Abstract: An intraoral camera system includes an imaging unit that generates image data by capturing an image of a tooth inside the mouth of a user, an area detector that detects the orientation of the imaging unit according to output by a multi-axis acceleration sensor and detects, according to the orientation detected, an area whose image is being captured by the imaging unit from among areas inside the mouth, the areas being determined by dividing a dentition into sections, and an identifier that narrows candidates which are combinations of tooth types and tooth positions down to fewer candidates to be used, according to the image data and the area detected, and identifies the type and the position of the tooth according to the narrowed candidates and the image data.

Abstract: Systems and methods of detecting and recognizing products on product storage structures of a product storage facility include an image capture device that moves about and captures images of the product storage structures at the product storage facility. A computing device processes the obtained images to detect and identify the products on the product storage structure, crops each of the identified individual products from the image to generate a plurality of cropped images and generates an image histogram template, feature vector template and location information template for each of the cropped images. The cropped images are stored in an electronic database and represent a reference model for each of the identified individual products and are stored in association with the generated image histogram template, feature vector template and location information template to facilitate recognition of products subsequently captured on the product storage structure by the image capture device.

Abstract: Systems and methods for image segmentation are described. Embodiments of the present disclosure receive an image depicting an object; generate image features for the image by performing a convolutional self-attention operation that outputs a plurality of attention-weighted values for a convolutional kernel applied at a position of a sliding window on the image; and generate label data that identifies the object based on the image features.

Abstract: A method of training a neural network model includes generating a positive image based on an original image, generating a positive text corresponding to the positive image based on an original text corresponding to the original image, the positive text referring to an object in the positive image, constructing a positive image-text pair for the object based on the positive image and the positive text, constructing a negative image-text pair for the object based on the original image and a negative text, the negative text not referring to the object, training the neural network model based on the positive image-text pair and the negative image-text pair to output features representing an input image-text pair, and identifying the object in the original image based on the features representing the input image-text pair.

Abstract: A system and method for polarization detection of a target in a strong background are provided. The system includes: an image signal acquisition module configured for image information acquisition and signal conversion of a plurality of channels; a YoLoV3 neural network-based feature extraction module configured to increase a number of network layers, obtain a fused feature map combined with intensity information and polarization information, and extract a feature difference between an intensity modality and a polarized light modality; a Darknet53 neural network-based feature extraction module configured to extract a feature value of a high dynamic range image; a residual network-based feature fusion module configured to learn weights of different modalities, obtain a fused feature map of intensity modality information and polarization modality information, and select and optimize the fused feature map by attention mechanism processing; and an image printing module configured to filter background information.

Abstract: A control circuit accesses a digitized image of a particular item and then processes that digitized image to thereby assign various aspects of the digitized image to various dimensions in N-dimensional Euclidean space to thereby provide an N-dimensional representation. The control circuit then accesses a database containing a plurality of various clustered groups of N-dimensional representations. At least some of the clustered groups have a corresponding identifying label. Conducting a nearest neighbor search serves to identify a clustered group to which the particular item most likely belongs to thereby provide an identified cluster group. When the identified cluster group does not have a corresponding identifying label, the control circuit can then further determine whether a predetermined condition has also been met. When true, the control circuit alerts a user via a user interface that the identified cluster group should be assigned an identifying label.

Abstract: Methods and systems are provided for generating enhanced image. A neural network system is trained where the training includes training a first neural network that generates enhanced images conditioned on content of an image undergoing enhancement and training a second neural network that designates realism of the enhanced images generated by the first neural network. The neural network system is trained by determine loss and accordingly adjusting the appropriate neural network(s). The trained neural network system is used to generate an enhanced aesthetic image from a selected image where the output enhanced aesthetic image has increased aesthetics when compared to the selected image.

Abstract: A method of verifying a user for transportation purposes is disclosed. The method may include using a communication apparatus to detect a face of the user. The method may include using the communication apparatus to instruct the user to perform a specific action, to validate that the specific action is performed by the user, to extract a frame from the specific action to use as an image, to obtain image parameters from the frame and to use the communication apparatus to send the image to a server for the server to determine whether the image is a genuine face by comparing the image parameters of the image with parameters in a database to obtain a comparison result and to use the comparison result to determine if the user should be verified.

Abstract: An object detection method includes inputting an input image to a learned machine learning model and generating a similarity image from an output of at least one specific layer, and generating a discriminant image to which at least an unknown label is assigned, by comparing a similarity of each pixel in the similarity image to a predetermined threshold value.

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed to categorize image text. An example apparatus includes region detection model training circuitry to identify candidate regions in an input image that include text, and generate bounding boxes around respective ones of the identified candidate regions. The example apparatus also includes mask application circuitry to improve optical character recognition (OCR) by applying a mask to the input image, wherein the mask removes content of the input image except for portions of the input image within the bounding boxes, and OCR circuitry to perform OCR on the masked input image to obtain text data within the bounding boxes.

Abstract: A method includes: obtaining a first image including a target item; selecting a plurality of reference images corresponding to the first image from a database; performing word segmentation on item text information corresponding to the plurality of reference images to obtain a plurality of words; and extracting a key word meeting a reference condition from the plurality of words, and determining the extracted key word as an item name of the target item.

Abstract: A system or technique can analyze metadata from a first set of images using a neural network to identify classification information about individuals depicted in the images. The system generates data based on this classification information, including attributes of the individuals, and compares this data against predefined criteria for image presentations. If any attributes are found to be non-compliant with the criteria, the system queries a database of second image metadata linked to a set of images. This database includes data generated from prior presentations and classification information provided by individuals depicted in the images. The system locates compliant attributes in the second image metadata, retrieves an image linked to these compliant attributes, and modifies a presentation to include the retrieved image.

Abstract: A method for estimating ripeness of produce includes illuminating the produce with light, measuring intensities of the light reflected from the produce at different frequencies, and determining a degree of ripeness of the produce from the relative intensities of the light reflected from the produce at the different frequencies.

Abstract: The disclosure relates to technology for monitoring driver attentiveness in a vehicle. A driver distraction system collects vehicle data and scene information from the vehicle while traveling on a route. The vehicle data and scene information are then processed to generate a reference heat map. At the same time, the driver distraction system may capture a gaze of a driver to track a gaze direction and duration of the driver while driving the vehicle on the route. The gaze direction and duration are processed to generate a driver gaze heat map. The driver gaze heat map and reference heat map are analyzed to determine a level of driver distraction of the driver in the vehicle, and a recommendation or warning is output to the driver of the vehicle according to the level of driver distraction.

Abstract: According to an aspect of the disclosure, a method of point cloud geometry encoding in a point cloud encoder is provided. In the method, geometry coding is performed on a point cloud at a first partition depth. Further, a plurality of largest coding units (LCUs) of the point cloud is determined at a second partition depth. A coding state of a LCU of the plurality of LCUs of the point cloud is set at the second partition depth. The geometry coding is performed on the plurality of LCUs of the point cloud at the second partition depth based on the coding state of the LCU at the second partition depth.

Abstract: Methods and systems for providing an interactive image scene graph pattern search are provided. A user is provide with an image having a plurality of selectable segmented regions therein. The user selects one or more of the segmented regions to build a query graph. Via a graph neural network, matching target graphs are retrieved that contain the query graph from a target graph database. Each matching target graph has matching target nodes that match with the query nodes of the query graph. Matching target images from an image database are associated with the matching target graphs. Embeddings of each of the query nodes and the matching target nodes are extracted. A comparison of the embeddings of each query node with the embeddings of each matching target node is performed. The user interface displays the matching target images that are associated with the matching target graphs.

Abstract: The present disclosure provides a retrieval and push method based on a fine art image tag, including the following steps: establishing a tag model database: training different tag contents with training samples of different subjects and different categories, and categorizing the training samples in the database according to knowledge point tags to obtain the tag model database; retrieval and push: uploading fine art work image samples to the trained tag model database, then extracting knowledge point tags of the fine art work image samples, retrieving associated fine art works, and then pushing the associated fine art works; real-time updating: recording knowledge point tags of input fine art work image samples by the tag model database in real time, and establishing a common tag and a model, and subsequently increasingly pushing contents relevant to the common tag; and generating a user portrait.

Abstract: An image processor receives first image data representing an image. The first image data comprising a plurality of color values corresponding to a plurality of pixels in the image. The image processor determines, using a trained machine learning model, second image data based on the first image data. The second image data comprises surface spectral reflection values corresponding to the plurality of pixels in the image, where the surface spectral reflection values are distributed across a plurality of wavelengths of visible light in the image. The image processor then performs at least one image processing operation with respect to the image using the second image data.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods that implement an image filter for enhancing light text and removing document shadows. In particular embodiments, the disclosed systems use a modified adaptive thresholding approach the relies on image gradients to efficiently guide the thresholding process. In addition, the disclosed systems use a machine-learning model to generate a document shadow map. The document shadow map can include text reflections. Accordingly, the disclosed systems remove text reflections from the document shadow map (e.g., by using an interpolated shadow intensity value of neighboring shadow map pixels). In turn, the disclosed systems use the document text mask and the document shadow map cleaned of text reflections to remove shadows from the digital image. Further, the disclosed systems enhance text in the shadow-removed digital image based on contrast stretching.

Abstract: A method, device and computer program product, the method comprising: obtaining access to a classifier trained upon a multiplicity of sets of decoded coefficients; obtaining a set of block coefficients associated with at least a part of the compressed image; and applying the classifier to the set of block coefficients, to obtain a classification of the compressed image.