Abstract: One embodiment provides a system that facilitates efficient collection of training data. During operation, the system obtains, by a recording device, a first image of a physical object in a scene which is associated with a three-dimensional (3D) world coordinate frame. The system marks, on the first image, a plurality of vertices associated with the physical object, wherein a vertex has 3D coordinates based on the 3D world coordinate frame. The system obtains a plurality of second images of the physical object in the scene while changing one or more characteristics of the scene. The system projects the marked vertices on to a respective second image to indicate a two-dimensional (2D) bounding area associated with the physical object.

Abstract: One embodiment provides a method for facilitating real-world interaction with virtual reality. During operation, the system receives, by a computing device from a virtual reality device associated with a user, instructions to configure physical components, wherein for a first physical component at a first location, the instructions indicate a type and an orientation, and wherein for a second physical component located at a second location, the instructions indicate a type, a length of extension, and an angle. The system executes, by a pose-adjusting unit, the instructions, which involves: physically moving the first physical component to the indicated orientation at the first location; physically extending the second physical component from the second location by the indicated length; and physically rotating the extended second physical component by the indicated angle. The system renders, on the virtual reality device, the configured physical components.

Abstract: One embodiment provides a system that facilitates efficient collection of training data for training an image-detection artificial intelligence (AI) engine. During operation, the system obtains a three-dimensional (3D) model of a physical object placed in a scene, generates a virtual object corresponding to the physical object based on the 3D model, and substantially superimposes, in a view of an augmented reality (AR) camera, the virtual object over the physical object. The system can further configure the AR camera to capture a physical image comprising the physical object in the scene and a corresponding AR image comprising the virtual object superimposed over the physical object, and create an annotation for the physical image based on the AR image.

Abstract: A method of monitoring user interactions with a networked device includes receiving log data of a control signal of a networked device, the networked device associated with a user. An interaction monitoring system may analyze the log data of the control signal to determine a temporal feature of the control signal. A processing device may then classify a user interaction level for the user based on the determined temporal features and generate a feedback response to the user based on the user interaction level.

Abstract: A method of monitoring user interactions with a networked device includes receiving image data associated with one or more images of a user, computing a plurality of visual features of the one or more images from the image data, and calculating, by a processing device, a style characteristic from the plurality of visual features. The method include calculating, by the processing device, a lifestyle characteristic based on the style characteristic and calculating, by the processing device, a user preference based on the lifestyle characteristic. The method includes determining a recommendation of a product based on the user preference and providing the recommendation to a user device associated with the user.

Abstract: A method of monitoring user interactions with a networked device includes receiving log data of a control signal of a networked device, the networked device associated with a user. An interaction monitoring system may analyze the log data of the control signal to determine a temporal feature of the control signal. A processing device may then classify a user interaction level for the user based on the determined temporal features and generate a feedback response to the user based on the user interaction level.

Abstract: One embodiment provides a system that facilitates efficient collection of training data. During operation, the system obtains, by a recording device, a first image of a physical object in a scene which is associated with a three-dimensional (3D) world coordinate frame. The system marks, on the first image, a plurality of vertices associated with the physical object, wherein a vertex has 3D coordinates based on the 3D world coordinate frame. The system obtains a plurality of second images of the physical object in the scene while changing one or more characteristics of the scene. The system projects the marked vertices on to a respective second image to indicate a two-dimensional (2D) bounding area associated with the physical object.

Abstract: A system and method schedules jobs in a cluster of compute nodes. A job with an unknown resource requirement profile is received. The job includes a plurality of tasks. Execution of some of the plurality of tasks is scheduled on compute nodes of the cluster with differing capability profiles. Timing information regarding execution time of the scheduled tasks is received. A resource requirement profile for the job is inferred based on the received timing information and the differing capability profiles. Execution of remaining tasks of the job is scheduled on the compute nodes of the cluster using the resource requirement profile.

Abstract: A method (100) is provided for characterizing a functionality of a flash (14) from at least one image (200) captured with a camera (12) using the flash (14). The method (100) includes: analyzing the image (200) by segmenting (106) the image (200) into at least one sub-region (206, 208), and applying a metric (108) to the sub-region (206, 208) to measure an image property within the sub-region (206, 208); and determining (110) the functionality of the flash (14) in response to a result of the applied metric.

Abstract: One embodiment of the present invention provides an energy-management system for managing energy within a predominantly closed power system. The system includes an upper-level energy-management module, at least one lower-level energy-management module coupled to the upper-level energy-management module, and at least one component coupled to the lower-level energy-management module. The component is configured to provide a performance model associated with the component to the lower-level energy-management module. The lower-level energy-management module is configured to provide an aggregated model to the upper-level energy-management module.

Abstract: One embodiment of the present invention provides a system for multi-domain clustering. During operation, the system collects domain data for at least two domains associated with users, wherein a domain is a source of data describing observable activities of a user. Next, the system estimates a probability distribution for a domain associated with the user. The system also estimates a probability distribution for a second domain associated with the user. Then, the system analyzes the domain data with a multi-domain probability model that includes variables for two or more domains to determine a probability distribution of each domain associated with the probability model and to assign users to clusters associated with user roles.

Abstract: A method for detecting a persistent change in a dynamically varying scene includes: obtaining a set of reference images of the scene; transforming the reference images into an abstract feature space; classifying pixels of the reference images in the abstract feature space; generating a stable reduced-reference image based on the classifications of corresponding pixels; obtaining a set of test images of the scene; transforming the test images into the abstract feature space; classifying pixels of the test images in the abstract feature space; generating a stable test image based on the classifications of corresponding pixels; and comparing the stable reduced-reference and test images to one another to detect a difference therein, the difference corresponding to a persistent change in the dynamically varying scene occurring between when the reference images and the test images were obtained.

Abstract: One embodiment of the present invention provides a system for clustering heterogeneous events. During operation, the system finds a partition of events into clusters such that each cluster includes a set of events. In addition, the system estimates probability distributions for various properties of events associated with each cluster. The system obtains heterogeneous event data, and analyzes the heterogeneous event data to determine the distribution of event properties associated with clusters and to assign events to clusters.

Abstract: A system for compiling a machine operable diagnostic system includes a header unit, a component unit, an effects unit, a test unit, and an instantiation unit. The header unit identifies general properties of a system from a generic description in an electronic format which includes at least one category of a plurality of variations. The component unit identifies a plurality of components of a modeled system from the generic description, and adds and separates the plurality of variations to the plurality of components, and each identified component includes at least one type of failure and a probability of the at least one type of failure. The effects unit identifies from the generic description a plurality of observable effects. The test unit identifies a plurality of tests and test outcomes designed to elicit the observable effects. The instantiation unit constructs a procedural instantiation of the machine operable diagnostic system.

Abstract: A recommender system uses an activity decision tree to model the changes in a user's behavior according to a plurality of time-related features. The system determines historical activities for the user, and generates a decision tree for the user's historical activities. Each leaf node of the decision tree is associated with an activity-prediction model that computes a probability for a corresponding activity. The system selects a path of the decision tree from a root node to a leaf node of the decision tree based on a target time. The selected path traverses two or more non-leaf nodes that are each associated with a temporal decision model that compares the target time against a temporal classifier. The system then determines a probability for a user activity based on an activity-prediction model of the selected path.

Abstract: One embodiment of the present invention provides a system for identifying anomalies. During operation, the system obtains work practice data associated with a plurality of users. The work practice data includes a plurality of user events. The system further categorizes the work practice data into a plurality of domains based on types of the user events, models user behaviors within a respective domain based on work practice data associated with the respective domain, and identifies at least one anomalous user based on modeled user behaviors from the multiple domains.

Abstract: One embodiment of the present invention provides a system for generating one or more recommendations for a customer. During operation, the system obtains transaction and image data for a plurality of existing customers. The system then trains one or more parameters of conditioning variables associated with one or more clusters based on image data as part of a predictive model. Next, the system determines a list of recommendable items for each cluster, based on the transaction data. The system obtains transaction and image data for a customer. The system then determines that the customer is a member of a cluster associated with the predictive model, based on the obtained transaction and image data. The system generates a recommendation for one or more recommendable items for the customer based on the determined cluster membership.

Abstract: One embodiment of the present invention provides managing component driver for an energy-management system that manages energy within a predominantly closed power system. The component driver includes a receiving mechanism configured to receive current status information for a component associated with the component driver, a model-construction mechanism configured to construct a performance model for the component, and a transmitting mechanism configured to transmit the performance model to an optimization and control module.

Abstract: A method (100) is provided for detecting an obstruction within a field of view of a camera (12) from an image (200) captured by the camera (12). The method (100) includes: analyzing the image (200) by applying edge detection (104) to the image (200), identifying (108) regions of the image (200) lacking edge content and comparing (112) a size of the identified regions to a threshold; and determining if there is an obstruction based upon a result of said comparison.

Abstract: A system includes a plurality of processors programmed to perform distributed, iterative computations on aggregate statistics of a data set. Each of the processors maintains global statistics describing the data set. The processors are programmed to iteratively perform a plurality of Gibbs sampling sweeps over a corresponding subset of the data set to update the corresponding global statistics. The processors are further programmed to maintain differential vectors representing the changes made to the global statistics by local computation. These can be broadcast to other processors to create a consistent global statistic across processors. The differential vectors can be saved for a number of past generations to provide updates for slow processors, processors that have been down for some time and newly added processors to the network. Additional methods are described for efficiently compressing multiple differential vector updates in to a smaller number of equally informative updates.