Abstract: In some embodiments, a computer-implemented method for simulating performance of a physical device is provided. Calculating a current time step of an operational simulation of the physical device includes, for each voxel of a simulated environment, concurrently with loading a set of field values for the voxel for a previous time step from a main memory, determining permittivity values for the voxel using feature parameter values. The computing system calculates a set of field values for the voxel for the current time step based on the set of field values for the voxel for the previous time step and the permittivity values.

Abstract: A system for optimizing physical designs provides integrated optimization of design geometry, design materials, and design subassemblies by mapping a catalog of actual or available construction materials and subassemblies to a differentiable representation tractable for computerized optimization. New subassemblies may be generated by using the differential representation in conjunction with a decoder trained on the actual or available subassemblies.

Abstract: In some embodiments, a computer-implemented method for designing a physical device is provided. A computing system generates an initial design based on a design specification. The initial design includes a list of features, and each feature of the list of features represents a convex shape. The computing system determines a set of signed distance fields that includes a signed distance field for each feature of the list of features, and determines a set of structural parameters using the set of signed distance fields. The computing system simulates performance of the initial design using the set of structural parameters to determine a performance loss value. The computing system determines at least one fabrication loss value using the set of signed distance fields. The computing system updates at least one feature of the list of features using the at least one fabrication loss value and a gradient of the performance loss value.

Abstract: In some embodiments, a computer-implemented method for designing a physical device is provided. A computing system determines whether a feature from a list of features is present in a set of structural parameters by, in response to determining whether a feature presence function indicates that the feature should be included in the set of structural parameters or not, updating the set of structural parameters to include the feature or refraining from updating the set of structural parameters to include the feature, accordingly. The computing system simulates performance of the initial design using the set of structural parameters to determine a performance loss value, determines a structural gradient based on the performance loss value, determines a feature gradient based on the performance loss value, and updates the features in the list of features based on the structural gradient and the feature gradient.

Abstract: In some embodiments, a computer-implemented method for designing a physical device is provided. A computing system generates an initial design that includes an input waveguide starting at an input location and extending to a end position, an output waveguide starting at a start position and extending to an output location, and a dispersive region. The computing system determines a set of structural parameters based on the initial design. The computing system simulates performance of the initial design using the set of structural parameters to determine a performance loss value based on at least one performance goal. The computing system updates at least one of the end position of the input waveguide, the start position of the output waveguide, or a size of the dispersive region in the initial design using a gradient of the performance loss value.

Abstract: In some embodiments, a computer-implemented method for designing a physical device is provided. A computing system generates an initial design based on a design specification. The initial design includes a list of geometric shape primitives. The computing system determines a set of structural parameters using the list of geometric shape primitives. The computing system simulates performance of the initial design using the set of structural parameters to determine a performance loss value. The computing system updates at least one of a size or a location of at least one of the geometric shape primitives using a gradient of the performance loss value.

Abstract: A multi-process model to support compiling applications for multiple platforms is described. In one embodiment, applications designed for execution on a mobile platform can be ported to and/or compiled for execution on a desktop/laptop platform without requiring modification of the core program code of the mobile application. The mobile application is executed using a multi-process (e.g., two or more process) model in which the core mobile application program generates content that is displayed by a host process. The host process enables automatic translation of program calls to generate mobile user interface elements into program calls that generate user interface elements of the host platform. The translation can be performed using a multi-process (e.g., two or more process) model in which the core application program generates content that is displayed by a host process.

Abstract: A computer-implemented physics solver operates within a neural network framework. A neural network accepts a coordinate for each location within a design domain and outputs a local composition for each location. A model of the design domain is formed in a physics solver. The model includes discrete elements that encompass the design domain. A solution of the model provides a value of a design objective. For a plurality of iterations, the following is performed: the neural network determines current local compositions for the locations in the design domain corresponding to the discrete elements; the current local compositions are input into the discrete elements of the physics solver to obtain a current value of the design objective; and the current value of the design objective is used to find a loss gradient of a loss function. The loss gradient is used to update the neural network during the iterations.

Abstract: A system and method may track data from one or more sensors during vehicle driving. Based on the sensors data, one or more alerts or potential hazards may be identified. The system and method may generate a drive summary including information and optional statistics about the alerts or potential hazards.

Abstract: A system and method may generate a more realistic augmented reality (AR) overlay by generating a segmentation image and blending it with one or more other images. The system may generate a segmentation image based on an input image. The segmentation image may be blended with an AR path overlay image to generate an object-masked AR path overlay image. The object-masked AR path overlay image may be blended with the input image to generate an output image.

Abstract: Systems and methods for a more usable Augmented Reality (AR) display of navigation indications is described. A live camera image of a scene may be captured from a device. Navigation instructions may be generated from a navigation system and a navigation indication may be generated for display. A computer vision-based positioning algorithm may be performed on the camera image to determine the relative position between the viewpoint of the device and one or more landmarks in the live camera image. The location or shape of the visual display of the navigation indication may be determined based on the computer vision-based positioning algorithm.

Abstract: A system and method may generate a more realistic augmented reality (AR) overlay by generating a segmentation image and blending it with one or more other images. The system may generate a segmentation image based on an input image. The segmentation image may be blended with an AR path overlay image to generate an object-masked AR path overlay image. The object-masked AR path overlay image may be blended with the input image to generate an output image.

Abstract: A multi-process model to support compiling applications for multiple platforms is described. In one embodiment, applications designed for execution on a mobile platform can be ported to and/or compiled for execution on a desktop/laptop platform without requiring modification of the core program code of the mobile application. The mobile application is executed using a multi-process (e.g., two or more process) model in which the core mobile application program generates content that is displayed by a host process. The host process enables automatic translation of program calls to generate mobile user interface elements into program calls that generate user interface elements of the host platform. The translation can be performed using a multi-process (e.g., two or more process) model in which the core application program generates content that is displayed by a host process.

Abstract: Systems and methods for a more usable Augmented Reality (AR) display of navigation indications is described. A live camera image of a scene may be captured from a device. Navigation instructions may be generated from a navigation system and a navigation indication may be generated for display. A computer vision-based positioning algorithm may be performed on the camera image to determine the relative position between the viewpoint of the device and one or more landmarks in the live camera image. The location or shape of the visual display of the navigation indication may be determined based on the computer vision-based positioning algorithm.

Abstract: Embodiments described herein provide for a multi-process model to support compiling applications for multiple platforms. In one embodiment, applications designed for execution on a mobile platform can be ported to and/or compiled for execution on a desktop/laptop platform without requiring modification of the core program code of the mobile application. The mobile application is executed using a multi-process (e.g., two or more process) model in which the core mobile application program generates content that is displayed by a host process. The host process enables automatic translation of program calls to generate mobile user interface elements into program calls that generate user interface elements of the host platform. The translation can be performed using a multi-process (e.g., two or more process) model in which the core application program generates content that is displayed by a host process.

Abstract: Embodiments described herein provide for a multi-process model to support compiling applications for multiple platforms. In one embodiment, applications designed for execution on a mobile platform can be ported to and/or compiled for execution on a desktop/laptop platform without requiring modification of the core program code of the mobile application. The mobile application is executed using a multi-process (e.g., two or more process) model in which the core mobile application program generates content that is displayed by a host process. The host process enables automatic translation of program calls to generate mobile user interface elements into program calls that generate user interface elements of the host platform. The translation can be performed using a multi-process (e.g., two or more process) model in which the core application program generates content that is displayed by a host process.

Abstract: Systems and methods for a more usable Augmented Reality (AR) display of navigation indications is described. A live camera image of a scene may be captured from a device. Navigation instructions may be generated from a navigation system and a navigation indication may be generated for display. A computer vision-based positioning algorithm may be performed on the camera image to determine the relative position between the viewpoint of the device and one or more landmarks in the live camera image. The location or shape of the visual display of the navigation indication may be determined based on the computer vision-based positioning algorithm.

Abstract: Embodiments described herein provide for a multi-process model to support compiling applications for multiple platforms. In one embodiment, applications designed for execution on a mobile platform can be ported to and/or compiled for execution on a desktop/laptop platform without requiring modification of the core program code of the mobile application. The mobile application is executed using a multi-process (e.g., two or more process) model in which the core mobile application program generates content that is displayed by a host process. The host process enables automatic translation of program calls to generate mobile user interface elements into program calls that generate user interface elements of the host platform. The translation can be performed using a multi-process (e.g., two or more process) model in which the core application program generates content that is displayed by a host process.

Abstract: A system and method may generate a more realistic augmented reality (AR) overlay by generating a segmentation image and blending it with one or more other images. The system may generate a segmentation image based on an input image. The segmentation image may be blended with an AR path overlay image to generate an object-masked AR path overlay image. The object-masked AR path overlay image may be blended with the input image to generate an output image.

Abstract: A system and method may track data from one or more sensors during vehicle driving. Based on the sensors data, one or more alerts or potential hazards may be identified. The system and method may generate a drive summary including information and optional statistics about the alerts or potential hazards.