Nutanix, Inc.

Nanometrics Inc.

Toyota Research Institute, Inc.

Leland Stanford Junior University

Onto Innovation Inc.

Toyota Jidosha Kabushiki Kaisha

Jie Li

Aroosh Sohi

Akshay Deodhar

Nikhil Bhatia

Silvio J Rabello

Arpit Singh

Basavaraj Kaladagi

Adrien David Gaidon

Vitor Guizilini

Rares A Ambrus

Leonidas John Guibas

William A McGahan

Or Litany

Jingsheng Shi

Karan Gupta

Kuan-Hui Lee

Amit Jain

Ranjan Parthasarathy

Kun-Hsin Chen

Vinod Gupta

Shunsho Kaku

Jeffrey Michael Walls

Pavan Kumar Konka

Pedro Vagos

Zhuan Liu

Nigel P Smith

Aman Nijhawan

Steven A Parkison

Harry Hai Yang

Anil Kumar Kushwah

Paul Michael Digioia

Krishna Ganti

Yongdong Liu

Wei Ming Chiew

Yiliang Liu

Avinash Manjaya Shetty

Petar Žuvela

Yang Xiao

William A Mcgahan

A method for 3D object detection is described. The method includes concurrently training a monocular depth network and a 3D object detection network. The method also includes predicting, using a trained monocular depth network, a monocular depth map of a monocular image of a video stream. The method further includes inferring a 3D point cloud of a 3D object within the monocular image according to the predicted monocular depth map. The method also includes predicting 3D bounding boxes from a selection of 3D points from the 3D point cloud of the 3D object based on a selection regression loss.

Monocular object detection via end-to-end differentiable pipeline

Optical measurement of a sample that includes a structure-of-interest (SOI) optically coupled to an unknown structure is optically measured by extracting from the resulting spectral signal the spectral variation that is correlated to key parameters associated with the SOI and removing the spectral variation from unknown structure that is irrelevant to the key parameters. An offline process is used to generate a physics-based model from a number of calibration measurements using reconstructed spectral signals after removing the spectral variation from the spectral signals that is irrelevant to the key parameters. A machine learning model may be additionally generated using at least the spectral variations correlated to key parameters associated with the SOI. In an in-line process, a sample is measured by filtering the spectral signals from a sample to remove spectral effects from the unknown structure and using the physics-based model or using the trained machine learning model.

Mitigation of undesired spectral effects in optical metrology

A method for 3D object detection is described. The method includes predicting, using a trained monocular depth network, an estimated monocular input depth map of a monocular image of a video stream and an estimated depth uncertainty map associated with the estimated monocular input depth map. The method also includes feeding back a depth uncertainty regression loss associated with the estimated monocular input depth map during training of the trained monocular depth network to update the estimated monocular input depth map. The method further includes detecting 3D objects from a 3D point cloud computed from the estimated monocular input depth map based on seed positions selected from the 3D point cloud and the estimated depth uncertainty map. The method also includes selecting 3D bounding boxes of the 3D objects detected from the 3D point cloud based on the seed positions and an aggregated depth uncertainty.

Systems and methods for uncertainty aware monocular 3D object detection

A method for an end-to-end boundary lane detection system is described. The method includes gridding a red-green-blue (RGB) image captured by a camera sensor mounted on an ego vehicle into a plurality of image patches. The method also includes generating different image patch embeddings to provide correlations between the plurality of image patches and the RGB image. The method further includes encoding the different image patch embeddings into predetermined categories, grid offsets, and instance identifications. The method also includes generating lane boundary keypoints of the RGB image based on the encoding of the different image patch embeddings.

End-to-end learned lane boundary detection based on a transformer

Contamination of metrology data by non-target signals caused by the measurement spot being incident on target and non-target areas may be reduced or eliminated using local gradient based mixed modeling or a machine learning model. The local gradient based mixed modeling approach uses a mixed model based on a model of the target region and a term of the local gradient of measured signals, which may be determined from scan data. The machine learning approach obtains mixed metrology data from a plurality of locations with respect to the target. The mixed metrology data is a mixture of target signals from the target and non-target signals from a non-target area and is used by a trained machine learning model to determine target metrology data that includes target signals from the target without the non-target signals and is used to determine parameters of interest for the target.

Measurements of structures in presence of signal contaminations

In some aspects, a non-transitory computer-readable storage medium having computer-executable instructions stored thereon that, upon execution by a processor, causes the processor to receive an indicator of a functionality; map the indicator to a first service and a second service dependent on the first service; identify a policy based on the first service and the second service; and determine, based on the policy, locations of deployment for the first service and the second service.

System and method for intent-based service deployment

Examples of services described herein expose an application programming interface (API) which may return the run-time configuration information. In this manner, software external to the service (e.g., an orchestrator) may query the service to determine the run-time configuration information, then provide both the initial configuration information known to the external software and the run-time configuration information to an Internet-facing gateway. Examples described herein may accordingly avoid or reduce instances of an upgrade dependency. When the service is upgraded, the run-time configuration information may be obtained by calling the API without a need to also upgrade the software external to the service.

Systems including interfaces for communication of run-time configuration information

A system may include a service deployment system configured to receive user indicators, map the user indicators to a plurality of services, determine an order of deployment associated with the plurality of services, and determine deployment locations associated with the plurality of services. The service deployment system may be configured to generate a plurality of actions, responsive to the plurality of services, the order of deployment, and the deployment locations.

A method and apparatus for data driven and cluster specific version/update control. The apparatus includes an automated multi-clusters management apparatus that interfaces with a plurality of remote clusters to provide data driven version/update control on a cluster by cluster basis. Generally, operation includes collection/identification of cluster specific data pertaining to software, hardware, and cluster requirements. The cluster specific data is later compared/analyzed against multi-cluster data pertaining to software releases, hardware characteristics, and known bugs/issues for each. The results of the comparison/analysis can then be ranked according to various metrics to different possible solutions and to differentiate the less desirable results from the more desirable results. Thus, the automated multi-cluster management apparatus provides for selection of versions/updates that is dependent on the cluster specific data. Additionally, the present disclosure provides for scheduling and distribution planning for selected versions/updates.

Method and apparatus for data driven and cluster specific version/update control

A plurality of overlay errors in a structure is determined using a target that includes a plurality of diffraction based overlay pads. Each diffraction based overlay pad has the same number of periodic patterns as the structure under test. Additionally, each diffraction based overlay pad includes a programmed shift between each pair of periodic patterns. The pads are illuminated and the resulting light is detected and used to simultaneously determine the plurality of overlay errors in the structure based on the programmed shifts. The overlay errors may be determined using a subset of elements of the Mueller matrix or by using the resulting spectra from the pads.

Simultaneous measurement of multiple overlay errors using diffraction based overlay

An empirical diffraction based overlay (eDBO) measurement of an overlay error is produced using diffraction signals from a plurality of diffraction based alignment pads from an alignment target. The linearity of the overlay error is tested using the same diffraction signals or a different set of diffraction signals from diffraction based alignment pads. Wavelengths that do not have a linear response to overlay error may be excluded from the measurement error.

Diffraction based overlay linearity testing

Asymmetry metrology is performed using at least a portion of Mueller matrix elements, including, e.g., the off-diagonal elements of the Mueller matrix. The Mueller matrix may be generated using, e.g., a spectroscopic or angle resolved ellipsometer that may include a rotating compensator. The Mueller matrix is analyzed by fitting at least a portion of the elements to Mueller matrix elements calculated using a rigorous electromagnetic model of the sample or by fitting the off-diagonal elements to a calibrated linear response. The use of the Mueller matrix elements in the asymmetry measurement permits, e.g., overlay analysis using in-chip devices thereby avoiding the need for special off-chip targets.

Scatterometry measurement of asymmetric structures

The process of modeling a complex two-dimensional periodic structure is improved by selectively truncating the Fourier expansion used in the calculation of resulting scatter signature from the model. The Fourier expansion is selectively truncated by determining the contribution for each harmonic order in the Fourier transform of the permittivity function and retaining the harmonic orders with a contribution that is above a threshold. The Fourier space may be compressed so that only the selected harmonic orders are used, thereby reducing the required memory and calculation times. The compressed Fourier space may be used in a real-time analysis of a sample or to generate a library that is used in the analysis of a sample.

Simulating two-dimensional periodic patterns using compressed fourier space

Growing community of inventors

San Jose, CA, United States of America

Jie Li