Abstract: A system and method for enriching a concept database. The method includes determining, based on at least one signature of a first multimedia data element (MMDE) and signatures of a plurality of existing concepts in the concept database, at least one first concept among the plurality of existing concepts, wherein each of the at least one first concept matches a portion of the at least one signature of the first MMDE; generating a reduced representation of the first MMDE, wherein generating the reduced representation further comprises removing the portion of the first MMDE matching the at least one first concept; comparing the reduced representation of the first MMDE to signatures representing a plurality of second MMDEs to determine a plurality of matching second MMDEs; generating, based on the reduced representation of the first MMDE and the plurality of matching second MMDEs, at least one second concept; and adding the generated at least one second concept to the concept database.

Abstract: Embodiments of the present disclosure provide a method and apparatus for generating a background-free image, a device, and a medium. The method may include: determining an image proportion of a target subject in an original image; clipping, in response to the image proportion being less than a preset threshold, a subject region image corresponding to the target subject from the original image; determining a corresponding subject region mask based on the subject region image; and synthesizing the subject region mask and the subject region image to generate a background-free image corresponding to the original image.

Abstract: A processor implemented method of processing a facial expression image, the method includes controlling a camera to capture a first facial expression image and a second facial expression image, acquiring a first expression feature of the first facial expression image, acquiring a second expression feature of the second facial expression image, generating a new expression feature dependent on differences between the acquired first expression feature and the acquired second expression feature, and adjusting a target facial expression image based on the new expression feature.

Abstract: Systems and methods for automatic information retrieval from imaged documents. Deep network architectures retrieve information from imaged documents using a neuronal visual-linguistic mechanism including a geometrically trained neuronal network. An expense management platform uses the neuronal visual-linguistic mechanism to determine geometric-semantic information of the imaged document.

Abstract: A query image is received, along with a query to initiate a search process to find other images based on the query image. The query includes a preference value associated with an attribute, the preference value indicative of a level of emphasis to be placed on the attribute during the search. A full query vector, which is within a first dimensional space and representative of the query image, is generated. The full query vector is projected to a reduced dimensional space having a dimensionality lower than the first dimensional space, to generate a query vector. An attribute direction corresponding to the attribute is identified. A plurality of candidate vectors of the reduced dimensional space is searched, based on the attribute direction, the query vector, and the preference value, to identify a target vector of the plurality of candidate vectors. A target image, representative of the target vector, is displayed.

Abstract: A method and a system for extracting information from a table in a document is provided. The method includes: receiving a document that includes information that is arranged in a table; determining three sets of coordinates that respectively relate to lines, words, and characters included in the document; extracting a list of lines based on the first set of coordinates; reconstructing the rows of the table based on list of lines and the second set of coordinates; reconstructing the columns of the table based on the reconstructed rows and the third set of coordinates; and outputting a reconstruction of the table. The three sets of coordinates are expressible in an hOCR format that is based on an open standard for representation of scanned information that is obtainable by using an optical character recognition (OCR) technique.

Abstract: Techniques for aligning images generated by two cameras are disclosed. This alignment is performed by computing a relative 3D orientation between the two cameras. A first gravity vector for a first camera and a second gravity vector for a second camera are determined. A first camera image is obtained from the first camera, and a second camera image is obtained from the second camera. A first alignment process is performed to partially align the first camera's orientation with the second camera's orientation. This process is performed by aligning the gravity vectors, thereby resulting in two degrees of freedom of the relative 3D orientation being eliminated. Visual correspondences between the two images are identified. A second alignment process is performed to fully align the orientations. This process is performed by using the identified visual correspondences to identify and eliminate a third degree of freedom of the relative 3D orientation.

Abstract: A system for using artificial intelligence to select a compatible element. The system includes at least a server wherein the at least a server is configured to receive training data. The at least a server is configured to receive at least a biological extraction from a user. The at least a server is configured to receive at least a datum of user activity data. The at least a server is configured to select at least a compatible element as a function of the training data, the at least a biological extraction, and the at least a user activity data. The at least a server is configured to transmit the at least a compatible element to a user client device.

Abstract: According to an aspect of some embodiments, an information processing apparatus is configured to perform: displaying an operation screen in a display, the operation screen generated by instructing a WebView to process the obtained web contents; if a predetermined instruction is given through the operation screen, calling the native library by using a script contained in the web contents, and instructing the called native library to generate image data corresponding to the file; and displaying the image data generated by the native library in the operation screen.

Abstract: A method of price text detection by an imaging controller comprises obtaining, by the imaging controller, an image of a shelf supporting labels bearing price text, generating, by the imaging controller, a plurality of text regions containing candidate text elements from the image, assigning, by the imaging controller, a classification to each of the text regions, selected from a price text classification and a non-price text classification. The imaging controller, within each of a subset of the text regions having the price text classification: detects a price text sub-region and generates a price text string by applying character recognition to the price text sub-region. The method further includes presenting, by the imaging controller, the locations of the subset of text regions, in association with the corresponding price text strings.

Abstract: The present application provides a method, apparatus and electronic equipment for recognizing a posture of a target, a first receiving signal and a second receiving signal upon scattering of a transmitting signal from a target to be recognized are acquired, a first baseband signal is determined according to the first receiving signal and the transmitting signal, and a second baseband signal is determined according to the second receiving signal and the transmitting signal; and a category of the posture of the target to be recognized is finally determined according to the first baseband signal and the second baseband signal. The first baseband signal and the second baseband signal carry various feature values related to the posture of the target, including but not limited to transversal velocity information and radial velocity information, etc.

Abstract: A product evaluation device includes a receiving means for receiving product specifying information for specifying a product attribute, an acquisition means for acquiring designation in formation corresponding to a product attribute specified by the product specifying information from a storage means storing, for each product attribute, a genuine image showing a characteristic part for evaluating whether a product is a genuine product and designation information designating the characteristic part in association with each other, a request means for requesting a user to take a photograph by designating a characteristic part of the product based on the designation information acquired from the storage means and requesting a photographed image obtained by the photographing, and an output means for outputting an evaluation result obtained based on the photographed image provided in response to the request and the genuine image stored in the storage means.

Abstract: A system and method for localizing an autonomous vehicle using a digital stereo camera mounted to the autonomous vehicle and oriented to obtain images of the road surface. The system comprises an image processor configured to analyze the images of the road surface and identify micro-features, such as pores, holes, cracks, and granular material embedded in the street. The image processor is configured to compile previously obtained images of the road surface into reference mosaics and to compare the micro-features identified in newly obtained images to the reference mosaic to determine the location of the autonomous vehicle. The system may be further configured with a communications module to transmit images and localization information to a database repository or server.

Abstract: A method includes obtaining, using at least one processor, an input image frame. The method also includes identifying, using the at least one processor, one or more regions of the input image frame containing redundant information. In addition, the method includes performing, using the at least one processor, an image processing task using the input image frame. The image processing task is guided based on the one or more identified regions of the input image frame. The method may further include obtaining, using the at least one processor, a coarse depth map associated with the input image frame. Performing the image processing task may include refining the coarse depth map to produce a refined depth map, where the refining of the coarse depth map is guided based on the one or more identified regions of the input image frame.

Abstract: An information processing apparatus (2000) detects an estimation position (24), which is a position where a target object is estimated to be present, from a captured image (20) generated by a camera (10). The target object is a belonging of a person. The information processing apparatus (2000) detects a plurality of parts (22) constituting a body of a person included in the captured image (20). For example, the parts (22) are joints. The information processing apparatus (2000) detects the estimation position (24) of the target object in the captured image (20) using information relating to a position of each of a plurality of the detected parts (22).

Abstract: A processor comprises a plurality of processing units on an integrated circuit interconnected by an exchange. The exchange has a group of exchange paths extending between first and second portions of the integrated circuit. Each group has a first exchange block in the first portion and a second exchange block in the second portion. The processor has a first external interface in the first portion a second external interface in the second portion and a routing bus which routes packets between the external interfaces and the exchange blocks. The first external interface exchanges packets between the integrated circuit and a host. The second interface exchanges packets between the integrated circuit and another integrated circuit. Errors may be trapped when packets are wrongly addressed. A network of such processors is also provided.

Abstract: An automated robot design pipeline facilitates the overall process of designing robots that perform various desired behaviors. The disclosed pipeline includes four stages. In the first stage, a generative engine samples a design space to generate a large number of robot designs. In the second stage, a metric engine generates behavioral metrics indicating a degree to which each robot design performs the desired behaviors. In the third stage, a mapping engine generates a behavior predictor that can predict the behavioral metrics for any given robot design. In the fourth stage, a design engine generates a graphical user interface (GUI) that guides the user in performing behavior-driven design of a robot. One advantage of the disclosed approach is that the user need not have specialized skills in either graphic design or programming to generate designs for robots that perform specific behaviors or express various emotions.

Abstract: Systems and methods are disclosed for processing digital images to predict at least one continuous value comprising receiving one or more digital medical images, determining whether the one or more digital medical images includes at least one salient region, upon determining that the one or more digital medical images includes the at least one salient region, predicting, by a trained machine learning system, at least one continuous value corresponding to the at least one salient region, and outputting the at least one continuous value to an electronic storage device and/or display.

Abstract: A method, system and computer program product for progressively updating at least one matrix of license plate identification values is disclosed. At a vehicle presence time, an image is captured within which is shown a uniquely identifiable license plate of a vehicle. Analytics is carried out on the image to obtain, in relation to the uniquely identifiable license plate, at least four values in relation to both a license plate number and at least one additional plate-identifying information. The matrix of license plate identification values is populated with the at least four values and stored in a database.

Abstract: A CNN of an image forming apparatus includes: an encoder which compresses, for each tile image obtained by dividing an image into specific size pieces, information of the tile image; a decoder which restores the information of the tile image compressed by the encoder; and a blank sheet determination portion which determines whether the tile image corresponds to a blank sheet image. A segmentation image generation portion uses, when the blank sheet determination portion determines the tile image as being the blank sheet image, the blank sheet image for an image of a part corresponding to the tile image in a segmentation image, and uses, when the blank sheet determination portion determines the tile image as not being the blank sheet image, an output image of the decoder for an image of a part corresponding to the tile image in the segmentation image.

Abstract: Image analysis using visual geometry as an anchor for optical character recognition can be configured to receive an image acquired by a camera. The image is analyzed to detect a location within the image having a specified geometry. The specified geometry can be a predefined, visual geometry. The image is divided to create an image segment, where the image segment is based on the location of the specified geometry within the image. The image segment is analyzed to detect one or more characters within the image segment. The one or more characters in the image segment are decoded. A character string is generated based on decoding the one or more characters in the image segment.

Abstract: An apparatus comprises processing circuitry configured to receive a plurality of training image data sets and a plurality of predetermined displacements. The processing circuitry is further configured to use the training image data sets and predetermined displacements to train a transformation regressor in combination with a discriminator in an adversarial fashion by repeatedly alternating a transformation regressor training process in which the transformation regressor is trained to predict displacements, and a discriminator training process in which the discriminator is trained to distinguish between predetermined displacements and displacements predicted by the transformation regressor.

Abstract: Methods, apparatus, and systems are provided for obfuscating facial identity in images by synthesizing a new facial image for an input image. A base face is detected from or selected for an input image. Facial images that are similar to the base face are selected and combined to create a new facial image. The new facial image is added to the input image such that the input image includes a combination of the base face and the new facial image. Where no base face is detected in the input image, a base face is selected from reference facial images based at least on pose keypoints identified in the input image. After a new facial image is generated based on the selected base face, a combination of the new facial image and the base facial image are added to the input image by aligning one or more pose keypoints.

Abstract: A redundant tracking system comprising multiple redundant tracking sub-systems, enabling seamless transitions between such tracking sub-systems, provides a solution to this problem by merging multiple tracking approaches into a single tracking system. This system is able to combine tracking objects with six degrees of freedom (6DoF) and 3DoF through combining and transitioning between multiple tracking systems based on the availability of tracking indicia tracked by the tracking systems. Thus, as the indicia tracked by any one tracking system becomes unavailable, the redundant tracking system seamlessly switches between tracking in 6DoF and 3DoF thereby providing the user with an uninterrupted experience.

Abstract: The subject technology receives first code in a first programming language, the first code including a reference to a first dataset. The subject technology converting the first code to: second code in a second programming language different than the first programming language, and at least one data query in a second format different than the first programming language. Further the subject technology sends the converted first code and the at least one data query to a data system for execution.

Abstract: Systems and methods are disclosed for processing digital images to predict at least one continuous value comprising receiving one or more digital medical images, determining whether the one or more digital medical images includes at least one salient region, upon determining that the one or more digital medical images includes the at least one salient region, predicting, by a trained machine learning system, at least one continuous value corresponding to the at least one salient region, and outputting the at least one continuous value to an electronic storage device and/or display.

Abstract: Automatically identifying insights from a dataset and presenting the insights graphically and in natural language text ranked by importance is provided. Different data types and structures in the dataset are automatic recognized and matched with a corresponding specific analysis type. The data is analyzed according to the determined corresponding analysis types, and insights from the analysis are automatically identified. The insights within a given insight type and between insight types are ranked and presented in order of importance. Insights include those having multiple pipelined attributes and other insights include multiple insights identified as having some relationship for the included insights.

Abstract: Systems and methods are provided for identifying a product in an image and outputting stock keeping units of the product. The system comprises three main components: a database server, a data analytics system and a standard dashboard. The database server contains real-time inventory images as well as historical images of each product type. The data analytics system is executed by a computer processor configured to apply object detection and classification and deep learning algorithms to detect product information captured by the image. The data analytics system is also configured to determine hierarchical classification categories for the product. The standard dashboard is configured to output a report regarding the product information.

Abstract: An image processing method includes a matching cost calculating process of calculating matching costs in a unit of sub-pixels having higher resolution than first and second images by using an image of a reference area contained in the first image in which a target object is imaged and images of a plurality of comparison areas contained in the second image in which the target object is imaged, and a synthesized cost calculating process of calculating synthesized costs related to the reference area based on comparison results of values related to the plurality of matching costs calculated in the matching cost calculating process.

Abstract: A method and an apparatus for detecting a salient object in an image includes separately performing convolution processing corresponding to at least two convolutional layers on a to-be-processed image to obtain at least two first feature maps of the to-be-processed image, performing superposition processing on at least two first feature maps included in a superposition set in at least two sets to obtain at least two second feature maps of the to-be-processed image, the at least two sets are in a one-to-one correspondence with the at least two second feature maps, and a resolution of a first feature map included in the superposition set is lower than or equal to a resolution of a second feature map corresponding to the superposition set, and splicing the at least two second feature maps to obtain a saliency map.

Abstract: An apparatus including a location device and a processor. The location device may be configured to determine location coordinates of the apparatus. The processor may be configured to receive video frames captured by a capture device, perform video analysis on the video frames to detect objects in the video frames and extract metadata corresponding to the objects detected in the video frames, correlate the metadata with the location coordinates, determine a distance from the apparatus to the objects in the video frames and calculate an absolute location of the objects in response to the distance and the location coordinates. The distance may be determined by comparing a size of the objects detected in the video frames with a known size of the objects. The absolute location for the objects in the video frames may be added to the metadata.

Abstract: The present disclosure relates to an image processing method and apparatus, an electronic device, and a storage medium. The method includes: obtaining video streams of a game tabletop; detecting target objects in a plurality of image frames included in the video streams; determining a current game stage based on the target objects; and determining game detecting results according to the target objects and the determined game stage.

Abstract: A method of recognizing animals includes recognizing a plurality of body parts of a plurality of animals based on at least one image of the animals, in which the plurality of body parts include a plurality of types of body parts, including determining first estimated positions of the recognized body parts in the at least one image. The method includes estimating a plurality of first associations of body parts based on the at least one image of the animals, each first association of body parts associates a body part of an animal with at least one other body part of the same animal, including determining relative positions of the body parts in each estimated first association of body parts in the at least one image.

Abstract: A process for locating real estate parcels for a user comprises accessing a library of parceled real estate image data to identify objects and features in a plurality of parcels identified by the user as having a feature of interest. A predictive model is constructed and applied to a geographic region selected by the user to generate a customized output of real estate parcels predicted to have the feature of interest.

Abstract: Methods, systems, and devices for feature-based image autofocus are described. A device may perform an autofocus procedure that includes determining a set of features based on determining a feature region associated with an image. The device may generate a feature weight map based on the set of features and estimate a direction of a target feature in the feature region. The device may generate a direction weight map that corresponds to the feature region. The device may determine a focus position of the image based on the generated feature weight map and the estimated direction of the target feature and perform an autofocus operation on the determined focus position of the image. The device may calculate a focus value based on the feature weight map, and the focus position of the image may be based on the focus value.

Abstract: A target detection method and apparatus, in which the method includes: obtaining a target candidate region in a to-be-detected image; determining at least two part candidate regions from the target candidate region by using an image segmentation network, where each part candidate region corresponds to one part of a to-be-detected target; and extracting, from the to-be-detected image, local image features corresponding to the part candidate regions; and learning the local image features of the part candidate regions by using a bidirectional long short-term memory LSTM network, to obtain a part relationship feature used to describe a relationship between the part candidate regions; and detecting the to-be-detected target in the to-be-detected image based on the part relationship feature. As a result, image data processing precision in target detection can be improved, application scenarios of target detection can be diversified, and target detection accuracy can be improved.

Abstract: An image capturing apparatus includes: an image capturing unit that includes an imaging device and outputs an image signal obtained by image capturing of a subject by the imaging device through an image capturing optical system; a control unit that, in a case of controlling an exposure for each of three or more division regions obtained by dividing an image represented by the image signal and in a case of dividing the image into a plurality of segment regions that are different from the division regions and among which a segment region extends across a boundary between some of the division regions, controls an exposure of the segment region in accordance with information about at least one division region among the some of the division regions over which the segment region extends; and a display unit that displays the image for which the exposure is controlled by the control unit.

Abstract: Embodiments of the present disclosure provide an image classification method for a computer device. The method includes obtaining an original image and a category of an object included in the original image; adjusting a display parameter of the original image to satisfy a value condition to obtain an adjusted original image; and transforming the display parameter of the original image according to a distribution condition that distribution of the display parameter needs to satisfy, to obtain a transformed image. The method also includes training a neural network model based on the category of the object and a training set constructed by the adjusted original image and the transformed image; and determining a category of an object included in a to-be-predicted image based on the trained neural network model.

Abstract: Methods and systems for mapping images from a first dynamic range to a second dynamic range using a set of reference color-graded images and neural networks are described. Given a first and a second image representing the same scene but at a different dynamic range, a neural network (NN) model is selected from a variety of NN models to determine an output image which approximates the second image based on the first image and the second image. The parameters of the selected NN model are derived according to an optimizing criterion, the first image and the second image, wherein the parameters include node weights and/or node biases for nodes in the layers of the selected NN model. Example HDR to SDR mappings using global-mapping and local-mapping representations are provided.

Abstract: An image capturing method can be applied to an application, and include: determining a dimension adjustment approach set for a preset object in an acquired image to be processed after a dimension adjustment function of the application is turned on; and obtaining and displaying an adjusted image by adjusting a dimension of the preset object in the image to be processed according to the dimension adjustment approach, so as to support real-time preview of the adjusted image. As the dimension of the preset object in the image to be processed is adjusted in the image capturing process, the image capturing experience of the user is improved, and satisfying photos or videos can be obtained quickly.

Abstract: A computer-implemented method of associating an annotation with an object in an image, comprising generating a dictionary including first vectors that associate terms of the annotation with concepts, classifying the image to generate a second vector based on classified objects and associated confidence scores for the classified objects, selecting a term of the terms associated with one of the first vectors having a shortest determined distance to the second vector, identifying a non-salient region of the image, and rendering the annotation associated with the selected term at the non-salient region.

Abstract: An image analysis device may obtain target image data representing a target image which is an analysis target, specify (m×n) partial images sequentially by scanning the target image data, wherein the (m×n) partial images are constituted of m partial images aligned along a first direction and n partial images aligned along a second direction, generate first probability data by using the (m×n) partial images and the first object data in the memory, and reduce the target image data so as to generate reduced image data. The image analysis device may execute image analysis according to a convolutional neural network by using the reduced image data as K pieces of channel data corresponding to K channels and using the first probability data as one piece of channel data corresponding to one channel, and output a result of the image analysis.

Abstract: The present invention relates to a tool system comprising a handheld implement having a passive vectorized tracking marker permanently integrated with the implement at a predetermined location on the implement in a predetermined orientation with respect to the implement; and a database comprising geometric information describing the at least one of a rotationally asymmetric shape of the tracking marker and a rotationally asymmetric pattern disposed on the tracking marker. The system may further comprise: a tracker configured for obtaining image information about the tracking marker when the tracking marker is in a field of view of the tracker; and a controller having a processor and memory, the controller in communication with the database and the tracker, the processor programmable for receiving and processing image information from the tracker; accessing the database to retrieve the geometric information; and comparing the image information with the geometric information.

Abstract: Systems and methods are disclosed configured to train an autoencoder. A data training set is generated comprising images of different faces. A first autoencoder configuration is generated, comprising a first encoder, and a first decoder. The first autoencoder configuration is trained using dataset images, wherein weights associated with the first encoder and weights associated with the first decoder are modified. A second autoencoder configuration is generated comprising the first encoder and a second decoder. The second decoder is trained using a plurality of images of a first target face. First encoder weights are substantially maintained, and weights associated with the second decoder are modified. An autoencoder comprising the trained first encoder and the trained second decoder is used to generate an output using a source image of a first face having a facial expression, where the facial expression of the first face from the source image is applied to the first specific target face.

Abstract: The present disclosure relates to an indoor positioning method for operating an indoor positioning system and an indoor positioning apparatus by executing an artificial intelligence (AI) algorithm and/or a machine learning algorithm in a 5G environment connected for the Internet of Things. The indoor positioning method according to an embodiment of the present disclosure includes receiving map data and map information data of an indoor map in response to a presence of the indoor map of an indoor space, acquiring an image of the indoor space at a device camera, comparing image information of the indoor map with the acquired image information of the indoor space based on the map data and the map information data of the indoor space, and performing indoor localizing of the indoor space based on a result of the comparing.

Abstract: A pedestrian re-identification method includes: obtaining a target video containing a target pedestrian and at least one candidate video; encoding each target video segment in the target video and each candidate video segment in the at least one candidate segment separately; determining a score of similarity between the each target video segment and the each candidate video segment according to encoding results, the score of similarity being used for representing a degree of similarity between pedestrian features in the target video segment and the candidate video segment; and performing pedestrian re-identification on the at least one candidate video according to the score of similarity.

Abstract: An information processing device includes a first obtaining unit that obtains, from an external information processing device, image information added to a captured image published for external access by the external information processing device under a preset obtainment condition, a first storage that stores the image information obtained by the obtaining unit, and a first controller that categorizes the image information stored in the first storage under a preset attribute condition. The image information includes at least positional information and time information each added to the captured image. The first controller categorizes the image information at least based on the positional information and the time information.

Abstract: A framework for identifying prominent influencers/celebrities on social media webpages and customizing selection of thumbnails for presentation on social media webpages is provided. A most prominent influencer/celebrity associated with a social media webpage is determined. A presented thumbnail associated with each of a plurality of videos on a social media webpage is identified, each thumbnail including a picture/image of the most prominent influencer/celebrity. A feature vector is created for each identified presented thumbnail. Also identified is a plurality of potential thumbnails, each representative of a common video to be posted on the webpage and including an image of the identified most prominent influencer/celebrity. The feature vectors for the identified presented thumbnails and potential thumbnails are grouped into vector clusters. A centroid vector is identified for each vector cluster and a most prominent cluster is identified.

Abstract: Embodiments of the invention relate to a method for determining that an object in a sequence of images is a human. The method may include the steps of detecting an object in a first image from the sequence of images and assigning a first score to the object. The object is tracked to a second image from the sequence of images and a second score is assigned to the object in the second image. The second score is compared to a threshold that is inversely related to the first score and a determination that the object in the second image is a human is made based on the comparison.

Abstract: A method for generating a reduced-blur digital image representing a scene, the method being computer-implemented and comprising the following successive steps: i) providing at least two digital source images, a same element of the scene being represented in at least two source images, ii) selecting a reference image among the source images, iii) for at least one source image different from the reference image, and for at least one pixel of the reference image, a) defining a pattern in the reference image comprising pixels of the reference image, the element being represented in said pattern, b) constructing a map of coordinates that associates coordinates of the pattern in the reference image with the coordinates of the most similar pattern in the source image, c) optionally, filtering of the map of coordinates, and d) generating a corrected image by assigning to a pixel of the corrected image, the position of the pixel of the reference image and a color extracted from the source image point which position i