Abstract: An information processing apparatus and an information processing method capable of accurately recognizing an object to be sensed are provided. One or a plurality of learning models are selected from among learning models corresponding to a plurality of categories, a priority of each of the selected learning models is set, observation data obtained by sequentially compounding pieces of sensor data applied from a sensor is analyzed using the learning model and the priority of the learning model, a setting of sensing at a next cycle is selected based on an analysis result, predetermined control processing is executed in such a manner that the sensor performs sensing at the selected setting of sensing, and the learning model corresponding to the category estimated as the category to which a current object to be sensed belongs with a highest probability is selected based on the categories to each of which the previously recognized object to be sensed belongs.

Abstract: Systems and methods for image processing are described. Embodiments identify a training set including a first image that includes a ground truth blur classification and second image that includes a ground truth blur map, generate a first embedded representation of the first image and a second embedded representation of the second image using an image encoder, predict a blur classification of the first image based on the first embedded representation using a classification layer, predict a blur map of the second image based on the second embedded representation using a map decoder, compute a classification loss based on the predicted blur classification and the ground truth blur classification, train the image encoder and the classification layer based on the classification loss, compute a map loss based on the blur map and the ground truth blur map, and train the image encoder and the map decoder.

Abstract: The subject technology receives image data and depth data. The subject technology selects an augmented reality content generator corresponding to a three-dimensional (3D) effect. The subject technology applies the 3D effect to the image data and the depth data based at least in part on the selected augmented reality content generator. The subject technology generates, using a processor, a message including information related to the applied 3D effect, the image data, and the depth data.

Abstract: A virtual assistant negotiation method is provided, which includes the following steps: transmitting an event information by a first electronic device corresponding to an initiator; receiving a plurality of candidate projects generated by second electronic devices corresponding to a plurality of participants according to the event information by the first electronic devices; selecting a portion of the candidate projects to serve as recommended projects by the first electronic device of the host; and making a decision according to the opinions, for the recommended projects, corresponding to the main participant among the participants by the first electronic device.

Abstract: According to embodiments of the present disclosure, a method and an apparatus for processing an image, a device, and a storage medium are provided. The method includes: performing an image processing operation on an initial image having a noise associated with an adversarial sample attack, to obtain an intermediate image, the image processing operation including at least one of: reducing resolution of the initial image, or smoothing at least a part of the initial image; determining an image enhancement model matching the image processing operation, the image enhancement model being trained based on a sample image and a reference image, and the reference image being obtained by performing at least the image processing operation on the sample image; and generating a target image by processing the intermediate image using the image enhancement model, the target image having an image quality higher than the intermediate image.

Abstract: A medical image processing apparatus according to an embodiment includes processing circuitry configured to generate an output data set apparently expressing a second data set obtained by transmitting and receiving an ultrasound wave, for each scanning line, as many times as a second number that is larger than a first number, by inputting a first data set to a trained model that generates the output data set on a basis of the first data set obtained by transmitting and receiving an ultrasound wave as many times as the first number for each scanning line.

Abstract: A system and method are disclosed for training a system or a model to allow estimation of the value of livestock that is farmed for monetary gain. The various aspects of the invention include generation of data that is used to supplement or augment capture or real data, wherein the subject of the data is an animal. Labels or attributes are generated and validated.

Abstract: An information processing apparatus includes an acquisition unit, a calculation unit, and a generation unit. The acquisition unit acquires information including information regarding multiple nodes and information regarding multiple links connecting the multiple nodes and acquires constraint information regarding node pairs included in the multiple nodes. The constraint information includes a positive constraint and a negative constraint. The calculation unit calculates, for each of multiple clusters, a classification proportion into which the multiple nodes are classified and calculates a degree of importance of each of the multiple clusters. The classification proportion represents a proportion in which each of the multiple nodes is classified as one of the multiple clusters. The generation unit generates a probability model for performing probabilistic clustering on the multiple nodes.

Abstract: Input images are partitioned into non-overlapping segments perpendicular to a disparity dimension of the input images. Each segment includes a contiguous region of pixels spanning from a first edge to a second edge of the image, with the two edges parallel to the disparity dimension. In some aspects, contiguous input image segments are assigned in a “round robin” manner to a set of sub-images. Each pair of input images generates a corresponding pair of sub-image sets. Semi-global matching processes are then performed on pairs of corresponding sub-images generated from each input image. The SGM processes may be run in parallel, reducing an elapsed time to generate respective disparity sub-maps. The disparity sub-maps are then combined to provide a single disparity map of equivalent size to the original two input images.

Abstract: A method for performing automatic visual inspection includes: capturing visual information of an object using a scanning system including a plurality of cameras; extracting, by a computing system including a processor and memory, one or more feature maps from the visual information using one or more feature extractors; classifying, by the computing system, the object by supplying the one or more feature maps to a complex classifier to compute a classification of the object, the complex classifier including: a plurality of simple classifiers, each simple classifier of the plurality of simple classifiers being configured to compute outputs representing a characteristic of the object; and one or more logical operators configured to combine the outputs of the simple classifiers to compute the classification of the object; and outputting, by the computing system, the classification of the object as a result of the automatic visual inspection.

Abstract: An electronic apparatus may include a memory that stores first information regarding a plurality of first artificial intelligence models trained to perform image processing differently from each other and second information regarding a second artificial intelligence model trained to identify a type of an image by predicting a processing result of the image by each of the plurality of first artificial intelligence models. The electronic apparatus may further include a processor configured to identify a type of an input image by inputting the input image to the second artificial intelligence model stored in the memory, and process the input image by inputting the input image to one of the plurality of first intelligence models stored in the memory based on the identified type.

Abstract: This disclosure relates to automatic creation of an image processing pipeline that has a dependency on image processing domain experts. Given the image processing problem has multiple viable solutions, search space of creating a solution for a given goal in a given constrained infrastructure is large and, choosing an optimal image processing solution is an effort, time and intellect intensive endeavor. The present disclosure models image processing goals in abstract task categories and externalizes domain knowledge as an object model and a set of rules defined over it. Rules codify task template selection criteria and algorithm selection criteria for implementing a task. Recommendations are provided to choose suitable algorithm (s) for carrying out image processing tasks. Effects of algorithm execution are recorded for subsequent tasks recommendations. The stored state of execution and trace information enables rerouting the execution and knowledge upgrade.

Abstract: An information processing system includes an image reading device and an information processing device. The image reading device reads a document to generate target image data. The information processing device processes the target image data. The information processing device includes a first conversion processing section, a second conversion processing section, and a selection section. The first conversion processing section is capable of converting image data to character code data. The second conversion processing section is capable of converting image data to character code data. The selection section selects conversion of the target image data to character code data by the first conversion processing section or the second conversion processing section.

Abstract: A method and apparatus for generating a face model, a storage medium, a processor, and a terminal are provided. The method includes that: feature extraction is performed on a currently input face image from at least one dimension to obtain a plurality of facial features; classification and identification are performed according to the plurality of facial features to obtain a facial feature identification result; a mapping relationship between the multiple facial features and face pinching parameters set in a current face pinching system is acquired; and a corresponding face model is generated according to the facial feature identification result and the mapping relationship. The present disclosure solves the technical problem that a manual face pinching function provided in a game in the related art is time-consuming and laborious, and it is difficult to obtain a face pinching effect that fully meets psychological expectations.

Abstract: The present invention relates to the field of medical monitoring, and in particular non-contact, video-based monitoring of pulse rate, respiration rate, motion, and oxygen saturation. Systems and methods are described for capturing images of a patient, producing intensity signals from the images, filtering those signals to focus on a physiologic component, and measuring a vital sign from the filtered signals.

Abstract: A computer implemented method, including receiving, by a monitoring system that is configured to monitor a property and from a first camera that is trained on a vicinity of an entry point of the property, first image data, determining that a visitor is located at the vicinity of the entry point of the property, generating, by the monitoring system, an appearance model of the visitor, receiving, by the monitoring system and from a second camera that is trained on an area of the property other than the vicinity of the entry point of the property, second image data, comparing, by the monitoring system, the second image data to the appearance model of the visitor, determining a confidence score that reflects a likelihood that the visitor is located at the area of the property other than the vicinity of the entry point, and performing a monitoring system action.

Abstract: A processor-implemented method of personalizing a speech recognition model includes: obtaining statistical information of first scaling vectors combined with a base model for speech recognition; obtaining utterance data of a user; and generating a personalized speech recognition model by modifying a second scaling vector combined with the base model based on the utterance data of the user and the statistical information.

Abstract: A computer-implemented method and system may include identifying, by a recommendation engine, a first plurality of images in a first theme that includes a first layout characteristic, determining a first plurality of vector representations of the first plurality of images using an image processing model that includes a layered matrix of nodes with weighted interconnections, identifying at least a second plurality of images in a second theme that includes a second layout characteristic that is different from the first layout characteristic, determining a second plurality of vector representations of the second plurality of images, computing a vector difference between the first plurality of vector representations and the second plurality of vector representations; determining and communicating a recommendation based on the computed vector difference, and mapping received feedback to the weightings of the interconnections to improve a future recommendation.

Abstract: A system for enhancing a set of object scan images may be provided. The set of object scan images may also here comprise at least a first scan image, located at a distance therefrom a second scan image.

Abstract: Techniques are described for processing digital video data using one or more machine learning models to determine an outcome of an item placement operation within a fulfillment center environment. Video data is processed using one or more machine learning models to determine an estimated likelihood that an occurrence of a particular fulfillment center operation is depicted within the two or more instances of digital video data. Upon determining that the estimated likelihood exceeds a predefined threshold confidence level, the video data is processed using second one or more machine learning models to determine a bin placement prediction and a confidence value. A data repository for a control system for the fulfillment center environment is updated, based on the bin placement prediction and the confidence value.

Abstract: An information processing apparatus is provided that detects a moving object from an image, and detects a human body from a target region that is an upper part of a region corresponding to the moving object, using a feature of an upper body of a human body, to thereby reduce processing load.

Abstract: Described herein are methods of generating learning data to facilitate de-biasing the labeled location of an object of interest within an image. Methods may include: receiving sensor data, where the sensor data is a first image; determining reference corner locations of an object in the first image using image processing; generating observed corner locations of the object in the first image from the determined reference corner locations; generating a bias transformation based, at least in part, on a difference between the reference corner locations and the observed corner locations of the object in the first image; receiving sensor data from another image sensor of a second image; receiving observed corner locations of an object in the second image from a user; and applying the bias transformation to the observed corner locations of the object in the second image to generate de-biased corners for the object in the second image.

Abstract: An information processing apparatus includes an obtaining unit that obtains image data which is data obtained by digitizing a document including plural areas in one document, a setting unit that sets a determination standard for each of the areas based on a degree of necessity of a confirming operation for a text or a picture included in the area of the image data, and a determination unit that determines a method of the confirming operation for each of the areas based on the determination standard.

Abstract: A method for detecting structural defects includes generating a first probabilistic map of a critical location of a structural component based on a physics model of the critical location, capturing data of the critical location, generating a second probabilistic map based on an analysis of the captured data of the critical location, aligning the first probabilistic map and the second probabilistic map, combining the aligned first and second probabilistic maps, and determining if the structural component includes a structural defect based on the combined probabilistic maps.

Abstract: An information processing apparatus includes a determination, first, second and third receiving, and first and second checking units, and a controller. When a determination result matches a first determination result, the first checking unit outputs a matching determination result as the final output. The second checking unit checks the first determination result against a second. The first and second determination results are respectively received by the first and second receiving units. When the first determination result matches the second, the second checking unit outputs the matching determination result as the final output. When the input information is received, the controller operates the determination, first receiving, and first checking units. When it is determined that the determination result does not match the first determination result, the controller operates selectively one of a unit set and the third receiving unit. The unit set is constituted by the second receiving and checking units.

Abstract: Embodiments provide a computer implemented method for identifying and correcting a misspelling in a question answering (QA) system, wherein the QA system is coupled to a document corpus, and the document corpus includes a plurality of documents related to a particular domain. The method includes the following steps: receiving an input question and a plurality of passages, wherein the plurality of passages are extracted from the document corpus by the QA system; providing at least one alternate form for each token extracted from the input question and the plurality of passages; identifying at least one misspelled token; and scoring at least one alternate form of each identified misspelled token.

Abstract: In one embodiment, a system accesses a probability model associated with an image depicting a body. The probability model includes probability values associated with regions of the image and each probability value represents a probability of the associated region of the image containing a particular body part. The system selects a subset (e.g., 3) of the probability values based on a comparison of the probability values. For each selected probability value, the system identifies surrounding probability values surrounding the selected probability value and computes a probabilistic maximum based on the selected probability value and the surrounding probability values. Each probabilistic maximum is associated with a location within the regions associated with the selected probability value and the surrounding probability values.

Abstract: To prevent typos in electronic documents, quality checks are performed to identify portions of the document that have been misspelled. Major sources of false positives are words that have been transliterated from other languages into English. The quality control system described herein identifies transliterated words in electronic documents. The system may be trained to evaluate a set of words, and determine likelihoods that individual words in the set of words are transliterated (i.e., they are not native to the language and/or alphabet of the electronic document). In some embodiments, the model may be further configured to identify a language and/or alphabet that the individual word was transliterated from.

Abstract: Methods, systems, and apparatuses are described for isolated word training and detection. Isolated word training devices and systems are provided in which a user may provide a wake-up phrase from 1 to 3 times to train the device or system. A concatenated phoneme model of the user-provided wake-up phrase may be generated based on the provided wake-up phrase and a pre-trained phoneme model database. A word model of the wake-up phrase may be subsequently generated from the concatenated phoneme model and the provided wake-up phrase. Once trained, the user-provided wake-up phrase may be used to unlock the device or system and/or to wake up the device or system from a standby mode of operation. The word model of the user-provided wake-up phrase may be further adapted based on additional provisioning of the wake-up phrase.

Abstract: Disclosed is a method for generating a semantic image labeling model, comprising: forming a first CNN and a second CNN, respectively; randomly initializing the first CNN; inputting a raw image and predetermined label ground truth annotations to the first CNN to iteratively update weights thereof so that a category label probability for the image, which is output from the first CNN, approaches the predetermined label ground truth annotations; randomly initializing the second CNN; inputting the category label probability to the second CNN to correct the input category label probability so as to determine classification errors of the category label probabilities; updating the second CNN by back-propagating the classification errors; concatenating the updated first and second CNNs; classifying each pixel in the raw image into one of general object categories; and back-propagating classification errors through the concatenated CNN to update weights thereof until the classification errors less than a predetermined

Abstract: An endoscope apparatus includes a processor comprising hardware, a processor performs a focus control that controls a position of a focus lens to an in-focus position based on an in-focus evaluation value, the focus lens being included in an optical system that forms an image of a captured image that is acquired by an imaging section, and the in-focus evaluation value being calculated from a first area within the captured image, and a change-in-scene detection process that detects whether or not a change in scene has occurred from a second area that includes an area that differs from the first area, wherein the processor is set to a standby state when the position of the focus lens has been controlled to the in-focus position, and resumes the focus control process when a change in scene has been detected when the focus control section is set to the standby state.

Abstract: This disclosure describes a computerized system for managing medical information, the system including at least one computing device configured to receive a plurality of medical documents, determine skip grams comprising tokens for the plurality of medical documents, wherein each of the skip grams comprises one or more tokens of at least one of the medical documents, and populate a trie data structure based on the skip grams. The at least one computing device is further configured to prune the plurality of nodes of the trie based on one or more criteria to produce a pruned trie of nodes, determine rules for associating medical billing codes with the skip grams of the pruned trie of nodes based on pointwise mutual information, and output the determined rules.

Abstract: Dense feature scale detection can be implemented using multiple convolutional neural networks trained on scale data to more accurately and efficiently match pixels between images. An input image can be used to generate multiple scaled images. The multiple scaled images are input into a feature net, which outputs feature data for the multiple scaled images. An attention net is used to generate an attention map from the input image. The attention map assigns emphasis as a soft distribution to different scales based on texture analysis. The feature data and the attention data can be combined through a multiplication process and then summed to generate dense features for comparison.

Abstract: An information processing apparatus includes: determination units that make determinations on an input using different methods, and obtain determination results for the input; a first output unit that outputs, when a certain percentage or more of the determination results match, a determination result matched at the certain percentage or more; a second output unit that outputs, when the first output unit does not find a determination result matched at the certain percentage or more, a final determination result for the input; and an accuracy rate calculation unit that calculates, when a determination result obtained by a determination unit of interest among the determination units corresponds to a determination result matched at the certain percentage or more or matches the determination result output by the second output unit, an accuracy rate of the determination unit of interest, regarding that the determination result obtained by the determination unit of interest is correct.

Abstract: A method includes determining whether a region to be recognized corresponding to a specified feature exists in an image to be processed. The method may also include determining a main subject region containing the main subject of the image to be processed in accordance with a preset feature parameter of the main subject region and the coordinates and size of the region to be recognized when the region to be recognized exists. The method may further include extracting an image of the main subject region as a foreground image for an extraction process of a foreground target, and taking the extracted image as the main subject of the image to be processed.

Abstract: Aspects of the disclosure are related to a method for determining a touch pressure level on a touchscreen, comprising: detecting a touch event by the touchscreen; obtaining data relating to features associated with the touch event comprising a capacitance value, a touch area, and/or a touch duration; and determining a touch pressure level based on one or more of the features.

Abstract: A generating apparatus according to an embodiment of the present application includes an obtaining unit and a generating unit. The obtaining unit obtains intermediate images in an intermediate layer of a neural network that recognizes the quantity of objects contained in an image. The generating unit generates a combination image by combining together the intermediate images obtained by the obtaining unit. For example, the generating unit generates the combination image by combining together the intermediate images to which a weight has been applied on the basis of a change in the recognition of the quantity of objects corresponding to a process applied to each of the intermediate images.

Abstract: Systems and methods for transmission and display of synchronized physical and visual data for three-dimensional display are disclosed. Physical and visual data may be encoded and interlaced to enable synchronized transmission of the data in efficient manners. A single data transport stream may be utilized to transmit both physical and visual data over a communication channel, allowing physical and visual data to be efficiently co-joined on a same surface at a receiving end to provide a realistic, true three-dimensional representation.

Abstract: The delivery of a package using an unmanned aerial vehicle (UAV) may include performing a surveillance action at a property of an authorized party. The surveillance action may include imaging the property, and may be performed before or after the package is delivered. Surveillance data such as image data gathered in a surveillance action may be modified to remove or obscure adjacent properties not associated with the property of the authorized party. Based on surveillance data, the probability or confidence value of a surveillance event may be determined and a surveillance alert may be generated. Surveillance actions may be scheduled by a central controller to ensure that the surveillance action may be performed without exhausting the resources, such as battery power or time, of the UAV. Surveillance actions may be received as an interrupt and may have an associated priority level.

Abstract: Systems and processes for language identification from short strings are provided. In accordance with one example, a method includes, at a first electronic device with one or more processors and memory, receiving user input including an n-gram and determining a similarity between a representation of the n-gram and a representation of a first language. The representation of the first language is based on an occurrence of each of a plurality of n-grams in the first language and an occurrence of each of the plurality of n-grams in a second language. The method further includes determining whether the similarity between the representation of the n-gram and the representation of the first language satisfies a threshold.

Abstract: The present disclosure relates to a fluoroscopic inspection system for automatic classification and recognition of cargoes. The system includes: an image data acquiring unit, configured to perform scanning and imaging for a container by using an X-ray scanning device to acquire a scanned image; an image segmenting unit, configured to segment the scanned image into small regions each having similar gray scales and texture features; a feature extracting unit, configured to extract features of the small regions; a training unit, configured to generate a classifier according to annotated images; and a classification and recognition unit, configured to recognize the small regions by using the classifier according to the extracted features, to obtain a probability of each small region pertaining to a certain category of cargoes, and merge small regions to obtain large regions each representing a category.

Abstract: It is inter alia disclosed to classify samples of a set of samples of an image regarding whether or not a respective sample comprises optical information on a specific object type based on classification information. Therein, the classification information comprises a plurality of classifiers, each classifier being associated exclusively with one color channel of a plurality of color channels.

Abstract: In an example embodiment, an image transformation is automatically performed on a digital image to improve perceived professionalism of a subject of the image. A machine learning algorithm is utilized to generate a professionalism score for the digital image, the utilizing a machine learning algorithm comprising: a training mode where a plurality of sample images with labeled professionalism scores are used to train a classification function in a model that produces as professionalism score as output; an analysis mode where the model is used to generate a professionalism score for the digital image. Then the professionalism score is used as an input to a continuous variable optimization algorithm to determine an optimum version of the digital image from a plurality of possible versions of the digital image on which one or more image transformations have been performed, using the classification function.

Abstract: A method includes acquiring image data from for a given time step. The image data represents at least one of an instrument and a target within an image space that includes patient anatomy. A given belief of system state is estimated to include an estimate of instrument state and target state determined from corresponding measurement and motion models computed based on the acquired image data and control input parameters for the given time step. An expected belief of system state is estimated for a subsequent time step based on the estimated given belief of system state and based on the measurement models and the motion models computed over a plurality of different control inputs parameters. An information metric is computed for each of the expected beliefs of system state and control input parameters are selected for the subsequent time step based on the computed information metrics.

Abstract: One image processing method has at least the following steps: receiving an image input in a device, wherein the image input is composed of at least one source image; receiving image selection information; regarding a source image included in the image input, checking the image selection information to determine whether the source image is selected or skipped; and performing an object oriented image processing operation upon each selected source image. Another image processing method has at least the following steps: receiving an image input in a device, wherein the image input is composed of at least one source image; receiving algorithm selection information; and regarding a source image included in the image input, checking the algorithm selection information to determine a selected image processing algorithm from a plurality of different image processing algorithms, and performing an image processing operation upon the source image based on the selected image processing algorithm.

Abstract: A method and apparatus for encoding a source media program is disclosed. In one embodiment, the method comprises the steps of transcoding the source media program to produce a constant video quality transcoded version of the media program, determining if a portion of constant video quality version of the transcoded media program fails to satisfy a constraint, if the portion of the transcoded media program fails to satisfy the constraint, adjusting at least one transcoding parameter of a set of transcoding parameters, transcoding the portion of the source media program according to the adjusted at least one transcoding parameter to produce a constant bit rate version of the portion of the transcoded media program, and substituting the constant bit rate version of the portion of the transcoded media program for the constant video quality portion of the transcoded media program.

Abstract: Embodiments relate to a system and a method for identifying, from contractual documents, (i) standard exact clauses matching clause examples and (ii) non-standard clauses semantically related to but not matching the clause examples. A standard feature data set comprising standard exact clauses matching clause examples is obtained. In addition, a mirror feature data set comprising semantically related clauses of the clause examples is obtained using semantic language analysis, where the mirror feature data set encompasses the standard feature data set. Non-standard clauses are obtained by extracting a difference between the mirror feature data set and the standard exact feature data set.

Abstract: An abnormality detection method of detecting abnormality of a blast furnace from tuyere images shot by cameras installed in vicinities of a plurality of tuyeres of the blast furnace includes: collecting, in a time-series manner, representative brightness vectors defined by representative brightnesses determined based on brightness values of respective pixels for each of the tuyeres image previously shot by the cameras at a same time; extracting a principal component vector by performing principal component analysis on the representative brightness vectors collected in the time-series manner; calculating, as an evaluation value, a length of a normal line drawn in a direction of the principal component vector from the representative brightness vector collected from the tuyere images shot by the cameras at the same time during an operation; and detecting the abnormality of the blast furnace by comparing the evaluation value with a predetermined threshold.

Abstract: A method and system of providing training information for training a categorizer includes receiving a query relating to at least one category and identifying at least one case within a data set that matches the query. The method and system receives one of a first indication that the identified at least one case belongs to the category, and a second indication that the identified at least one case does not belong to the category. Training information is modified based on receiving one of the first indication and second indication.

Abstract: At least one example embodiment discloses an image feature extracting method. The method includes determining a probabilistic model based on pixel values of pixels in a kernel, determining image feature information of a current pixel of the pixels in the kernel and determining whether to change the image feature information of the current pixel based on a random value and a probability value of the current pixel, the probability value being based on the probabilistic model.