Abstract: A video quality evaluation method comprises acquiring an image sequence and audio information by decoding a to-be-evaluated video, wherein the to-be-evaluated video is non-offending video; extracting an action feature vector and a face feature vector from the image sequence, and extracting an audio feature vector from the audio information; constructing a video feature vector according to at least one of the action feature vector, the face feature vector and the audio feature vector; and determining a quality score of the to-be-evaluated video according to the video feature vector.

Abstract: A mechanism is described for facilitating hemisphere cube map projection format imaging environments, according to one embodiment. A method of embodiments, as described herein, includes capturing, by a camera coupled to one or more processors, an image having image content, wherein the image content being represented by the image is omnidirectional such that the image content is mapped on a sphere while representing less than the sphere; mapping the image to a cubic representation based on six faces of a cube, wherein one or more of the six faces are classified as inactive regions such that they remain unoccupied or partially occupied; and arranging, based on the cubic representation, the six faces in a compact representation by avoiding inclusion of the inactive regions.

Abstract: A method and system for providing location-specific information to users, wherein this information is directly maintained and supplied by the provider of the service. This information is supplied by an application running on a personal communication device, where the application is developed and provided by a third-party servicer and the third party servicer makes various program options/APIs of the application available to and selectable by its customers. Other embodiments of the invention provide techniques for changing the appearance and/or functionality of a location aware application on a portable communication device. The method includes capturing or retrieving a digital image of a subject using a portable communications device; analyzing the digital image to determine a subject of the digital image; detecting a location associated with the digital image; and displaying information related to the subject of the digital image and the location associated with the digital image.

Abstract: An apparatus and method for efficient image optimized image stitching. For example, one embodiment of an apparatus comprises: feature search area identification circuitry/logic to narrow down a feature search area based on possible overlap between two image frames; feature detection circuitry/logic to identify a plurality of feature points in a first image frame of the two image frames; feature matching circuitry/logic to map one or more of the plurality of feature points from the first image frame to corresponding feature points in the right image frame; image frame stitching and blending circuitry/logic to stitch the first image frame and second image frame based on the mapping of the feature points between the two image frames and to blend a portion of the first image frame with a portion of the second image frame.

Abstract: A client device operates by: extracting a set of deletion candidate data from application data stored in memory, wherein the deletion candidate data corresponds to a plurality of photographs; generating feature detection data for each of the plurality of photographs by performing a computer vision function on the each of the plurality of photographs; generating a subset of the set of deletion candidate data for deletion by selecting ones of the set of deletion candidate data that compare favorably to deletion criteria data, wherein the deletion criteria data indicates at least one unfavorable feature and wherein the subset of the set of deletion candidate data is generated by selecting ones of the plurality of photographs with feature detection data that indicates the at least one unfavorable feature is included in the ones of the plurality of photographs; and facilitating deletion of at least a portion of the subset of the set of deletion candidate data from the memory.

Abstract: According to one embodiment, a ranging device includes storage and a processor. The storage is configured to store a statistical model generated by learning a bokeh that occurs in a first image affected by aberration of an optical system of a capture unit. The processor is configured to acquire a second image captured by the capture unit, acquire focus position information when the second image was captured, acquire a bokeh value indicating a bokeh of a subject in the second image, which is output from the statistical model by inputting the acquired second image to the statistical model, and convert the acquired bokeh value into a distance to the subject based on the acquired focus position information.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for identifying the language of a spoken utterance. One of the methods includes receiving input features of an utterance; and processing the input features using an acoustic model that comprises one or more convolutional neural network (CNN) layers, one or more long short-term memory network (LSTM) layers, and one or more fully connected neural network layers to generate a transcription for the utterance.

Abstract: An medical image processing method comprises generating a first segmented medical image in accordance with a first segmentation model and based on an original medical image that comprises a plurality of pixels. The method also comprises determining a foreground point and a background point according to the initial target region of the first segmented image. The method further comprises: for each pixel of the plurality of pixels of the original image, determining a first image distance between the respective pixel and the foreground point and a second image distance between the respective pixel and the background point. The method further comprises obtaining a foreground point range image and a background point range image corresponding to the original image, and generating a second segmented image in accordance with a second segmentation model based on the original image, the foreground point range image, and the background point range image.

Abstract: A process mining system performs process mining using visual logs generated from video streams of worker devices. Specifically, for a given worker device, the process mining system obtains a series of images capturing a screen of a worker device while the worker device processes one or more tasks related to an operation process. The process mining system determines activity labels for a plurality of images. An activity label for an image may indicate an activity performed on the worker device when the image was captured. The activity label is determined by extracting information from pixels of the image and inferring the activity of the worker device from the extracted information. The process mining system generates event logs from the visual logs of worker devices and uses the event logs for process mining.

Abstract: Provided is a neural-network-based classification method, including: generating, by a neural network, one or more score vectors corresponding to one or more samples respectively; determining a first subset of the one or more samples according to the one or more score vectors and a first decision threshold, wherein the first subset is associated with a first class; and selecting samples to be re-examined from the one or more samples according to the first subset.

Abstract: A computing device can obtain data describing at least one document, the at least one document referencing at least one media object, wherein a portion of the at least one media object includes one or more characters. The computing device can obtain data describing the one or more characters in the at least one media object in the at least one document. The computing device can generate an updated copy of the at least one document that includes the data describing the one or more characters in the at least one media object. The computing device can present, on a display screen of the computing device and through an interface, the updated copy of the at least one document, wherein the one or more characters in the at least one media object are able to be selected or searched.

Abstract: An image processing apparatus receives destination information for use in data transmission, performs control, based on the received destination information including a destination in an email address format, so that a first screen, which is used to transmit data external to the image processing apparatus, and on which a transmission destination of the data is displayed, based on the received destination information, is displayed on the operation unit, and performs control, based on the received destination information including only a destination in a fax format so that a second screen, different from the first screen and used to perform fax transmission, on which a transmission destination of the fax transmission is displayed, based on the received destination information, is displayed on the operation unit.

Abstract: A system and method for measuring three-dimensional (3D) coordinate values of an environment is provided. The method including moving a 2D scanner through the environment. A 2D map of the environment is generated using the 2D scanner. A path is defined through the environment using the 2D scanner. 3D scan locations along the path are defined using the 2D scanner. The 2D scanner is operably coupled to a mobile base unit. The mobile base unit is moved along the path based at least in part on the 2D map and the defined path. 3D coordinate values are measured at the 3D scan locations with a 3D scanner, the 3D scanner being coupled to the mobile base unit.

Abstract: A system is configured to receive an indication that an apparatus in a first assembled state should comprise a component with a first digital fingerprint and a component with a second digital fingerprint. The system is configured to receive video footage of apparatuses in the first assembled state. The system is configured to isolate an image of an apparatus in the first assembled state. The system is configured to split the image into a frame comprising a first component and a frame comprising a second component. The system is configured to generate first and second filtered images. The system is configured to identify feature points in the filtered images. The system is configured to determine that the first set of feature points matches the first digital fingerprint and that the second set of feature points matches the second digital fingerprint. The system is configured to update a component database.

Abstract: There is a need for more effective and efficient predictive data analysis solutions and/or more effective and efficient solutions for generating image representations of categorical data. In one example, embodiments comprise receiving a categorical input feature, generating an image representation of the categorical input feature, generating an image-based prediction based at least in part on the image representation, and performing one or more prediction-based actions based at least in part on the image-based prediction.

Abstract: Example aspects of the present disclosure are directed to systems and methods for learning data augmentation strategies for improved object detection model performance. In particular, example aspects of the present disclosure are directed to iterative reinforcement learning approaches in which, at each of a plurality of iterations, a controller model selects a series of one or more augmentation operations to be applied to training images to generate augmented images. For example, the controller model can select the augmentation operations from a defined search space of available operations which can, for example, include operations that augment the training image without modification of the locations of a target object and corresponding bounding shape within the image and/or operations that do modify the locations of the target object and bounding shape within the training image.

Abstract: Aspects of the subject disclosure may include, for example, a device, that includes a processing system including a processor and a memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations including receiving a manifest for a point cloud, wherein the point cloud is partitioned into a plurality of cells; determining an occlusion level for a cell of the plurality of cells with respect to a predicted viewport; reducing a point density for the cell provided in the manifest based on the occlusion level, thereby determining a reduced point density; and requesting delivery of points in the cell, based on the reduced point density. Other embodiments are disclosed.

Abstract: The present disclosure describes systems, non-transitory computer-readable media, and methods for utilizing a machine learning model trained to determine subtle pose differentiations to analyze a repository of captured digital images of a particular user to automatically capture digital images portraying the user. For example, the disclosed systems can utilize a convolutional neural network to determine a pose/facial expression similarity metric between a sample digital image from a camera viewfinder stream of a client device and one or more previously captured digital images portraying the user. The disclosed systems can determine that the similarity metric satisfies a similarity threshold, and automatically capture a digital image utilizing a camera device of the client device. Thus, the disclosed systems can automatically and efficiently capture digital images, such as selfies, that accurately match previous digital images portraying a variety of unique facial expressions specific to individual users.

Abstract: Techniques relating to managing “bad” or “imperfect” data being imported into a database system are described herein. A lifecycle technology solution helps receive data from a variety of different data sources of a variety of known and/or unknown formats, standardize it, fit it to a known taxonomy through model-assisted classification, store it to a database in a manner that is consistent with the taxonomy, and allow it to be queried for a variety of different usages. Auto-classification, enrichment, clustering model and model stacks, and/or other disclosed techniques, may be used in these and/or other regards.

Abstract: Systems, methods, and non-transitory computer readable media are configured to apply a spectral clustering technique to at least a portion of a similarity graph to generate clusters of geographic sub-regions constituting geographic regions. A tf-idf technique is performed to determine pages of a social networking system associated with a geographic region as potential local suggestions for a user associated with a geographic sub-region in the geographic region. References to at least a portion of the pages are presented as local suggestions to the user.

Abstract: A computer-implemented method of detecting logos in a graphical rendering may comprise detecting, using a first and a second trained object detector, logos in the graphical rendering and outputting a first and a second list of detections and filtering, using at least a first and a second prior performance-based filter, the received first and second lists of detections into a first group of kept detections, a second group of discarded detections and a third group of detections. Detections in the third group of detections may be clustered in at least one cluster comprising detections that are of a same class and that are generally co-located within the electronic image. A cluster score may then be assigned to each cluster. A set of detections of logos in the graphical rendering may then be output, the set comprising the detections in the first group and a detection from each of the clusters whose assigned cluster score is greater than a respective threshold.

Abstract: A plant operation data monitoring device comprises: an input section that receives operation data on a plant; and a calculator that includes databases storing the operation data received, and a computing section executing a program. The computing section stores the operation data received in a first database of the databases in time series. The computing section determines from peak values of the operation data stored whether gradients of the operation data are positive or negative, and then stores the gradients in a second database of the databases for positive gradients or in the second database of the databases for negative gradients in time series. The computing section determines threshold values for abnormality determination about the positive and negative gradients, divides the positive gradients and the negative gradients into normal values and abnormal values, and additionally stores the divided gradients in the second database for the positive or negative gradients.

Abstract: Methods, apparatus, and processor-readable storage media for determining similarity between time series using machine learning techniques are provided herein. An example computer-implemented method includes obtaining a primary time series and a set of multiple candidate time series; calculating, using machine learning techniques, similarity measurements between the primary time series and each of the candidate time series; for each of the similarity measurements, assigning weights to the candidate time series based on similarity to the primary time series relative to the other candidate time series; generating, for each of the candidate time series, a similarity score based on the weights assigned to each of the candidate time series across the similarity measurements; and outputting, based on the similarity scores, identification of at least one candidate time series for use in one or more automated actions relating to at least one system.

Abstract: A computing system, a method, and a computer-readable storage medium for adjusting eye gaze are described. The method includes capturing a video stream including images of a user, detecting the user's face region within the images, and detecting the user's facial feature regions within the images based on the detected face region. The method includes determining whether the user is completely disengaged from the computing system and, if the user is not completely disengaged, detecting the user's eye region within the images based on the detected facial feature regions. The method also includes computing the user's desired eye gaze direction based on the detected eye region, generating gaze-adjusted images based on the desired eye gaze direction, wherein the gaze-adjusted images include a saccadic eye movement, a micro-saccadic eye movement, and/or a vergence eye movement, and replacing the images within the video stream with the gaze-adjusted images.

Abstract: Described are various embodiments of a pupil tracking system and method, and digital display device and digital image rendering system and method using same. In one embodiment, a computer-implemented method for improving a perceptive experience of light field content projected via a light field display within a light field viewing zone comprises sequentially acquiring a user feature location, and comparing a velocity computed therefrom with a designated threshold velocity. Upon the velocity corresponding with a transition from a relatively dynamic to a relatively static state, a rendering geometry of the light field content is adjusted to project the light field content within an adjusted light field viewing zone in accordance with a newly acquired user feature location.

Abstract: The technology disclosed extends Human-in-the-loop (HITL) active learning to incorporate real-time human feedback to influence future sampling priority for choosing the best instances to annotate for accelerated convergence to model optima. The technology disclosed enables the user to communicate with the model that generates machine annotations for unannotated instances. The technology disclosed also enables the user to communicate with the sampling logic that selects instances to be annotated next. The technology disclosed enables the user to generate ground truth annotations, from scratch or by correcting erroneous model annotations, which guide future model predictions to more accurate results.

Abstract: A capturing device has a lens block that includes a lens for focusing light from a subject during a daytime, the subject includes a vehicle. The capturing device further includes an image sensor that captures an image based on light from the subject focused by the lens, and a processor that generates a face image of an occupant riding in the vehicle based on a first number of captured imaged of the subject which are captured by the image sensor at different times. The processor generates the face image of the occupant further based on a second number of captured images in which a luminance value of a region of interest is equal to or smaller than a threshold value among the first number of captured images of the subject.

Abstract: A method, computer program product, and computer system for detection and utilization of similarities among tables in multiple data systems that include a first data system and a second data system. A semantic dataset is generated. A first measure (X) of similarity between semantic data in columns of the first and second data systems is computed using the semantic dataset. A first, different measure (Y) of similarity between semantic data in columns of the first and second data systems is computed using the semantic dataset. A third measure (Z) similarity between columns of the first and second data systems is computed based on data in cells within the columns. A weighted combination (U) of X, Y, and Z between the columns of tables in the first and second data systems is computed. X, Y, Z, U or combinations thereof are used to improve a computer system.

Abstract: A control apparatus includes an acquisition unit configured to acquire information on a rotation of the first imaging unit in the rotation direction and position information of a designated range designated on a first image imaged by the first imaging unit, a setting unit configured to set the second imaging range so as to include the designated range based on the information on the rotation and the position information, and a control unit configured to control the second imaging unit to instruct the second imaging unit to image the second imaging range and to acquire a second image by changing at least one of the imaging direction and the angle of view of the second imaging unit.

Abstract: An example system includes a processor to receive raw and unlabeled videos. The processor is to extract speech from the raw and unlabeled videos. The processor is to extract positive frames and negative frames from the raw and unlabeled videos based on the extracted speech for each object to be detected. The processor is to extract region proposals from the positive frames and negative frames. The processor is to extract features based on the extracted region proposals. The processor is to cluster the region proposals and assign a potential score to each cluster. The processor is to train a binary object detector to detect objects based on positive samples randomly selected based on the potential score.

Abstract: A method includes processing, using at least one processor of an electronic device, each of multiple images using a photometric augmentation engine, where the photometric augmentation engine performs one or more photometric augmentation operations. The method also includes applying, using the at least one processor, multiple layers of a convolutional neural network to each of the images, where each layer generates a corresponding feature map. The method further includes processing, using the at least one processor, at least one of the feature maps using at least one feature augmentation engine between consecutive layers of the multiple layers, where the at least one feature augmentation engine performs one or more feature augmentation operations.

Abstract: A method for determining a class to which an input image belongs in an inference process, includes: storing a frequent feature for each class in a frequent feature database; inputting an image as the input image; extracting which class the input image belongs to; extracting a plurality of features that appear in the inference process; extracting one of the features that satisfies a predetermined condition as a representative feature; reading out the frequent feature corresponding to an extracted class from the frequent feature database; extracting one or a plurality of ground features based on the frequent feature and the representative feature; storing a concept data representing each feature in an annotation database; reading out the concept data corresponding to the one or the plurality of ground features from the annotation database; generating explanation information; and outputting the ground feature and the explanatory information together with the extracted class.

Abstract: A method for combining multiple images is disclosed herein. A target mapping matrix is determined based on a first image and a second image. The target mapping matrix is associated with a target correspondence between the first image and the second image. The first image and the second image are combined into a combined image based on the first target mapping matrix. The combined image is output by implementing the disclosed method.

Abstract: A system and method for generating an interface for providing recommendations based on contextual insights, the method including: generating at least one signature for at least one multimedia content element identified within an interaction between a plurality of users; generating at least one contextual insight based on the generated at least one signature and user interests of the plurality of users, wherein each contextual insight indicates a current user preference; searching for at least one content item that matches the at least one contextual insight; and generating an interface for providing the at least one content item within the interaction between the plurality of users.

Abstract: The invention provides a method, apparatus, device and storage medium for predicting a protein binding site. The method comprises the steps of: receiving a protein sequence to be predicted, dividing the protein sequence by using a preset sliding window and sliding step to obtain a plurality of amino acid sub-sequences, building word vectors for the protein sequence according to the amino acid sub-sequences, extracting document features from word elements, building document feature vectors for the protein sequence according to the extracted document features, extracting protein chain biological features from the amino acid sub-sequences, building biological feature vectors for the protein sequence according to the extracted biological features, classifying the amino acid sub-sequences expressed with the document feature vectors and the biological feature vectors by using a preset amino acid residue classification model to obtain amino acid residue types for the protein sequence.

Abstract: Disclosed examples include image processing methods and systems to process image data, including computing a plurality of scaled images according to input image data for a current image frame, computing feature vectors for locations of the individual scaled images, classifying the feature vectors to determine sets of detection windows, and grouping detection windows to identify objects in the current frame, where the grouping includes determining first clusters of the detection windows using non-maxima suppression grouping processing, determining positions and scores of second clusters using mean shift clustering according to the first clusters, and determining final clusters representing identified objects in the current image frame using non-maxima suppression grouping of the second clusters.

Abstract: There is provided a recognition system adaptable to a portable device or a wearable device. The recognition system senses a body heat using a thermal sensor, and performs functions such as the living body recognition, image denoising and body temperature prompting according to detected results.

Abstract: An image reading device includes: an area sensor in which color filters of three colors of R, G, and B are arranged in a Bayer array and a light receiving amount is detected by a light receiving element for each color filter; and a hardware processor that: reads a document by using the light receiving elements in a first group in the area sensor, reads the document by using the light receiving elements in a second group in the area sensor, at a region shifted by ½ pixels in a sub-scanning direction from a reading region of the light receiving elements in the first group, and interpolates R-color read data and B-color read data using G-color read data and synthesizes image data having a resolution twice the resolution of the area sensor.

Abstract: A model generation method includes updating, by at least one processor, a weight matrix of a first neural network model at least based on a first inference result obtained by inputting, to the first neural network model which discriminates between first data and second data generated by using a second neural network model, the first data, a second inference result obtained by inputting the second data to the first neural network model, and a singular value based on the weight matrix of the first neural network model. The model generation method also includes at least based on the second inference result, updating a parameter of the second neural network model.

Abstract: Various embodiments are provided for implementing an approximation nearest neighbour (ANN) search in a computing environment are provided. An approximation nearest neighbour (ANN) of a plurality of feature vectors in hyper-planes with dynamically variable subspaces by searching an inverted index may be retrieved.

Abstract: A method is for matching a set of first classes assigned to a first data set with a set of second classes assigned to a second data set. The method includes constructing, via a set of pre-processing functions, a plurality of alignment profiles such that at least one alignment profile is assigned to each of the first classes and each of the second classes. The method includes generating a comparison matrix for each group of the alignment profiles, such that each group includes at least one of the first classes and at least one of the second classes. The method includes training a first machine learning model, through supervised training, based on the generated comparison matrices and based on probabilistic labels generated by a second machine learning model.

Abstract: A dynamic graph includes a plurality of nodes and edges at a plurality of time steps; each node corresponds to a geographic location in a first area where pest infestation information is available for a subset of locations. Each edge connects two of the nodes which are geographically proximate, has a direction based on wind direction, and has a weight based on relative wind speed. Assign node features based on weather data as well as labels corresponding to pest infestation severity. Train a graph convolutional network on the dynamic graph. Based on predicted future weather conditions for a second area different than the first area, use the trained graph convolutional network to predict, via inductive learning, pest infestation severity for future times for a new set of nodes corresponding to new geographic locations in the second area for which no pest infestation information is available.

Abstract: A method for ascertaining a depth information image for an input image. The input image is processed using a convolutional neural network, which includes multiple layers that sequentially process the input image, and each converts an input feature map into an output feature map. At least one of the layers is a depth map layer, the depth information image being ascertained as a function of a depth map layer. In the depth map layer, an input feature map of the depth map layer is convoluted with multiple scaling filters to obtain respective scaling maps, the multiple scaling maps are compared pixel by pixel to generate a respective output feature map in which each pixel corresponds to a corresponding pixel from a selected one of the scaling maps.

Abstract: Disclosed are a method and an apparatus for detecting abnormal objects in a video. The method for detecting abnormal objects in a video reconstructs a restored batch by applying each input batch to which an inpainting pattern is applied to a trained auto-encoder model, and fuses a time domain reconstruction error using time domain restored frames output by extracting and restoring a time domain feature point by applying a spatial domain reconstruction error and a plurality of successive frames using a restored frame output by combining the reconstructed restoring batch to a trained LSTM auto-encoder model to estimate an area where an abnormal object is positioned.

Abstract: Disclosed systems and methods relate to remote sensing, deep learning, and object detection. Some embodiments relate to machine learning for object detection, which includes, for example, identifying a class of pixel in a target image and generating a label image based on a parameter set. Other embodiments relate to machine learning for geometry extraction, which includes, for example, determining heights of one or more regions in a target image and determining a geometric object property in a target image. Yet other embodiments relate to machine learning for alignment, which includes, for example, aligning images via direct or indirect estimation of transformation parameters.

Abstract: An object of the present invention is to provide a tool system allowing a tool to be controlled on a work object basis before the work is started. A tool system includes a portable tool and an identification unit. The tool includes a driving unit to operate with power supplied from a battery pack. The identification unit identifies, by a contactless method, a current work object, to which the tool is set in place, out of a plurality of work objects.

Abstract: A modulated segmentation system can use a modulator network to emphasize spatial prior data of an object to track the object across multiple images. The modulated segmentation system can use a segmentation network that receives spatial prior data as intermediate data that improves segmentation accuracy. The segmentation network can further receive visual guide information from a visual guide network to increase tracking accuracy via segmentation.

Abstract: There is provided with an information processing apparatus. An acquisition unit acquires a plurality of pattern discrimination results each indicating a location of a pattern that is present in an image. A selection unit selects a predetermined number of pattern discrimination results from the plurality of pattern discrimination results. A determination unit determines whether or not the selected predetermined number of pattern discrimination results are to be merged, based on a similarity of the locations indicated by the predetermined number of pattern discrimination results. A merging unit merges the predetermined number of pattern discrimination results for which it was determined by the determination unit that merging is to be performed. A control unit controls the selection unit, the determination unit, and the merging unit to repeatedly perform respective processes.

Abstract: A lip-language identification method and an apparatus thereof, an augmented reality device and a storage medium. The lip-language identification method includes: acquiring a sequence of face images for an object to be identified; performing lip-language identification based on a sequence of face images so as to determine semantic information of speech content of the object to be identified corresponding to lip actions in a face image; and outputting the semantic information.

Abstract: A refrigerator according to the present invention comprises: a storage chamber for storing articles; a camera for photographing the inner space of the storage chamber; a control part which visually recognizes a first article image captured by the camera so as to acquire article information corresponding to the first article image; a memory for storing the acquired article information so as to generate an article image history; and a display electrically connected to the control part. Further, the control part may: acquire, through the camera, a second article image in which an article is partially hidden by any other article; detect, from the second article image, a partial article image of the article partially hidden by the other article; and identify an article matching the partial article image on the basis of the article image history.