Abstract: In a computer-implemented method and system for capturing the condition of a structure, the structure is scanned with an unmanned aerial vehicle (UAV). Data collected by the UAV corresponding to points on a surface of a structure is received and a 3D point cloud is generated for the structure, where the 3D point cloud is generated based at least in part on the received UAV data. A 3D model of the surface of the structure is reconstructed using the 3D point cloud.

Abstract: A system and methods are provided in which an artificial intelligence inference module identifies targeted information in large-scale unlabeled data, wherein the artificial intelligence inference module autonomously learns hierarchical representations from large-scale unlabeled data and continually self-improves from self-labeled data points using a teacher model trained to detect known targets from combined inputs of a small hand labeled curated dataset prepared by a domain expert together with self-generated intermediate and global context features derived from the unlabeled dataset by unsupervised and self-supervised processes. The trained teacher model processes further unlabeled data to self-generate new weakly-supervised training samples that are self-refined and self-corrected, without human supervision, and then used as inputs to a noisy student model trained in a semi-supervised learning process on a combination of the teacher model training set and new weakly-supervised training samples.

Abstract: A method for identifying errors associated with named entity recognition includes recognizing a candidate named entity within a text and extracting a chunk from the text containing the candidate named entity. The method further includes creating a feature vector associated with the chunk and analyzing the feature vector for an indication of an error associated with the candidate named entity. The method also includes correcting the error associated with the candidate named entity.

Abstract: Various embodiments described herein relate to techniques for forecasting with state transitions and confidence factors. In this regard, a system is configured to segment data associated with one or more assets to determine a set of classifications for one or more attributes related to the one or more assets. The system is also configured to generate a state machine associated with a Markov chain model based on the set of classifications for the data. Furthermore, the system is configured to perform a machine learning process associated with the state machine to determine one or more behavior changes associated with the one or more attributes related to the one or more assets. The system is also configured to predict, based on the one or more behavior changes associated with the one or more attributes related to the one or more assets, a change in demand data for the one or more assets during a future interval of time.

Abstract: Certain aspects of the present disclosure provide techniques for automatically detecting and classifying tumor regions in a tissue slide. The method generally includes obtaining a digitized tissue slide from a tissue slide database and determining, based on output from a tissue classification module, a type of tissue of shown in the digitized tissue slide. The method further includes determining, based on output from a tumor classification model for the type of tissue, a region of interest (ROI) of the digitized tissue slide and generating a classified slide showing the ROI of the digitized tissue slide and an estimated diameter of the ROI. The method further includes displaying on an image display unit, the classified slide and user interface (UI) elements enabling a pathologist to enter input related to the classified slide.

Abstract: Machine-readable storage media having instructions stored therein that, when executed by a processor of a mobile device, configure the mobile device to capture a check image for funds to be deposited into a recipient account. The mobile device configured to display a request to a user of the mobile device to provide one or more portions of a MICR line for the received check image and receive user inputs from the user specifying the one or more portions of the MICR line. The mobile device configured to transmit a message to a bank account computer system associated with the recipient account, the message including data specifying the one or more portions of the MICR line.

Abstract: Intent determination based on one or more multi-model structures can include generating an output from each of a plurality of domain-specific models in response to a received input. The domain-specific models can comprise simultaneously trained machine learning models that are trained using a corresponding local loss metric for each domain-specific model and a global loss metric for the plurality of domain-specific models. The presence or absence of an intent corresponding to one or more domain-specific models can be determined by classifying the output of each domain-specific model.

Abstract: A method is provided for Cross Video Temporal Difference (CVTD) learning. The method adapts a source domain video to a target domain video using a CVTD loss. The source domain video is annotated, and the target domain video is unannotated. The CVTD loss is computed by quantizing clips derived from the source and target domain videos by dividing the source domain video into source domain clips and the target domain video into target domain clips. The CVTD loss is further computed by sampling two clips from each of the source domain clips and the target domain clips to obtain four sampled clips including a first source domain clip, a second source domain clip, a first target domain clip, and a second target domain clip. The CVTD loss is computed as |(second source domain clip?first source domain clip)?(second target domain clip?first target domain clip)|.

Abstract: Systems and methods for performing direct conversion of image sensor data to image analytics are provided. One such system for directly processing sensor image data includes a sensor configured to capture an image and generate corresponding image data in a raw Bayer format, and a convolution neural network (CNN) coupled to the sensor and configured to generate image analytics directly from the image data in the raw Bayer format. Systems and methods for training the CNN are provided, and may include a generative model that is configured to convert RGB images into estimated images in the raw Bayer format.

Abstract: The present disclosure relates generally to artificial intelligence (AI), machine learning (ML), and deep learning technologies. More specifically, the disclosure relates to a vehicle image composite system that employs computer vision (CV) along with a Generative Adversarial Network (GAN) to generate realistic composite car images. For example, in one or more embodiments, the composite car image generator system trains a Convolutional Neural Network (CNN) to learn the Make Model Year parameters of all vehicle images provided. Once trained, the determined Make Model Year parameters of the vehicles allow the CNN to produce realistic composite images of a vehicle of any make, model, year, and trim level.

Abstract: An electronic device and a method for document segmentation are provided. The method includes: obtaining a first feature map and a second feature map corresponding to an original document; performing a first upsampling on the second feature map to generate a third feature map; concatenating the first feature map and the third feature map to generate a fourth feature map; inputting the fourth feature map to a first inverted residual block (IRB) and performing a first atrous convolution operation based on a first dilation rate to generate a fifth feature map; inputting the fourth feature map to a second IRB and performing a second atrous convolution operation based on a second dilation rate to generate a sixth feature map; concatenating the fifth feature map and the sixth feature map to generate a seventh feature map; performing a convolution operation on the seventh feature map to generate a segmented document.

Abstract: A method may include obtaining first sensor data captured by a first sensor system and second sensor data captured by a second sensor system of a different type from the first sensor system. The method may include detecting a first object included in the first sensor data and a second object included in the second sensor data. The method may include assigning a first label to the first object and a second label to the second object after comparing the first and the second sensor data. The first and second labels may indicate degrees to which the first and the second objects match. Responsive to the first and second labels indicating that the first and the second objects match, the method may include designating a matched object representative of the first object and the second object and sending the matched object to a downstream computing system of an autonomous vehicle.

Abstract: An image processing apparatus comprises: an obtaining unit configured to obtain an image and distance information concerning a distance from an in-focus plane, which corresponds to each pixel included in the image; a setting unit configured to set an image processing condition according to the distance information based on an output characteristic of an output apparatus concerning a sharpness; and a processing unit configured to perform image processing for the image using the distance information obtained by the obtaining unit and the image processing condition set by the setting unit, wherein the processing unit changes, in accordance with the distance information, a band of a spatial frequency of the image to which the image processing is applied.

Abstract: According to some embodiments, disclosed are systems and methods for machine learning-based image detection and the determination of slippery conditions based therefrom. The disclosed systems and method identify a set of images that depict captured imagery in relation to at least one area of a floor at a location. These images are then analyzed via at least one slippery condition detection machine learning algorithm, which results in a determination of a classification of the area of the floor (e.g., does a puddle exist or other type of slippery condition). This information is stored and later used for training of the at least one slippery condition detection machine learning algorithm. Moreover, the information is communicated to beacons in/around the location, to alert users to the condition.

Abstract: An Artificial Intelligence system, an apparatus and, a computer program product and a method for automatic improvement of artificial intelligence classification models. A model-performance measurement of the classification model is iteratively improved by at least a predetermined target goal in each iteration. The iterative improvement comprises generating a hypotheses graph for improving the classification model, based on a list of hypotheses and scores thereof. Each hypothesis relates to a strategy for potentially improving the classification model, and is associated with a score indicating a likelihood that an application thereof improves the model-performance measurement. Each node of the hypotheses graph comprises a hypothesis of the list of hypotheses.

Abstract: A fingerprint sensing module comprising a fingerprint sensor device having a sensing array arranged on a first side of the device, the sensing array comprising an array of fingerprint sensing elements. The fingerprint sensor device comprises connection pads for connecting to external circuitry. The fingerprint sensing module further comprises a fingerprint sensor device cover structure, arranged to cover the fingerprint sensor device, having a first side configured to be touched by a finger, thereby forming a sensing surface of the sensing module, and a second side facing the sensing array, wherein the cover structure comprises conductive traces for electrically connecting the fingerprint sensor module to external circuitry, and wherein a surface area of the cover structure is larger than a surface area of the sensor device. The fingerprint sensor device comprises wire-bonds electrically connecting the connection pads of the fingerprint sensing device to the conductive traces of the cover structure.

Abstract: Various embodiments include methods and neural network computing devices implementing the methods, for generating an approximation neural network. Various embodiments may include performing approximation operations on a weights tensor associated with a layer of a neural network to generate an approximation weights tensor, determining an expected output error of the layer in the neural network due to the approximation weights tensor, subtracting the expected output error from a bias parameter of the layer to determine an adjusted bias parameter and substituting the adjusted bias parameter for the bias parameter in the layer. Such operations may be performed for one or more layers in a neural network to produce an approximation version of the neural network for execution on a resource limited processor.

Abstract: Provided are a method and a system for analyzing image data obtained by photographing a tunnel by a drone using artificial intelligence, in tunnel maintenance inspection, rapidly and accurately finding a part that requires maintenance of the tunnel, and calculating a maintenance solution and a maintenance estimate for the part. The system for maintaining a tunnel by analyzing tunnel image data received from a drone using artificial intelligence, includes: the drone that photographs a tunnel to generate the tunnel image data; a position signal generating apparatus that is provided inside the tunnel and generates a position signal for determining position information of the drone in the tunnel; and an artificial intelligence tunnel maintenance apparatus that finds a part of the tunnel that requires maintenance, and calculates an optimal maintenance solution and an optimal maintenance estimate necessary for the tunnel maintenance.

Abstract: A client device configured with a neural network includes a processor, a memory, a user interface, a communications interface, a power supply and an input device, wherein the memory includes a trained neural network received from a server system that has trained and configured the neural network for the client device. A server system and a method of training a neural network are disclosed.

Abstract: A method of producing a model to detect changes in forest cover is disclosed. The method includes obtaining forest-cover classification data of a land area. The land area includes one or more subregions having unchanged forest-cover classifications between a first time and a second time. The method further includes obtaining image data of the subregions at multiple times. For at least one forest-cover classification, the method includes applying a statistical analysis to the image data to determine one or more threshold values representing measurement variations. The method further includes comparing subsequently obtained image data to the one or more threshold values and classifying the one or more subregions as changed or unchanged based on the comparison of subsequently obtained image data to the one or more threshold values.