Abstract: A method, an apparatus, device and a storage medium for generating a target re-recognition model are provided. The method may include: acquiring a set of labeled samples, a set of unlabeled samples and an initialization model obtained through supervised training; performing feature extraction on each sample in the set of the unlabeled samples by using the initialization model; clustering features extracted from the set of the unlabeled samples by using a clustering algorithm; assigning, for each sample in the set of the unlabeled samples, a pseudo label to the sample according to a cluster corresponding to the sample in a feature space; and mixing a set of samples with a pseudo label and the set of the labeled samples as a set of training samples, and performing supervised training on the initialization model to obtain a target re-recognition model.

Abstract: A system for monitoring vehicle traffic may include a camera positioned to capture images within a license plate detection zone, the images may represent license plates of vehicles. The system may include an electronic device identification sensor that detects and stores electronic device identifiers of electronic devices located within an electronic device detection zone, and a computing system that detects, using the images, a license plate ID of a vehicle, compare the license plate ID of the vehicle to a database of trusted vehicle license plate IDs, identifies the vehicle as a suspicious vehicle, the identification based at least in part on the comparison of the license plate ID of the vehicle to the database of trusted vehicle license plate IDs, and correlates the license plate ID of the vehicle with at least one of the plurality of stored electronic device identifiers.

Abstract: Systems, methods, and computer-readable media are disclosed for visual labeling of training data items for training a machine learning model. Training data items may be generated for training the machine learning model. Visual labels, such as QR codes, may be created for the training data items. The creation of the training data item and the visual label may be automated. The visual labels and the training data items may be combined to obtain a labeled training data item. The labeled training data item may comprise a separator to distinguish the training data item from the visual label. The labeled training data item may be used for training and validation of the machine learning model. The machine learning model may analyze the training data item, attempt to identify the training data item, and compare the identification against the embedded label.

Abstract: An eyewear having an electronic processor configured to identify a hand gesture including a sign language, and to generate speech that is indicative of the identified hand gesture. The electronic processor uses a convolutional neural network (CNN) to identify the hand gesture by matching the hand gesture in the image to a set of hand gestures, wherein the set of hand gestures is a library of hand gestures stored in a memory. The hand gesture can include a static hand gesture, and a moving hand gesture. The electronic processor is configured to identify a word from a series of hand gestures.

Abstract: Embodiments of the disclosed technologies are capable of inputting, to a machine-learned classifier that has been created using a set of neural network-based models, multi-band digital image data that represents an aerial view of an agricultural field location containing an horticultural product; outputting, by the classifier, annotated image data, the annotated image data comprising annotation data indicative of individual instances of the horticultural product in the agricultural field location; using the annotated image data, computing a first predicted yield for the agricultural field location; adjusting the first predicted yield by a scaling factor to produce a second predicted yield for the agricultural field location.

Abstract: A method of controlling operation of equipment that spools an elongated member on and off a rotatable drum, in one or more embodiments, includes obtaining video data of a position of the elongated member on the rotatable drum. The method can also include feeding data into a trained artificial neural network and processing the data fed into the trained artificial neural network to determine at least one of a calculated position of the elongated member on the rotatable drum, a calculated fleet angle, or both. The method can also include actuating the rotatable drum to one of spool elongated member on and off the rotatable drum.

Abstract: The present invention provides a high-precision identification method and system for substations, including building a Mask RCNN objection recognition network model based on convolutional neural networks; inputting acquired image information of a object into the Mask RCNN object recognition network model for preliminary recognition and outputting a recognition result of the object; using an information entropy to create a semantic decision tree and correcting the recognition result of the object according to a principle of relative correlation between different objects and outputting a final recognition decision result; reading the recognition decision result to obtain a true type of the object to be recognized. The present invention greatly improves the accuracy of image recognition of substations, and has a positive role in the research and development of automatic inspection equipment for inspection robots.

Abstract: An apparatus includes: a camera configured to view a driver of a vehicle; and a processing unit comprising a first model and a second model; wherein the first model is configured to process the images of the driver and to output feature information based on the images, and wherein the second model is configured to receive the feature information from the first model, and to generate an output that indicates whether the driver is drowsy or not.

Abstract: An ultrasonic flaw detection device (A) includes: an ultrasonic probe (2) that emits ultrasonic waves to an inspection object (P) and detects reflected waves; a sheet material (1) attached to a surface of the inspection object and having two-dimensional patterns, the two-dimensional patterns being arranged on the inspection object and indicating positions on the inspection object; an imaging device (3) attached to the ultrasonic probe and imaging the two-dimensional patterns (1a); and a processing unit (21) that reads position information indicating a position on the inspection object from a captured image captured by the imaging device and relates a detection result of the ultrasonic probe to the position information, wherein the processing unit determines an index indicating the degree of quality of the detection result based on an air pocket area (HA) that is an area in which air pockets are shown in a predetermined area (H) of the captured image.

Abstract: A method of generating an image recognition model for recognising an input image and a system thereof are provided. The method includes appending at least one feature extraction layer to the image recognition model, extracting a plurality of feature vectors from a set of predetermined images, grouping the plurality of feature vectors into a plurality of categories, clustering the plurality of feature vectors of each of the plurality of categories into at least one cluster, determining at least one centroid for each of the at least one cluster, such that each of the at least one cluster comprises at least one centroid, such that each of the at least one centroid is represented by a feature vector, generating a classification layer based on the feature vector of the at least one centroid of the plurality of categories, and appending the classification layer to the image recognition model. In addition, a method of classifying an input image and a system thereof are provided.

Abstract: Disclosed herein are embodiments of systems, methods, and products comprise an analytic server, which provides a terrain segmentation and classification tool for synthetic aperture radar (SAR) imagery. The server accurately segments and classifies terrain types in SAR imagery and automatically adapts to new radar sensors data. The server receives a first SAR imagery and trains an autoencoder based on the first SAR imagery to generate learned representations of the first SAR imagery. The server trains a classifier based on labeled data of the first SAR imagery data to recognize terrain types from the learned representations of the first SAR imagery. The server receives a terrain query for a second SAR imagery. The server translates the second imagery data into the first imagery data and classifies the second SAR imagery terrain types using the classifier trained for the first SAR imagery. By reusing the original classifier, the server improves system efficiency.

Abstract: A method for splitting text line images includes receiving a text line image and identifying that the text line image comprises a plurality of zones, wherein each zone includes text whose font differs from the text of adjacent zones. The method further includes selecting a splitting position between multiple zones and splitting the text line image at the splitting position into a plurality of image segments, wherein each image segment contains at least one zone of the text line image and performing optical character recognition on each image segment to recognize a text segment of the image segment. In certain implementations, the method further includes generating one or more confidence measurements and selecting a splitting position that corresponds to a large gradient in the confidence measurement.

Abstract: A three-dimensional (3D) video reconstruction method, an encoder and a decoder are provided, comprising obtaining a list of video content screens or video content frames of an object from the 3D video; obtaining a list of depth screens of the 3D video; adding a shape screen to each video frame of the 3D video; superimposing each of the video content screens or video content frames with the depth screen and the shape screen to form a shape identification library; and storing the shape identification library at a header of a compressed file for unmasking of the object. The shape recognition list format may significantly reduce the storage size and increase the compression ratio by replacing the original shape with the identifications, and help improve the rendering quality.

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

Abstract: The present disclosure relates to an information processing device and method, and a program that enable processing, related to management of a vehicle, to be performed even in a case where no vehicle registration number is recognizable. Feature information of a vehicle is extracted from a captured image on the basis of a detection result of a number plate of the vehicle in the captured image, and the feature information of the vehicle extracted by the feature extraction unit is registered in a predetermined database. The present disclosure can be applied to, for example, an information processing device, an image processing device, a communication device, an electronic device, an information processing method, a program, and the like.

Abstract: A system receives an image including a live facial image of the user and an identity document including a photograph of the user. Moreover, the system calculates a facial match score by comparing facial features in the live facial image to facial features in the photograph. The system recognizes data objects and characters in the identity document using optical character recognition (OCR) and computer vision, and then identifies, based on the recognized data objects and characters, a type of the identity document. Further, the system calculates a document validity score by comparing the recognized characters and data objects to character strings and data objects known to be present in the identified type of the identity document. Additionally, the system determines and outputs the user's identity verification status based on comparing the facial match score to a facial match threshold and comparing the document validity score to a document validity threshold.

Abstract: A model including an information extraction layer that obtains image information of a training object in a depth image; a pixel point positioning layer that performs position estimation on a three-dimensional coordinate of human-body key points, defines a body part of the training object as a body component, and calibrates a three-dimensional coordinate of all human-body key points corresponding to the body component; a feature extraction layer that extracts a key-point position feature, a body moving speed feature, and a key-point moving speed feature for action recognition; a vector dimensionality reduction layer that combines the key-point position feature, the body moving speed feature, and the key-point moving speed feature as a multidimensional feature vector, and performs dimensionality reduction on the multidimensional feature vector; and a feature vector classification layer that classifies the multidimensional feature vector that is performed with dimensionality reduction, to recognize a fitness act

Abstract: A matching accelerator in the form of a hardware accelerator configured to perform matrix multiplication and/or additional operations is used to optimize keypoint matching. An SSE calculation may be determined by utilizing the matching accelerator to perform matrix multiplication to obtain a cost matrix for two sets of keypoint descriptors from two images. The hardware accelerator may determine a best cost calculation for each keypoint in each direction, which is utilized to perform keypoint matching.

Abstract: Implementations are described herein for tracking objects with changing appearances across temporally-disparate images. In various implementations, a first probability distribution over a plurality of classes may be determined for a first biological object depicted in a first image captured at a first point in time. The classes may represent stages of growth of biological objects. Additional probability distribution(s) over the plurality of classes may be determined for candidate biological object(s) depicted in a second image captured at a second point in time subsequent to the first point in time. The candidate biological object(s) may potentially match the first biological object depicted in the first image.

Abstract: A dance segment recognition method includes: extracting an audio segment from a video segment in response to the audio segment including a music segment with a time length greater than or equal to a pre-specified time length, using a video sub-segment that is included in the video segment and that corresponds to the music segment as a candidate dance segment and in response to a quantity of image frames showing a dance move in a plurality of image frames of the candidate dance segment being greater than a first pre-specified threshold, determining the candidate dance segment to be a dance segment. The dance segment recognition method can increase the speed of recognizing a dance segment in a video segment.