Abstract: A system and method for assessing video quality of a video-based application trains a neural network using training data of video samples and assesses video of the video-based application using the neural network to generate the subjective video quality information of the video-based application. Data augmentation is performed on video data, which is labeled with at least one subjective quality level, to generate the training data of video samples.

Abstract: A system includes a computing device that includes a memory configured to store instructions. The system also includes a processor to execute the instructions to perform operations that include receiving data representing an image, the image being represented in the data by a collection of visual elements. Operations also include determining whether to select the image for presentation by one or more entities using a machine learning system, the machine learning system being trained using data representing a plurality of training images and data representing one or more attributes regarding image presentation by the one or more entities.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training an image embedding model. In one aspect, a method comprises: obtaining training data comprising a plurality of training examples, wherein each training example comprises: an image pair comprising a first image and a second image; and selection data indicating one or more of: (i) a co-click rate of the image pair, and (ii) a similar-image click rate of the image pair; and using the training data to train an image embedding model having a plurality of image embedding model parameters.

Abstract: An image processing method for identifying text on production line components obtains an image to be recognized and a standard image for reference and extracts a first text area of the image to be recognized. A second text area of the standard image is obtained, and a text window is extracted based on the second text area. The method further obtains a target text area of the image to be recognized based on the first text area and the text window, and obtains a first set of first text sub-areas, and obtains a second set of second text sub-areas, by dividing the second text area into sub-windows of the text window. The method further marks the image to be recognized as a qualifying image when each first text sub-area of the first set is the same as a corresponding second text sub-area of the second set.

Abstract: A method for ascertaining a depth information image for an input image. The input image is processed using a convolutional neural network, which includes multiple layers that sequentially process the input image, and each converts an input feature map into an output feature map. At least one of the layers is a depth map layer, the depth information image being ascertained as a function of a depth map layer. In the depth map layer, an input feature map of the depth map layer is convoluted with multiple scaling filters to obtain respective scaling maps, the multiple scaling maps are compared pixel by pixel to generate a respective output feature map in which each pixel corresponds to a corresponding pixel from a selected one of the scaling maps.

Abstract: The method is for dividing dark objects, substructures and background of an image from an electron microscope into segments by analyzing pixel values. The segments are transformed and aligned so that the transformed objects, sub-structures and background are meaningfully comparable. The transformed segments are clustered into classes which are used for ontological investigation of samples that are visualized by using electron microscopy. A triangle inequality comparison can be used to further cluster groups of objects to transfer understanding from different interactions between objects and to associate interactions with each other.

Abstract: A system and method relate to providing machine learning predictions with defenses against patch attacks. The system and method include obtaining a digital image and generating a set of location data via a random process. The set of location data include randomly selected locations on the digital image that provide feasible bases for creating regions for cropping. A set of random crops is generated based on the set of location data. Each crop includes a different region of the digital image as defined in relation to its corresponding location data. The machine learning system is configured to provide a prediction for each crop of the set of random crops and output a set of predictions. The set of predictions is evaluated collectively to determine a majority prediction from among the set of predictions. An output label is generated for the digital image based on the majority prediction. The output label includes the majority prediction as an identifier for the digital image.

Abstract: Systems and methods for detecting hail damage on a vehicle are described including, receiving an image of at least a section of a vehicle. Detecting a plurality of hail damage including, detecting a plurality of damaged areas distributed over the entire section of the vehicle, and differentiating the plurality of damaged areas from one or more areas of noise, processing the received image to classify one or more sections of the vehicle as one or more panels of the vehicle bodywork, and using the detected areas of damage, the classification of the seriousness of the damage and the classification of one or more panels to compute a panel damage density estimate.

Abstract: System and techniques for vehicle environment modeling with a camera are described herein. A device for modeling an environment comprises: a hardware sensor interface to obtain a sequence of unrectified images representative of a road environment, the sequence of unrectified images including a first unrectified image, a previous unrectified image, and a previous-previous unrectified image; and processing circuitry to: provide the first unrectified image, the previous unrectified image, and the previous-previous unrectified image to an artificial neural network (ANN) to produce a three-dimensional structure of a scene; determine a selected homography; and apply the selected homography to the three-dimensional structure of the scene to create a model of the road environment.

Abstract: A system for generating a 3D segmentation of a target volume is provided. The system accesses views of an X-ray scan of a target volume. The system applies a 2D CNN to each view to generate a 2D multi-channel feature vector for each view. The system applies a space carver to generate a 3D channel volume for each channel based on the 2D multi-channel feature vectors. The system then applies a linear combining technique to the 3D channel volumes to generate a 3D multi-label map that represents a 3D segmentation of the target volume.

Abstract: An example electronic system is described in which an imaging device includes a lens and an image sensor. The imaging device is aligned with an optical target. The optical target includes a text character of a defined text size. An image capturer activates the imaging device to capture an electronic image of the optical target. The electronic image includes the text character of the optical target. An optical recognizer generates an optical recognition result for the character based on the captured electronic image. A sharpness detector compares the optical recognition result with a true value of the text character included in the optical target. Based on the comparison, a designated or defined text size is selected as a designated resolution. The designated resolution is then associable with the imaging device, the optical target, the electronic image, or a component thereof.

Abstract: An apparatus for feature recognition of two-dimensional prints is illustrated. The apparatus comprise a processor and a memory communicatively connected to the processor. The memory contains instructions configuring the processor to receive a two-dimensional print of a part for manufacture, scale two-dimensional print so that the two-dimensional print is within a predetermined area, identify a curve feature of the two-dimensional print as a function of scaling of the two-dimensional print, wherein the curve feature comprises a plurality of line segments, and classify a line type of the curve feature using line observations as a function of the curve feature identification.

Abstract: Methods and systems of implementing a convolutional neural network are described. In an example, a structure may receive input signals and distribute the input signals to a plurality of unit cells. The structure may include a plurality of multi-kernel modules that may include a respective set of unit cells. A unit cell may correspond to an element of a kernel being implemented in the convolutional neural network and may include a storage component configured to store a weight of a corresponding element of the kernel. A first pass gate of the unit cell may be activated to pass a stored weight of the unit cell to a plurality of operation circuits in the corresponding unit cell, such that the stored weight may be applied to the input signals. The structure may generate a set of outputs based on the application of the stored weights to the input signals.

Abstract: A system for vector extraction comprising a vector extraction engine stored and operating on a network-connected computing device that loads raster images from a database stored and operating on a network-connected computing device, identifies features in the raster images, and computes a vector based on the features, and methods for feature and vector extraction.

Abstract: Systems and methods for training an artificial intelligence (AI) classifier of scanned items. The items may include a training set of sample raw scans. The set may include in-class objects and not-in-class raw scans. An AI classifier may be configured to sample raw scans in the training set, measure errors in the results, update classifier parameters based on the errors, and detect completion of training.

Abstract: An information processing system includes: a sample data acquisition unit that acquires, for each sample, sample data in which a first cluster and a second cluster are associated with each other, the first cluster including a plurality of sets of a biological element detected from the sample and a biological element quantity indicating a quantity of the biological element, the second cluster including a plurality of sets of a morpheme regarding text describing an environment in which the sample is present and an appearance frequency of the morpheme; and a generation unit that analyzes a plurality of pieces of the sample data with the biological element quantity and the appearance frequency as parameters and generates information indicating a relationship between the environment and the first cluster.

Abstract: Embodiments are disclosed for training a neural network classifier to learn to more closely align an input image with its attribution map. In particular, in one or more embodiments, the disclosed systems and methods comprise receiving a training image comprising a representation of one or more objects, the training image associated with at least one label for the representation of the one or more objects, generating a perturbed training image based on the training image using a neural network, and training the neural network using the perturbed training image by minimizing a combination of classification loss and attribution loss to learn to align an image with its corresponding attribution map.

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: A vision-LiDAR fusion method and system based on deep canonical correlation analysis are provided. The method comprises: collecting RGB images and point cloud data of a road surface synchronously; extracting features of the RGB images to obtain RGB features; performing coordinate system conversion and rasterization on the point cloud data in turn, and then extracting features to obtain point cloud features; inputting point cloud features and RGB features into a pre-established and well-trained fusion model at the same time, to output feature-enhanced fused point cloud features, wherein the fusion model fuses RGB features to point cloud features by using correlation analysis and in combination with a deep neural network; and inputting the fused point cloud features into a pre-established object detection network to achieve object detection. A similarity calculation matrix is utilized to fuse two different modal features.

Abstract: Systems and methods that facilitate determining interaction between medications and the brain using a brain measure and a brain model. Hidden nervous system states are difficult to predict, diagnose, and treat with therapeutic medications. A Dual Neural Machine Translation (d-NMT) algorithmic system that utilizes sets of parameters for a relapsing-remitting MS model based on patient medical records and adjusts a method of parameterization to produce a model that can match patients' medical records and medical images. These parameters are can be used by a therapeutic determining model to recommend therapies, doses, and time courses accurately.