Abstract: Techniques for performing a prediction task using a multi-modal statistical model configured to receive input data from multiple modalities including input data from a first modality and input data from a second modality different from the first modality.

Abstract: A method, system and computer program product to generate a training data set for image segmentation applications, comprising providing a set of input documents of a first format. The input documents each comprise one or more pages. The input documents are split into individual document pages and parsed. Parsing comprises identifying a predefined set of items including position information of the position of the predefined set of items in the individual document pages; generating a bitmap image of a second format for each individual document page of the first format. The bitmap image comprises a predefined number of pixels. A mask is generated for each individual document. The mask comprises the predefined number of pixels of the corresponding bitmap image. Generating the mask comprises assigning an encoded class label to each pixel of the mask based on the position information of identified items of the predefined set of items.

Abstract: Disclosed herein are methods, apparatus, and systems for iris and periocular biometric recognition. A method for biometric recognition includes capturing one or more image frames of at least a facial portion of a user, determining an iris image portion and a periocular region image portion from the one or more image frames, performing iris encoding on a live iris image portion to generate an iris template, performing periocular region encoding on a live periocular region image portion to generate a periocular region template, performing iris matching on the iris template to generate an iris result, performing periocular region matching on the periocular region template to generate a periocular region result, and using one or more of the iris result and the periocular region result to enable the user to access an object.

Abstract: A method for optical imaging of single protein molecules including tethering single protein molecules via a flexible polymer linker to a glass slide having a surface coated with an indium tin oxide (ITO) so that the single protein molecules are tethered to the coated surface. The single protein molecules are driven into oscillation by applying an alternating electric field to the coated surface and the glass slide is located in the field of view of an objective lens. Incident light is directed onto the coated surface from an angle to generate an evanescent field and produce scattered light. The scattered light is collected and imaged by a CMOS imager to record a sequence of images of the scattered light. A Fast Fourier Transform (FFT) filter is applied to each pixel of the recorded image sequence to produce an oscillation amplitude image from which size, charge, and mobility of the plurality of single protein molecules can be determined.

Abstract: Certain aspects involve a machine-learning query system that uses a dual deep learning network to service queries and other requests. In one example, a machine-learning query system services a query received from a client computing system. A dual deep learning network included in the machine-learning query system matches an unstructured input data object, received from the client computing system, to an unstructured reference data object. The matching may include generating an input feature vector by an embedding subnetwork, based on the unstructured input data object. The matching may also include generating an output probability by a relationship subnetwork, based on the input feature vector and a relationship feature vector that is based on the unstructured reference data object. The machine-learning query system may transmit a responsive message to the client system.

Abstract: Digital imaging systems and methods are described for determining a user-specific skin redness value of a user's skin after removing hair. An example method may be performed by one or more processors and may include aggregating training images comprising pixel data of skin of individuals after removing hair. A skin redness model may be trained using the training images to output skin redness values associated with a degree of skin redness from least to most red. The method may include receiving an image of a user including pixel data of the user's skin after hair is removed from the skin, analyzing the image using the skin redness model to determine a user-specific skin redness value, generating a user-specific recommendation designed to address a feature identifiable within the pixel data of the user's skin, and rendering the recommendation on a display screen of a user computing device.

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: Method for storing a compressed digital image includes performing an artificial intelligence (“AI”) diagnostic test on the digital image; determining an uncompressed diagnostic result associated with the digital image based on the AI diagnostic test; generating one or more compressed digital images by compressing the digital image, each compressed digital image having a respective compression ratio; decompressing each compressed digital image to generate a respective decompressed digital image; performing the diagnostic test on each decompressed digital image; determining a decompressed diagnostic result associated with each decompressed digital image based on the respective diagnostic test; identifying one or more decompressed digital images having a respective decompressed diagnostic result that is the same as the uncompressed diagnostic result; selecting from the identified one or more decompressed digital images, the decompressed digital image associated with the highest compression ratio; and storing the

Abstract: A license plate identification method is provided, including steps of: obtaining a to-be-processed image including all characters on a license plate; extracting several feature maps corresponding to character features of the to-be-processed image through a feature map extraction module; for each of the characters, extracting a block and a coordinate according to the feature maps through a character identification model based on a neural network; and obtaining a license plate identification result according to the respective blocks and the respective coordinates of the characters.

Abstract: Apparatus and methods are described including introducing a cell suspension comprising red blood cells into a carrier that is a closed cavity that includes a base surface, via an inlet defined by the carrier. The cells in the cell suspension are allowed to settle on the base surface of the carrier to form a monolayer of cells on the base surface of the carrier. At least one microscope image of at least a portion of the monolayer of cells is acquired. Other applications are also described.

Abstract: A biometric capture device is operative to adjust one or more environmental parameters to enhance a range (e.g., distance) at which a biometric may be captured from a subject. For example, a sample biometric capture device may be, include, or otherwise incorporate a retinal or iris scanner configured to capture an image of the retina or iris (e.g., a biometric) when the retina or iris is illuminated by infrared light. Generally, the amount of infrared light required to accurately image the retina or iris increases with the distance of the subject's retina or iris from the image capture device. The biometric capture device may capture a facial image using a first image sensor, identify a face in the facial image, capture an iris image using a second image sensor guided by the facial image, and identify a person using the iris image.

Abstract: The present invention relates to a parameter training method for a convolutional neural network, CNN, for classifying data, the method comprising the implementation by data processing means (11c) of servers (1a, 1b, 1c) of steps of: (a1) Obtaining parameters of a set of at least one first CNN; (a2) For a first CNN of said set: Training, based on a database of already-classified public training data, parameters of a final representation block (B) of a second CNN corresponding to the first selected CNN to which said representation block (B) has been added; Retraining, based on a database of already-classified confidential training data of a secondary server (1a, 1b), parameters of the second CNN; Transmitting to the main server (1c) parameters of a third CNN corresponding to the second CNN without a final representation block (B); (a3) Replacing a first CNN of said set of first CNNs with the third CNN; (a4) Aggregating said set of at least one first CNN into a fourth CNN.

Abstract: A system and method for managing and tracking fishing/hunting rules and regulations is provided. The system generally comprises a global positioning system (GPS), a processor operably connected to the GPS, a power supply, a display operably connected to the processor, and a non-transitory computer-readable medium coupled to the processor and having instructions stored thereon. The system and method are designed to easily and conveniently compile and store information or data related to wildlife sports. The system may use this information to help a user comply with state and federal rules and regulations via the creation of registration strategies and via indicia.

Abstract: Method and apparatus embodiments can generate a volume fluorescence image of a tooth. Method and apparatus embodiments can project structured light patterns onto a tooth and generate a contour (volume) image of the tooth surface from acquired corresponding structured light projection images; then acquire one or more fluorescence images of the tooth generated under blue-UV illumination. A composite image that shows fluorescence image content mapped to the generated contour image can be transmitted, stored, modified and/or displayed.

Abstract: An information processing apparatus 100 comprising: first inference unit configured to perform a first inference to medical image data and obtain information related to a diagnostic name identified from the medical image data as a first inference result; and second inference unit configured to perform a second inference to the medical image data and the information related to the diagnostic name and obtain information related to an image finding as a second inference result.

Abstract: The present invention relates to a system and method for diagnosing potential diseases by making facial analysis via image processing techniques from photo and video data available in cloud storing unit and informing the user.

Abstract: A classification model building apparatus and a classification model building method thereof are provided. The classification model building apparatus introduces a clustering algorithm to assist in training a deep learning model for classification and takes a sum of a clustering loss function, a center concentration loss function and a classification loss function multiplied by different weights, respectively, as a total loss function for training the deep learning model. Based on the total loss function, the classification model building apparatus adjusts parameters of the deep learning model through a backpropagation algorithm to build a classification model.

Abstract: Methods and apparatus are provided for mapping and displaying a three dimensional (3D) surgical pathway. Entry and target voxels are selected along with a semi-target voxel therebetween. Where voxels representing bone are between the entry and target voxels, the semi-target voxel is selected such that voxels representing bone are not within a shortest line between it and the entry voxel. A series of voxels through the semi-target voxel define the pathway. For each voxel V in the series between the semi-target voxel and one of the endpoint voxels, the immediately succeeding voxel of voxel V is selected from among the group of neighbor voxels of voxel V by comparing selection weights of each voxel determined a selected basis including relative distances with respect to the endpoint voxel and the semi-target voxel. The pathway voxels are selectively highlighted in a displayed view to provide a visualization of the 3D surgical pathway.

Abstract: A medical evaluation system operates by: receiving a set of medical scans of a medical scan protocol captured for a patient, the set of medical scans corresponding to a proper subset of a plurality of sequence types; generating abnormality data by performing an inference function on the set of medical scans, wherein the inference function utilizes a computer vision model trained on a plurality of medical scans corresponding to the proper subset of the plurality of sequence types; calculating a confidence score for the abnormality data; generating first additional sequence data, wherein when the confidence score compares unfavorably to a confidence score threshold, the first additional sequence data indicates at least one first additional medical scan of the patient, corresponding to a first at least one of the plurality of sequence types not included in the proper subset of the plurality of sequence types, and when the confidence score compares favorably to the confidence score threshold, the first additional

Abstract: Systems and methods for determining illumination angle and viewing angle in color observation are described. The system includes an optical capturing device to acquire an image of its surroundings and a processing unit. When operated, the processing unit executes the following steps: identify a position of a light source in the image acquired by the optical capturing device; identify a position of a viewer in the image acquired by the optical capturing device; determine a first angular position of the light source with respect to the optical capturing device and a second angular position of the viewer with respect to the optical capturing device and provide the first angular position and the second angular position to define an actual viewing geometry.