Abstract: A computer-implemented method of field calibrating a device is disclosed. In one example, a reference image is acquired, via an infrared camera, while the device is in a factory-calibrated state. The reference image measures a factory-calibrated spatial relationship of one or more infrared-visible fiducial markers of the device relative to the infrared camera while the device is in the factory-calibrated state. Later, a field image is acquired, via the infrared camera. The field image measures an updated spatial relationship of the one or more infrared-visible fiducial markers relative to the infrared camera. The device is field calibrated based on the reference image and the field image.

Abstract: The disclosed techniques provide enhanced eye tracking systems utilizing joint estimation of biological parameters and hardware parameters. By the use of joint estimation of biological parameters, e.g., direction and position of an eye, with concurrent estimation of hardware parameters, e.g., camera position or camera direction, a system can self-calibrate and provide eye tracking estimations that can allow a system to accommodate for deformations and other changes of a device. The disclosed techniques include a method to model changes of a device, as well as detect and compensate for them while the eye-tracking device is in normal use, thereby preserving the performance of the system without requiring a factory-calibration procedure to be repeated.

Abstract: A computer-implemented method of field calibrating a device is disclosed. In one example, a reference image is acquired, via an infrared camera, while the device is in a factory-calibrated state. The reference image measures a factory-calibrated spatial relationship of one or more infrared-visible fiducial markers of the device relative to the infrared camera while the device is in the factory-calibrated state. Later, a field image is acquired, via the infrared camera. The field image measures an updated spatial relationship of the one or more infrared-visible fiducial markers relative to the infrared camera. The device is field calibrated based on the reference image and the field image.

Abstract: Techniques for training an embedding using a limited training set are described. In some examples, the embedding is trained by generating a plurality of vectors from a random sample of the limited set of training data classes using a layer of the particular machine learning classification model, randomly selecting samples from the plurality of vectors into a set of samples, computing at least one distance for each sampled class from a center parameter for the class using the set of samples, generating a discrete probability distribution over the classes for a query point based on distances to a center parameter for each of the classes in the embedding space, calculating a loss value for the modified prototypical network, the calculation of the loss value being for a fixed geometry of the embedding space and including a measure of the difference between distributions, and back propagating.

Abstract: Approaches herein provide for multi-camera calibration and subsequent application in augmented reality applications. The approach obtains image data of a three-dimensional (3D) calibration object from different directions using the at least one camera. A determination is made for extrinsic parameters associated with a relative position of the 3D calibration object to the at least one camera. A further determination is made for intrinsic parameters associated with the at least one camera relative to the 3D calibration object facing in the different positions. Revised intrinsic parameters are generated from the intrinsic parameters. The revised intrinsic parameters provide a calibration output that may be applied to generate views of an item in an augmented reality application.

Abstract: Objects can be rendered in three-dimensions and viewed and manipulated in an augmented reality environment. For example, contour information for an object represented in image data for a plurality of viewpoints can be obtained. A region of space that includes the object can be partitioned into a plurality of sub regions. A position of each sub region can be identified in the image data for each viewpoint. A value can be determined for each sub region indicating whether that sub region as projected in the image data is within the contour of the object. Once the total values for each sub region are determined, a visual hull technique can be used to generate a visual hull of the object based on the values. Thereafter, the visual hull can be used to generate a three-dimensional representation of the object for a number of different viewpoints.

Abstract: Methods, apparatus and computer-readable media for detecting potential defects in a part are disclosed. A potential defect may be automatically detected in a part, and may be reported to an operator in various ways so that the operator may review the defect and take appropriate action.

Abstract: Methods, apparatus and computer-readable media for detecting potential defects in a part are disclosed. A potential defect may be automatically detected in a part, and may be reported to an operator in various ways so that the operator may review the defect and take appropriate action.

Abstract: Methods, apparatus and computer-readable media for detecting potential defects in a part are disclosed. A potential defect may be automatically detected in a part, and may be reported to an operator in various ways so that the operator may review the defect and take appropriate action.

Abstract: A method for determining a pose of a camera includes obtaining both a first image of a scene and a second image of the scene, where both the first and second images are captured by the camera. A first set of features is extracted from the first image and a second set of features is extracted from the second set of features. The method includes calculating a value of a visual-motion parameter based on the first set of features and the second set of features without matching features of the first set with features of the second set. The method also includes determining the pose of the camera based, at least, on the value of the visual motion parameter.

Abstract: A first map comprising local features and 3D locations of the local features is generated, the local features comprising visible features in a current image and a corresponding set of covisible features. A second map comprising prior features and 3D locations of the prior features may be determined, where each prior feature: was first imaged at a time prior to the first imaging of any of the local features, and lies within a threshold distance of at least one local feature. A first subset comprising previously imaged local features in the first map and a corresponding second subset of the prior features in the second map is determined by comparing the first and second maps, where each local feature in the first subset corresponds to a distinct prior feature in the second subset. A transformation mapping a subset of local features to a subset of prior features is determined.

Abstract: In an example embodiment, a method of calculating end-of-life (EOL) predictions for a physical asset is provided. A state-space model for the physical asset is obtained, the state-space model being a physics-based model describing a state of the physical asset at a particular time given measurements or observations for the physical asset. Then a current state of the physical asset is inferred. Then a long-term prediction is derived for the physical asset based on the inferred current state of the physical asset and the state-space model for the physical asset. Then an EOL probability distribution function is generated for the physical asset based on the long-term prediction, the EOL probability distribution function describing a range of estimates of EOL for the physical asset and their corresponding confidence intervals.

Abstract: A method for determining a pose of a camera includes obtaining both a first image of a scene and a second image of the scene, where both the first and second images are captured by the camera. A first set of features is extracted from the first image and a second set of features is extracted from the second set of features. The method includes calculating a value of a visual-motion parameter based on the first set of features and the second set of features without matching features of the first set with features of the second set. The method also includes determining the pose of the camera based, at least, on the value of the visual motion parameter.

Abstract: Methods, apparatus and computer-readable media for detecting potential defects in a part are disclosed. A potential defect may be automatically detected in a part, and may be reported to an operator in various ways so that the operator may review the defect and take appropriate action.

Abstract: In an example embodiment, a method of calculating end-of-life (EOL) predictions for a physical asset is provided. A state-space model for the physical asset is obtained, the state-space model being a physics-based model describing a state of the physical asset at a particular time given measurements or observations for the physical asset. Then a current state of the physical asset is inferred. Then a long-term prediction is derived for the physical asset based on the inferred current state of the physical asset and the state-space model for the physical asset. Then an EOL probability distribution function is generated for the physical asset based on the long-term prediction, the EOL probability distribution function describing a range of estimates of EOL for the physical asset and their corresponding confidence intervals.

Abstract: A technique is provided for measuring an object based on multiple views. The technique includes registering each of the plurality of images and a first model of the object with one another and reconstructing a second model of the object based on the plurality of images and the first model registered with one another.

Abstract: A method for obtaining multi-material decomposition images of an object is presented. The method includes acquiring an image pair from a dual energy computed tomography scan of the imaged object. The method then includes selecting a material basis for multi-material decomposition of the image pair. The method further includes applying a physicochemical model for the material basis. Also, the method includes performing multi-material decomposition using at least one constraint imposed by the physicochemical model.

Abstract: Methods and systems for processing images are provided. One method includes obtaining a plurality of images corresponding to at least one area of an object and performing a rectification of at least some of the plurality of images using a reference structure. The method also includes performing a gradient vector field analysis on the rectified plurality of images of the object to identify anomaly regions within the object.

Abstract: An apparatus for multi-energy tissue quantification includes an x-ray imaging system comprises an x-ray source configured to emit a beam of x-rays toward an object to be imaged, a detector configured to receive the x-rays attenuated by the object, and a data acquisition system (DAS) operably coupled to the detector. A computer operably connected to the x-ray source and the DAS is programmed to cause the x-ray source to emit x-rays at each of a first kVp and a second kVp toward the detector, acquire x-ray data from x-rays emitted at the first and second kVp through a region of interest (ROI), and perform a first multi-material decomposition based on the acquired x-ray data. The computer is also programmed to quantify a volume fraction of a first material in the ROI based on the first multi-material decomposition and display the volume fraction of the first material to a user.

Abstract: Approaches are described for generating an initial reconstruction of CT data acquired using a wide-cone system. In one implementation, frequency data may be patched in to a first scan, such as an axial scan, from a second scan, such as a helical scan. In one embodiment, an initial reconstruction may be processed (such as via a non-linear operation) to correct frequency omissions and/or errors in the reconstruction. Corrected frequency information may be utilized to improve the reconstructed image.