A method comprises identifying a representation of first statistical information in a document, identifying descriptive text that describes the first statistical information, determining whether the descriptive text accurately describes the first statistical information, and upon determination that the descriptive text does not accurately describe the first statistical information, generating alternative descriptive text that accurately describes the first statistical information.

Computer implemented systems and methods are disclosed for extracting specific data using machine learning algorithms. In accordance with some embodiments, a memory device that stores at least a set of computer executable instructions for a machine learning algorithm and a pre-fill engine; and at least one processor that executes the instructions that cause the pre-fill engine to perform functions that include: receiving electronic documents, seed dataset documents, and pre-fill questions; determining output data that enable navigation through the electronic documents using the machine learning algorithm; determining output questions that enable navigation through the electronic documents using the machine learning algorithm; determining output documents to enable navigation through the electronic documents using the machine learning algorithm; and presenting one or more answers for one or more of the output questions using a graphical user interface.

Computer implemented systems and methods are disclosed for extracting specific data using machine learning algorithms. In accordance with some embodiments, a memory device that stores at least a set of computer executable instructions for a machine learning algorithm and a pre-fill engine; and at least one processor that executes the instructions that cause the pre-fill engine to perform functions that include: receiving electronic documents, seed dataset documents, and pre-fill questions; determining output data that enable navigation through the electronic documents using the machine learning algorithm; determining output questions that enable navigation through the electronic documents using the machine learning algorithm; determining output documents to enable navigation through the electronic documents using the machine learning algorithm; and presenting one or more answers for one or more of the output questions using a graphical user interface.

Systems, methods and techniques for automatically recognizing two-dimensional real world objects with an augmented reality display device, and augmenting or enhancing the display of such real world objects by superimposing virtual images such as a still or video advertisement, a story or other virtual image presentation. In non-limiting embodiments, the real world object includes visible features including visible security features and a recognition process takes the visible security features into account when recognizing the object and/or displaying superimposed virtual images.

Systems, methods and techniques for automatically recognizing two-dimensional real world objects with an augmented reality display device, and augmenting or enhancing the display of such real world objects by superimposing virtual images such as a still or video advertisement, a story or other virtual image presentation. In non-limiting embodiments, the real world object includes visible features including visible security features and a recognition process takes the visible security features into account when recognizing the object and/or displaying superimposed virtual images.

Various implementations of the present disclosure relate to style transfer. In some implementations, a computer-implemented method comprises: obtaining a target object having a first style, a style of the target object being editable; obtaining a reference image including a reference object; obtaining a second style of the reference object, the second style of the reference object being extracted from the reference image; and applying the second style to the target object.

In some aspects, a method includes performing optical character recognition (OCR) based on data corresponding to a document to generate text data, detecting one or more bounded regions from the data based on a predetermined boundary rule set, and matching one or more portions of the text data to the one or more bounded regions to generate matched text data. Each bounded region of the one or more bounded regions encloses a corresponding block of text. The method also includes extracting features from the matched text data to generate a plurality of feature vectors and providing the plurality of feature vectors to a trained machine-learning classifier to generate one or more labels associated with the one or more bounded regions. The method further includes outputting metadata indicating a hierarchical layout associated with the document based on the one or more labels and the matched text data.

A computer model to identify a type of physical card is trained using simulated card images. The physical card may exist with various subtypes, some of which may not exist or be unavailable when the model is trained. To more robustly identify these subtypes, the training data set for the computer model includes simulated card images that are generated for the card type. The simulated card images are generated based on a semi-randomized background that varies in appearance, onto which an identifying marking of the card type is superimposed, such that the training data for the computer model includes additional randomized sample card images and ensure the model is robust to further variations in subtypes.

A computer model to identify a type of physical card is trained using simulated card images. The physical card may exist with various subtypes, some of which may not exist or be unavailable when the model is trained. To more robustly identify these subtypes, the training data set for the computer model includes simulated card images that are generated for the card type. The simulated card images are generated based on a semi-randomized background that varies in appearance, onto which an identifying marking of the card type is superimposed, such that the training data for the computer model includes additional randomized sample card images and ensure the model is robust to further variations in subtypes.

Structured documents are processed using convolutional neural networks. For example, the processing can include receiving a rendered form of a structured document; mapping a grid of cells to the rendered form; assigning a respective numeric embedding to each cell in the grid, comprising, for each cell: identifying content in the structured document that corresponds to a portion of the rendered form that is mapped to the cell, mapping the identified content to a numeric embedding for the identified content, and assigning the numeric embedding for the identified content to the cell; generating a matrix representation of the structured document from the numeric embeddings assigned to the cells of the grids; and generating neural network features of the structured document by processing the matrix representation of the structured document through a subnetwork comprising one or more convolutional neural network layers.