Abstract: Described are techniques for oblique image rectification. The techniques include receiving an original image depicting an oblique view of a circular object and pre-processing the original image into an edge image. The techniques further include generating, by a machine learning model based on the edge image, a heatmap including an ellipse formed by the oblique view of the circular object. The techniques further include computing ellipse parameters describing the ellipse of the heatmap. The techniques further include performing, using the ellipse parameters, an affine transformation on the original image to generate a rectified image, where the rectified image converts the ellipse to a circle.

Abstract: Described are techniques for oblique image rectification. The techniques include receiving an original image depicting an oblique view of a circular object and pre-processing the original image into an edge image. The techniques further include generating, by a machine learning model based on the edge image, a heatmap including an ellipse formed by the oblique view of the circular object. The techniques further include computing ellipse parameters describing the ellipse of the heatmap. The techniques further include performing, using the ellipse parameters, an affine transformation on the original image to generate a rectified image, where the rectified image converts the ellipse to a circle.

Abstract: Automated analog gauge reading is provided. The method comprises a computer system receiving input of an image and detecting at least one analog gauge in the image. The computer system corrects the orientation of the analog gauge in the image and detects scene text and tick labels on the analog gauge. The computer system determines a position of a pointer on the analog gauge relative to the scene text and outputs a gauge reading value based on an arithmetic progression of tick labels and angle of the pointer with respect to minimum and maximum values on the analog gauge.

Abstract: A computer-implemented method, comprising receiving one or more inputs associated with a master services agreement (MSA), and generating an output that includes smart contract information; performing a simulation and receiving user input to interactively adjust the smart contract information and implement a smart purchase agreement (SPA) in blockchain and in a state engine; and automatically generating a smart purchase order (SPO) under the MSA, wherein the SPO is automatically validated for a conflict against the MSA in real-time, and for the case of a conflict, a conflict warning is provided.

Abstract: A computer-implemented method, comprising receiving one or more inputs associated with a master services agreement (MSA), and generating an output that includes smart contract information; performing a simulation and receiving user input to interactively adjust the smart contract information and implement a smart purchase agreement (SPA) in blockchain and in a state engine; and automatically generating a smart purchase order (SPO) under the MSA, wherein the SPO is automatically validated for a conflict against the MSA in real-time, and for the case of a conflict, a conflict warning is provided

Abstract: Example implementations described herein are directed to an interface for a worker dispatch system configured to dispatch routes and schedules for workers on a manufacturing floor. Example implementations can involve a system configured to execute a lot ordering and route optimization for each of a plurality of workers; determine rack placement of the plurality of racks and total travel distance for each worker; generate a deployment plan for each worker from the lot ordering, the route optimization, the rack placement, and the total travel distance, the deployment plan involving a pick-up route and a schedule for each worker; provide the deployment plan to an interface configured to receive modifications to the deployment plan; and upon receiving instructions to deploy the deployment plan through the interface, provide the deployment plan to a device of each worker.

Abstract: Example implementations described herein are directed to systems and methods for skill assessment, such as hand washing compliance in hospitals, or assembling products in factories. Example implementations involve body part tracking (e.g., hands), skeleton tracking and deep neural networks to detect and recognize sub-tasks and to assess the skill on each sub-task. Furthermore, the order of the sub-tasks is checked for correctness. Beyond monitoring individual users, example implementations can be used for analyzing and improving workflow designs with multiple sub-tasks.

Abstract: A computer-implemented method of localization for an indoor environment is provided, including receiving, in real-time, a dynamic query from a first source, and static inputs from a second source; extracting features of the static inputs by applying a metric learning convolutional neural network (CNN), and aggregating the extracted features of the static inputs to generate a feature transformation; and iteratively extracting features of the dynamic query on a deep CNN as an embedding network and fusing the feature transformation into the deep CNN, and applying a triplet loss function to optimize the embedding network and provide a localization result.

Abstract: A computer-implemented method of localization for an indoor environment is provided, including receiving, in real-time, a dynamic query from a first source, and static inputs from a second source; extracting features of the static inputs by applying a metric learning convolutional neural network (CNN), and aggregating the extracted features of the static inputs to generate a feature transformation; and iteratively extracting features of the dynamic query on a deep CNN as an embedding network and fusing the feature transformation into the deep CNN, and applying a triplet loss function to optimize the embedding network and provide a localization result.

Abstract: A method of visualizing recommended pieces of media content is provided. The method includes identifying at least one piece of media content associated with a received content feature associated with a viewed piece of media content, selecting at least one additional piece of media content linked to the viewed piece of media content by a sequential relationship, generating a two-dimensional visualization based on a content similarity between the identified at least one piece of media content, the viewed piece of media content, and the selected at least one additional piece of media content and the sequential relationship; and displaying the generated two-dimensional visualization.

Abstract: Example implementations described herein are directed to an interface for a worker dispatch system configured to dispatch routes and schedules for workers on a manufacturing floor. Example implementations can involve a system configured to execute a lot ordering and route optimization for each of a plurality of workers; determine rack placement of the plurality of racks and total travel distance for each worker; generate a deployment plan for each worker from the lot ordering, the route optimization, the rack placement, and the total travel distance, the deployment plan involving a pick-up route and a schedule for each worker; provide the deployment plan to an interface configured to receive modifications to the deployment plan; and upon receiving instructions to deploy the deployment plan through the interface, provide the deployment plan to a device of each worker.

Abstract: Example implementations described herein are directed to systems and methods for skill assessment, such as hand washing compliance in hospitals, or assembling products in factories. Example implementations involve body part tracking (e.g., hands), skeleton tracking and deep neural networks to detect and recognize sub-tasks and to assess the skill on each sub-task. Furthermore, the order of the sub-tasks is checked for correctness. Beyond monitoring individual users, example implementations can be used for analyzing and improving workflow designs with multiple sub-tasks.

Abstract: Example implementations are directed to systems and methods for developing a topic model for a set of video series that include overlapping topics, wherein each video series includes segments directed to one or more of topics, wherein the topic model is based on topic distributions detected from the segments; analyzing auxiliary information, for each video series, to determine sequence information for the segments of the video series; and generating an array for topic transitions using sequence pattern mining on the distributions and the sequence information, wherein a next segment is determined for a selected segment based on an alignment decision using the array and sequence based scoring.

Abstract: A method of visualizing recommended pieces of media content is provided. The method includes identifying at least one piece of media content associated with a received content feature associated with a viewed piece of media content, selecting at least one additional piece of media content linked to the viewed piece of media content by a sequential relationship, generating a two-dimensional visualization based on a content similarity between the identified at least one piece of media content, the viewed piece of media content, and the selected at least one additional piece of media content and the sequential relationship; and displaying the generated two-dimensional visualization.

Abstract: Example implementations are directed to systems and methods for developing a topic model for a set of video series that include overlapping topics, wherein each video series includes segments directed to one or more of topics, wherein the topic model is based on topic distributions detected from the segments; analyzing auxiliary information, for each video series, to determine sequence information for the segments of the video series; and generating an array for topic transitions using sequence pattern mining on the distributions and the sequence information, wherein a next segment is determined for a selected segment based on an alignment decision using the array and sequence based scoring.