multimodal transformer model, generating updated pseudo-labels based on label completion predictions from the third multimodal transformer model, and training the third multimodal transformer model using a noise-aware loss function and the updated pseudo-labels to generate an updated third multimodal transformer model.

Abstract: A method and system for characterization of Cannabaceae plants using macro photography images is disclosed. The method comprises the steps of receiving one or more macro photography images of a Cannabaceae plant; performing feature extraction analysis of trichomes using image processing, and performing plant characterization analysis using a neural network which analyzes the macro photography images. The training phase of the neural network comprises using results of chemical composition laboratory tests performed on the plants for which the macro photography images have been used in the training phase. The invention calculates and reports an assessment of maturity of the plant for harvesting, diagnosis of the existence of diseases, insects, or pests, assessment of the presence and concentrations of central ingredients, recommendations for treatment during plants drying, curing or storage production processes, and assessment of the quality and pricing of Cannabaceae plants products.

Abstract: The embodiments of the disclosure provide a video description information generation method, a video processing method, and video description information generation apparatus, and a video processing apparatus. The video description information generation method includes: obtaining a frame-level video feature sequence corresponding to a video; generating a global part-of-speech sequence feature of the video according to the video feature sequence, the global part-of-speech sequence feature being a feature of a sequence of a combination of parts of speech in the video; and generating natural language description information of the video according to the global part-of-speech sequence feature and the video feature sequence.

Abstract: Aspects of the technology described herein relate to techniques for guiding an operator to use an ultrasound device. Thereby, operators with little or no experience operating ultrasound devices may capture medically relevant ultrasound images and/or interpret the contents of the obtained ultrasound images. For example, some of the techniques disclosed herein may be used to identify a particular anatomical view of a subject to image with an ultrasound device, guide an operator of the ultrasound device to capture an ultrasound image of the subject that contains the particular anatomical view, and/or analyze the captured ultrasound image to identify medical information about the subject.

Abstract: Accurately detection of logos in media content on media presentation devices is addressed. Logos and products are detected in media content produced in retail deployments using a camera. Logo recognition uses saliency analysis, segmentation techniques, and stroke analysis to segment likely logo regions. Logo recognition may suitably employ feature extraction, signature representation, and logo matching. These three approaches make use of neural network based classification and optical character recognition (OCR). One method for OCR recognizes individual characters then performs string matching. Another OCR method uses segment level character recognition with N-gram matching. Synthetic image generation for training of a neural net classifier and utilizing transfer learning features of neural networks are employed to support fast addition of new logos for recognition.

Abstract: An analysis device includes an analysis unit configured to receive scattered light, transmitted light, fluorescence, or electromagnetic waves from an observed object located in a light irradiation region light-irradiated from a light source and analyze the observed object on the basis of a signal extracted on the basis of a time axis of an electrical signal output from a light-receiving unit configured to convert the received light or electromagnetic waves into the electrical signal.

Abstract: A style-based generative network architecture enables scale-specific control of synthesized output data, such as images. During training, the style-based generative neural network (generator neural network) includes a mapping network and a synthesis network. During prediction, the mapping network may be omitted, replicated, or evaluated several times. The synthesis network may be used to generate highly varied, high-quality output data with a wide variety of attributes. For example, when used to generate images of people's faces, the attributes that may vary are age, ethnicity, camera viewpoint, pose, face shape, eyeglasses, colors (eyes, hair, etc.), hair style, lighting, background, etc. Depending on the task, generated output data may include images, audio, video, three-dimensional (3D) objects, text, etc.

Abstract: Methods, apparatus, and articles of manufacture providing an efficient safety mechanism for signal processing hardware are disclosed. An example apparatus includes an input interface to receive an input signal; a hardware accelerator to process the input signal, the hardware accelerator including: unprotected memory to store non-critical data corresponding to the input signal; and protected memory to store critical data corresponding to the input signal; and an output interface to transmit the processed input signal.

Abstract: Systems and methods are disclosed for tracking objects in a physical environment using visual sensors onboard an autonomous unmanned aerial vehicle (UAV). In certain embodiments, images of the physical environment captured by the onboard visual sensors are processed to extract semantic information about detected objects. Processing of the captured images may involve applying machine learning techniques such as a deep convolutional neural network to extract semantic cues regarding objects detected in the images. The object tracking can be utilized, for example, to facilitate autonomous navigation by the UAV or to generate and display augmentative information regarding tracked objects to users.

Abstract: According to one embodiment, a reading support system includes a processing device. The processing device includes an extractor and a type determiner. The extractor extracts a plurality of regions from a candidate region. The candidate region is a candidate of a region in which a meter is imaged. The regions respectively include a plurality of characters of the meter. The type determiner determines a type of the meter based on positions of the regions.

Abstract: A target custody platform comprising a data acquisition engine, a data analysis engine, a machine learning engine, and a data presentation layer configured to task a plurality of satellites for imagery data wherein the imagery data and metadata is used in conjunction with other types of data including identification data and weather data as inputs into a one or more machine and/or deep learning algorithms configured to predict a the likelihood a target of interest will travel along a project path.

Abstract: A system for detecting and tracking an object that is exhibiting an anomalous behavior from a helicopter is disclosed. The system includes a search light connected to the helicopter; a camera; and a processor including an object detection module coupled to the search light and the camera, the object detection module being configured to receive a plurality of images from the camera, compare the plurality of images against a pattern database, determine the object is exhibiting the anomalous behavior and instruct the search light to point toward the object.

Abstract: Autonomous aerial navigation in low-light and no-light conditions includes using night mode obstacle avoidance intelligence, training, and mechanisms for vision-based unmanned aerial vehicle (UAV) navigation to enable autonomous flight operations of a UAV in low-light and no-light environments using infrared data.

Abstract: The present disclosure provides a portable apparatus for detecting early crop diseases based on spatial frequency domain imaging. The apparatus includes an end cover, a spatial frequency domain imaging apparatus, a dark box body, a telescopic section, and an opening-and-closing apparatus connected in sequence. The detection method includes: putting a crop sample to be detected into the dark box body from a bottom; projecting structured light of sine grey scale patterns with different spatial frequencies to the crop sample; after the sine gray scale pattern is switched each time, acquiring, by a camera, a diffuse reflection image of a surface of the crop sample once; after capturing all diffuse reflection images, performing uniformity correction on the images, demodulating the images, and extracting an alternating current component; and inputting an alternating current component image to a trained disease detection model, and determining whether the crop sample has a disease.

Abstract: Augmented reality (AR) servicing guidance uses a mobile device (30, 90) with a display and a camera. Computer vision (CV) processing (102) such as Simultaneous Location and Mapping (SLAM) is performed to align AR content (80) with a live video feed (96) captured by the camera. The aligned AR content is displayed on the display of the mobile device. In one example, the mobile device is a head-up display (HUD) (90) with a transparent display (92, 94) and camera (20). In another example the mobile device is a cellphone or tablet computer (30) with a front display (34) and a rear-facing camera (36). The AR content (80) is generated by CV processing (54) to align (70) AR content with recorded video (40) of a service call. The aligning (72) includes aligning locations of interest (LOIs) identified in the recorded video of the service call by a user input.

Abstract: Systems and methods for displaying augmented reality content relative to an identified object are disclosed. The systems and methods can include a set of operations. An operation can include providing a first image. An operation can include determining interest points in the first image. An operation can include identifying an object in the first image. An operation can include identifying augmented reality content associated with the object. An operation can include determining a first transformation for displaying the augmented reality content in the first image relative to the identified object. An operation can include providing the interest points and the first transformation. An operation can include determining a second transformation for displaying the augmented reality content in a second image relative to the identified object using, at least in part, the first transformation and the interest points. An operation can include displaying, by a user device, the augmented reality content.

Abstract: Images of an object may be captured via a camera at a mobile computing device at different viewpoints. The images may be used to identify components of the object and to identify damage estimates estimating damage to some or all of the components. Capture coverage levels corresponding with the components may be determined, and then recording guidance may be provided for capturing additional images to increase the capture coverage levels.

Abstract: This disclosure is enables various technologies involving various actions based on an object detecting a defined area and the defined area detecting the object.

Abstract: An example virtual reality (VR) system includes a VR headset, a plurality of monitoring stations, and an image processing device. Each of the monitoring stations includes an image sensor to capture images of the VR headset and an environment in which the VR headset is used. The image processing device is to process the images captured by the monitoring stations to detect an object in the environment, and to transmit, to the VR headset, information that includes notification of the object detected.

Abstract: Methods and apparatus to measure brand exposure in media streams are disclosed. Example apparatus disclosed herein are to determine a first histogram based on at least one of luminance components or chrominance components of a first frame of video, and determine a second histogram based on at least one of luminance components or chrominance components of a second frame of the video. Disclosed example apparatus are also to detect a transition in the video based on the first histogram and the second histogram, and responsive to the detection of the transition in the video. Disclosed example apparatus are further to process a region of interest within at least one of the first frame or the second frame to detect a logo in the region of interest.

Abstract: A system and method are presented to create custom versions for users of recorded sessions of individuals. Individuals are recorded at a booth responding to prompts. Audio and visual data recorded at the booth are divided into time segments according to the timing of the prompts. Depth sensors at the booth are used to assign score values to time segments. Prompts are related to criteria that were selected as being relevant to an objective. Users are associated with subsets of criteria in order to identify subsets of prompts whose responses are relevant to the users. Time segments of audio and visual data created by the identified subset of prompts are selected. The selected time segments are ordered according to herd behavior analysis. Lesser weighted time segments may be redacted. The remaining portions of ordered time segments are presented to the user as a custom version.

Abstract: A location in a video may be specified and content related to the location may be accessed. A method for accessing the related content may include receiving a reference to a pixel location in a frame of a video feed of a filmed occurrence and accessing a spatio-temporal index corresponding to the filmed occurrence. The spatio-temporal index may index information relating to events or objects of the filmed occurrence and corresponding pixel locations at which the events or the objects are detected in the video feed. The method may further include querying the spatio-temporal index using the pixel location to determine particular information of an indexed event or an indexed object and receiving the particular information wherein the particular information indicates at least one of spatial and temporal alignment parameters for aligning the indexed event with a corresponding event in at least one other video feed of the filmed occurrence.

Abstract: Data processing systems and methods are disclosed for augmenting video content with one or more augmentations to produce augmented video. Elements within video content may be identified by spatiotemporal indices and may have associated values. An advertiser can pay to have an augmentation added to an element that, for example, advertises the advertiser's goods and/or includes a link that, when activated, takes a user to the advertiser's web site. Elements may have associated contexts that can be used to determine augmentations and element value, such as a position and/or current use of the element.

Abstract: Methods, systems and apparatuses may provide for technology that selects a player from a plurality of players based on an automated analysis of two-dimensional (2D) video data associated with a plurality of cameras, wherein the selected player is nearest to a projectile depicted in the 2D video data. The technology may also track a location of the selected player over a subsequent plurality of frames in the 2D video data and estimate a location of the projectile based on the location of the selected player over the subsequent plurality of frames.

Abstract: Methods, systems, and media for adaptive presentation of a video content item based on an area of interest are provided.

Abstract: Aspects of the present invention provide an approach for reinforcement learning for unsupervised video summarization with piecewise linear interpolation. A candidate importance score is anticipated by training a video summarization network using a set of video frames in a video. The candidate importance score is interpolated with respect to other candidate importance scores in the video in order to generate a selection probably for each frame of the video. A video summarization that includes a set of selected frames is generated frame selection probability.

Abstract: The present disclosure provides a method for acquiring an item placing state, including: sending an acquisition request, wherein the acquisition request includes one or more association relationships, each of the association relationships includes an association relationship between a hot zone and a camera equipment, and one hot zone corresponds to one kind of item; and receiving a placing state of an item corresponding to at least one hot zone, wherein the placing state of the item corresponding to the at least one hot zone is determined according to a current image and a standard image of the at least one hot zone in response to the acquisition request. The present disclosure further provides a processing method, a system for acquiring an item placing state and a processing system.

Abstract: A video analytics system for real-time monitoring and assessment of airplane ramp safety processes comprises rules relating to airplane ramp safety processes, portable video capture devices located at different predetermined positions with respect to an airplane at an airport ramp which capture video and data related to movements and positioning of objects and services being provided relative to the aircraft at the airport ramp and transmit captured video footage and data via a wireless network to a server which performs video analytics on the captured video and data to generate video analytics data, compare the video analytics data to the rules relating to the airplane safety processes, to determine any safety conditions, store the results of the comparison and output a real-time notification if a safety condition is determined.

Abstract: The present disclosure provide a method and an Internet of Things system for counting and regulating pedestrian volume in a public place of a smart city. The method includes receiving, based on the user platform, a query request for an intended place initiated by a user; transmitting, based on the service platform, the query request to the management platform, and generating, based on the management platform, a query instruction; issuing, based on the management platform, the query instruction to a corresponding sensor network sub-platform according to the regional location; sending, based on the sensor network sub-platform, the query instruction to the corresponding object platform; obtaining, based on the object platform, a query result according to the query instruction; and feeding back, based on the object platform, the query result to the user platform through the corresponding sensor network sub-platform, the management platform, and the service platform respectively.

Abstract: A map in a cloud service stores physical objects previously detected by other vehicles that have previously traveled over the same road that a current vehicle is presently traveling on. New data received by the cloud service from the current vehicle regarding new objects that are being encountered by the current vehicle can be compared to the previous object data stored in the map. Based on this comparison, an operating status of the current vehicle is determined. In response to determining the status, an action such as terminating an autonomous navigation mode of the current vehicle is performed.

Abstract: An object recognition method uses a sensor configured to acquire a position of an object existing in a surrounding environment as point clouds including a plurality of detection points in a top view. The method includes: grouping point clouds according to a proximity; determining, when performing polygon approximation on the grouped point clouds, whether at least part of the detection points constituting the grouped point clouds are located in a blind zone of an approximate polygon acquired by the polygon approximation with respect to the sensor; recognizing the grouped point clouds as point clouds corresponding to plural objects when it is determined that the detection points are located in the blind zone with respect to the sensor; and recognizing the grouped point clouds as point clouds corresponding to a single object of the approximate polygon when it is determined that the detection points are not located in the blind zone.

Abstract: The technology provides a sign detection and classification methodology. A unified pipeline approach incorporates generic sign detection with a robust parallel classification strategy. Sensor information such as camera imagery and lidar depth, intensity and height (elevation) information are applied to a sign detector module. This enables the system to detect the presence of a sign in a vehicle's externa environment. A modular classification approach is applied to the detected sign. This includes selective application of one or more trained machine learning classifiers, as well as a text and symbol detector. Annotations help to tie the classification information together and to address any conflicts with different the outputs from different classifiers. Identification of where the sign is in the vehicle's surrounding environment can provide contextual details.

Abstract: A method for monitoring driver alertness in a vehicle is disclosed. Image data representing images of an eye of a driver are obtained, the images having been captured while the driver is operating the vehicle. An eye openness measure of the eye of the driver is computed, based on the image data. A determination is made that the driver is in an unsafe state for driving based on the eye openness measure, and an alert is generated, while the driver is operating the vehicle, based on determining that the driver is in the unsafe state.

Abstract: Method and system for detecting vehicle occupant actions are disclosed. For example, the method includes receiving, by one or more processors, previously classified image data from a previously classified image database, the previously classified image data representing driver postures that are rotated and scaled to be standardized for a range of different drivers and different locations of a vehicle interior sensor within a given vehicle, wherein the previously classified image data is representative of previously classified vehicle occupant actions; receiving, by the one or more processors, current image data from the vehicle interior sensor subsequent to the vehicle interior sensor being registered within the given vehicle, wherein the current image data is representative of current vehicle occupant actions; and determining, by the one or more processors, a vehicle occupant action based at least in part upon a comparison of the current image data and the previously classified image data.

Abstract: A method comprises receiving an input of at least one image associated with an issue. In the method, text is extracted from the at least one image, and an intent is determined from the extracted text. The method further includes recommending a response to the issue based at least in part on the determined intent, and transmitting the recommended response to a user. The extracting, determining and recommending are performed at least in part using one or more machine learning models.

Abstract: A text recognition method includes: acquiring an image including text information, the text information including M characters, M being a positive integer greater than 1; performing text recognition on the image to acquire character information about the M characters; recognizing reading direction information about each character in accordance with the character information about the M characters, the reading direction information being used to indicate a next character corresponding to a current character in a semantic reading order; and ranking the M characters in accordance with the reading direction information about the M characters to acquire a text recognition result of the text information.

Abstract: A system for monitoring gluten consumption, especially in celiac people, which allows the feeding of food consumption data and updating, in real time, of the estimated amount of gluten consumed daily. Still, the present invention refers to a system for prediction that associates the possibility of an indisposition being associated or not with an undue consumption of gluten.

Abstract: An inverted microscope system provided with a transillumination subsystem that illuminates a sample includes: an eyepiece lens; an objective; a tube lens; a projection device, disposed below the objective, that projects a projected image based on projected image data onto an image plane where an optical image is formed; a first modulation element included in the transillumination subsystem; a second modulation element disposed between the objective and the tube lens; and a processor and a memory, the processor being configured to perform the following steps: generating an analysis result that specifies a candidate cell that is a reproductive cell suitable for fertilization, based on at least the digital image data acquired by the imaging device; and generating the projected image data based on the generated analysis result, wherein the projected image includes a first assisting image that specifies the candidate cell as an assisting image that assists with micro-insemination.

Abstract: A computing device for extracting target data from a source document includes: a memory storing target data extraction rules; a processor connected with the memory, the processor configured to: obtain text recognition data extracted from an image of the source document, the text recognition data indicating locations of text structures in the source document; define text lines based on the text recognition data; identify a reference string from the text recognition data; select a subset of the text lines based on a location of the reference string and the target data extraction rules; and output the subset of the text lines as the target data.

Abstract: The present disclosure describes a method, an apparatus, and a non-transitory computer-readable medium for detecting sensitive text information such as privacy-related text information from a signal and modifying the signal by removing the detected sensitive text information therefrom. The apparatus receives the signal such as an image, a video clip, or an audio clip, and recognizes a text string therefrom. The apparatus then detects, from the text string, a substring based on a similarity between the substring and a regular expression, and modifies the signal by removing information related to the detected substring from the signal.

Abstract: Systems and methods here may be used for pre-processing images, including using a computer for receiving a pixelated image of a paper document of an original size, downscaling the received pixelated image, employing a neural network algorithm to the downscaled image to identify four corners of the paper document in the received pixelated image, re-enlarging the downscaled image to the original size, identifying each of four corners of the paper document in the pixelated image, determining a quadrilateral composed of lines that intersect at four angles at the four corners of the paper document in the pixelated image, defining a projective plane of the pixelated image, and determining an inverse transformation of the pixelated image to transform the projective plane quadrilateral into a right angled rectangle.

Abstract: Systems and methods are disclosed to receive a training data set comprising a plurality of document images, wherein each document image of the plurality of document images is associated with respective metadata identifying a document field containing a variable text; generate, by processing the plurality of document images, a first heat map represented by a data structure comprising a plurality of heat map elements corresponding to a plurality of document image pixels, wherein each heat map element stores a counter of a number of document images in which the document field contains a document image pixel associated with the heat map element; receive an input document image; and identify, within the input document image, a candidate region comprising the document field, wherein the candidate region comprises a plurality of input document image pixels corresponding to heat map elements satisfying a threshold condition.

Abstract: A receipt capture tool residing on a customer mobile device may be initiated when a customer completes an in-store or online purchase. The receipt capture tool may prompt the customer to capture an image of a receipt detailing a purchase and an item (e.g., product or service) purchased. For instance, the photo of a physical receipt may be taken by the mobile device, or an electronic receipt or email detailing the purchasing transmitted from a physical merchant or online merchant server may be stored. Receipt information may be extracted and saved with other information pertinent to the item purchased, including warranty information. If the customer needs to return or repair the item purchased at a future date, the receipt and warranty information may be subsequently accessed via their mobile device. The receipt and warranty information may also be stored in a searchable database to facilitate easy retrieval by the customer.

Abstract: Actors may be detected and tracked within a scene using multiple imaging devices provided in a network that are aligned with fields of view that overlap at least in part. Processors operating on the imaging devices may evaluate the images using one or more classifiers to recognize body parts within the images, and to associate the body parts with a common actor within the scene. Each of the imaging devices may generate records of the positions of the body parts and provide such records to a central server, that may correlate body parts appearing within images captured by two or more of the imaging devices and generate a three-dimensional model of an actor based on positions of the body parts. Motion of the body parts may be tracked in subsequent images, and the model of the actor may be updated based on the motion.

Abstract: Optical sensors and their making methods are described herein. In some embodiments, a described sensing apparatus includes: an image sensor; a collimator above the image sensor, wherein the collimator includes an array of apertures; and an optical filtering layer above the collimator, wherein the optical filtering layer is configured to filter a portion of light to be transmitted into the array of apertures.

Abstract: A display apparatus and a bonding method therefor are provided. The display apparatus includes a display module, the back face of the display module having a heat dissipation film layer, a first open hole being disposed in the heat dissipation film layer, a composite film layer being provided in the first open hole, and a fingerprint module being attached to the composite film layer; the composite film layer includes a support layer and a buffer layer attached to one another, the support layer being closest to the fingerprint module.

Abstract: The present disclosure relates to a display panel and a display device using the same. The display panel includes a plurality of first unit pixel areas in which N subpixels (N is a positive integer greater than or equal to 3) for displaying data are disposed, and a plurality of second unit pixel areas in which N?1 subpixels for displaying data are disposed. Each of the first unit pixel areas and each of the second unit pixel areas have the same size. Each of the first and second unit pixel areas is a pixel area in a parallelogram or rhombus shape.

Abstract: A piezoelectric sensor, a manufacturing method thereof and an electronic device are provided. The piezoelectric sensor includes a substrate, an active layer, the active layer being disposed at a side of the substrate: a first electrode, the first electrode being disposed at a side of the active laver a wav from the substrate, and the first electrode including a plurality of sub-electrodes disposed at intervals: a piezoelectric layer, the piezoelectric layer being disposed at a side of the first electrode away from the active layer; and a second electrode, the second electrode being disposed at a side of the piezoelectric layer away from the first electrode. The active layer is configured to be capable of switching between an insulating state and a conducting state, and in the conducting state the active layer is capable of conducting the plurality of sub-electrodes.

Abstract: A detection device includes a sensor area provided with a plurality of detection elements each comprising a photoelectric conversion element in a detection area, and a detector configured to detect a potential difference between a voltage generated in a first detection element and a voltage generated in a second detection element adjacent to the first detection element in the detection area.