公开(公告)号：US20220351524A1

公开(公告)日：2022-11-03

申请号：US17864026

申请日：2022-07-13

Applicant: NVIDIA Corporation

Inventor： Trung Pham ,  Berta Rodriguez Hervas ,  Minwoo Park ,  David Nister ,  Neda Cvijetic

IPC: G06V20/56 ,  G06N3/04 ,  G05B13/02 ,  G06T5/00 ,  G06T3/40 ,  G06T7/11 ,  G06T11/20 ,  G06K9/62 ,  G06N3/08 ,  G06V30/262

Abstract: In various examples, live perception from sensors of a vehicle may be leveraged to detect and classify intersection contention areas in an environment of a vehicle in real-time or near real-time. For example, a deep neural network (DNN) may be trained to compute outputs—such as signed distance functions—that may correspond to locations of boundaries delineating intersection contention areas. The signed distance functions may be decoded and/or post-processed to determine instance segmentation masks representing locations and classifications of intersection areas or regions. The locations of the intersections areas or regions may be generated in image-space and converted to world-space coordinates to aid an autonomous or semi-autonomous vehicle in navigating intersections according to rules of the road, traffic priority considerations, and/or the like.

公开(公告)号：US12286115B2

公开(公告)日：2025-04-29

申请号：US17116138

申请日：2020-12-09

Applicant: NVIDIA Corporation

Inventor： Trung Pham ,  Berta Rodriguez Hervas ,  Minwoo Park ,  David Nister ,  Neda Cvijetic

IPC: B60W30/18 ,  G06N3/04 ,  G06N3/08 ,  G06T7/33 ,  G06V10/764 ,  G06V10/82 ,  G06V20/56 ,  G06V20/64 ,  G06N3/045

Abstract: In various examples, a three-dimensional (3D) intersection structure may be predicted using a deep neural network (DNN) based on processing two-dimensional (2D) input data. To train the DNN to accurately predict 3D intersection structures from 2D inputs, the DNN may be trained using a first loss function that compares 3D outputs of the DNN—after conversion to 2D space—to 2D ground truth data and a second loss function that analyzes the 3D predictions of the DNN in view of one or more geometric constraints—e.g., geometric knowledge of intersections may be used to penalize predictions of the DNN that do not align with known intersection and/or road structure geometries. As such, live perception of an autonomous or semi-autonomous vehicle may be used by the DNN to detect 3D locations of intersection structures from 2D inputs.

公开(公告)号：US12248319B2

公开(公告)日：2025-03-11

申请号：US18340255

申请日：2023-06-23

Applicant: NVIDIA Corporation

Inventor： Minwoo Park ,  Xiaolin Lin ,  Hae-Jong Seo ,  David Nister ,  Neda Cvijetic

IPC: G05D1/00 ,  G05D1/228 ,  G06F18/214 ,  G06F18/23 ,  G06F18/2411 ,  G06N3/04 ,  G06N3/045 ,  G06N3/08 ,  G06V10/14 ,  G06V10/44 ,  G06V10/48 ,  G06V10/75 ,  G06V10/764 ,  G06V10/766 ,  G06V10/776 ,  G06V10/82 ,  G06V10/94 ,  G06V20/56

Abstract: In various examples, systems and methods are disclosed that preserve rich spatial information from an input resolution of a machine learning model to regress on lines in an input image. The machine learning model may be trained to predict, in deployment, distances for each pixel of the input image at an input resolution to a line pixel determined to correspond to a line in the input image. The machine learning model may further be trained to predict angles and label classes of the line. An embedding algorithm may be used to train the machine learning model to predict clusters of line pixels that each correspond to a respective line in the input image. In deployment, the predictions of the machine learning model may be used as an aid for understanding the surrounding environment—e.g., for updating a world model—in a variety of autonomous machine applications.

公开(公告)号：US20240127454A1

公开(公告)日：2024-04-18

申请号：US18391276

申请日：2023-12-20

Applicant: NVIDIA Corporation

Inventor： Trung Pham ,  Berta Rodriguez Hervas ,  Minwoo Park ,  David Nister ,  Neda Cvijetic

IPC: G06T7/11 ,  G05B13/02 ,  G06F18/21 ,  G06F18/24 ,  G06N3/04 ,  G06N3/08 ,  G06T3/4046 ,  G06T5/70 ,  G06T11/20 ,  G06V10/26 ,  G06V10/34 ,  G06V10/44 ,  G06V10/82 ,  G06V20/56 ,  G06V30/19 ,  G06V30/262

CPC classification number: G06T7/11 ,  G05B13/027 ,  G06F18/21 ,  G06F18/24 ,  G06N3/04 ,  G06N3/08 ,  G06T3/4046 ,  G06T5/70 ,  G06T11/20 ,  G06V10/267 ,  G06V10/34 ,  G06V10/454 ,  G06V10/82 ,  G06V20/56 ,  G06V30/19173 ,  G06V30/274 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30252 ,  G06T2210/12

Abstract: In various examples, live perception from sensors of a vehicle may be leveraged to detect and classify intersection contention areas in an environment of a vehicle in real-time or near real-time. For example, a deep neural network (DNN) may be trained to compute outputs—such as signed distance functions—that may correspond to locations of boundaries delineating intersection contention areas. The signed distance functions may be decoded and/or post-processed to determine instance segmentation masks representing locations and classifications of intersection areas or regions. The locations of the intersections areas or regions may be generated in image-space and converted to world-space coordinates to aid an autonomous or semi-autonomous vehicle in navigating intersections according to rules of the road, traffic priority considerations, and/or the like.

公开(公告)号：US20240101118A1

公开(公告)日：2024-03-28

申请号：US18537527

申请日：2023-12-12

Applicant: NVIDIA Corporation

Inventor： Sayed Mehdi Sajjadi Mohammadabadi ,  Berta Rodriguez Hervas ,  Hang Dou ,  Igor Tryndin ,  David Nister ,  Minwoo Park ,  Neda Cvijetic ,  Junghyun Kwon ,  Trung Pham

IPC: B60W30/18 ,  B60W30/09 ,  B60W30/095 ,  B60W60/00 ,  G06N3/08 ,  G06V10/25 ,  G06V10/75 ,  G06V10/764 ,  G06V10/80 ,  G06V10/82 ,  G06V20/56 ,  G06V20/70 ,  G08G1/01

CPC classification number: B60W30/18154 ,  B60W30/09 ,  B60W30/095 ,  B60W60/0011 ,  G06N3/08 ,  G06V10/25 ,  G06V10/751 ,  G06V10/764 ,  G06V10/803 ,  G06V10/82 ,  G06V20/56 ,  G06V20/588 ,  G06V20/70 ,  G08G1/0125

Abstract: In various examples, live perception from sensors of a vehicle may be leveraged to detect and classify intersections in an environment of a vehicle in real-time or near real-time. For example, a deep neural network (DNN) may be trained to compute various outputs—such as bounding box coordinates for intersections, intersection coverage maps corresponding to the bounding boxes, intersection attributes, distances to intersections, and/or distance coverage maps associated with the intersections. The outputs may be decoded and/or post-processed to determine final locations of, distances to, and/or attributes of the detected intersections.

公开(公告)号：US11921502B2

公开(公告)日：2024-03-05

申请号：US18151012

申请日：2023-01-06

Applicant: NVIDIA Corporation

Inventor： Minwoo Park ,  Xiaolin Lin ,  Hae-Jong Seo ,  David Nister ,  Neda Cvijetic

IPC: G05D1/00 ,  G05D1/02 ,  G06F18/214 ,  G06F18/23 ,  G06F18/2411 ,  G06N3/04 ,  G06N3/08 ,  G06V10/44 ,  G06V10/48 ,  G06V10/75 ,  G06V10/764 ,  G06V10/766 ,  G06V10/776 ,  G06V10/82 ,  G06V10/94 ,  G06V20/56

CPC classification number: G05D1/0077 ,  G05D1/0088 ,  G06F18/2155 ,  G06F18/23 ,  G06F18/2411 ,  G06N3/0418 ,  G06V10/457 ,  G06V10/48 ,  G06V10/751 ,  G06V10/764 ,  G06V10/776 ,  G06V10/82 ,  G06V10/955 ,  G06V20/588 ,  G05D2201/0213

Abstract: In various examples, systems and methods are disclosed that preserve rich spatial information from an input resolution of a machine learning model to regress on lines in an input image. The machine learning model may be trained to predict, in deployment, distances for each pixel of the input image at an input resolution to a line pixel determined to correspond to a line in the input image. The machine learning model may further be trained to predict angles and label classes of the line. An embedding algorithm may be used to train the machine learning model to predict clusters of line pixels that each correspond to a respective line in the input image. In deployment, the predictions of the machine learning model may be used as an aid for understanding the surrounding environment—e.g., for updating a world model—in a variety of autonomous machine applications.

公开(公告)号：US11508049B2

公开(公告)日：2022-11-22

申请号：US16570187

申请日：2019-09-13

Applicant: NVIDIA Corporation

Inventor： Hae-Jong Seo ,  Abhishek Bajpayee ,  David Nister ,  Minwoo Park ,  Neda Cvijetic

IPC: G06K9/00 ,  G06T7/00 ,  G06K9/62 ,  G06N3/04 ,  G06N3/08 ,  G06V10/98

Abstract: In various examples, a deep neural network (DNN) is trained for sensor blindness detection using a region and context-based approach. Using sensor data, the DNN may compute locations of blindness or compromised visibility regions as well as associated blindness classifications and/or blindness attributes associated therewith. In addition, the DNN may predict a usability of each instance of the sensor data for performing one or more operations—such as operations associated with semi-autonomous or autonomous driving. The combination of the outputs of the DNN may be used to filter out instances of the sensor data—or to filter out portions of instances of the sensor data determined to be compromised—that may lead to inaccurate or ineffective results for the one or more operations of the system.

公开(公告)号：US20200293796A1

公开(公告)日：2020-09-17

申请号：US16814351

申请日：2020-03-10

Applicant: NVIDIA Corporation

Inventor： Sayed Mehdi Sajjadi Mohammadabadi ,  Berta Rodriguez Hervas ,  Hang Dou ,  Igor Tryndin ,  David Nister ,  Minwoo Park ,  Neda Cvijetic ,  Junghyun Kwon ,  Trung Pham

IPC: G06K9/00 ,  G06K9/62 ,  G06N3/08 ,  B60W30/09 ,  B60W30/095 ,  B60W60/00 ,  G08G1/01

Abstract: In various examples, live perception from sensors of a vehicle may be leveraged to detect and classify intersections in an environment of a vehicle in real-time or near real-time. For example, a deep neural network (DNN) may be trained to compute various outputs—such as bounding box coordinates for intersections, intersection coverage maps corresponding to the bounding boxes, intersection attributes, distances to intersections, and/or distance coverage maps associated with the intersections. The outputs may be decoded and/or post-processed to determine final locations of, distances to, and/or attributes of the detected intersections.

公开(公告)号：US20200249684A1

公开(公告)日：2020-08-06

申请号：US16781893

申请日：2020-02-04

Applicant: NVIDIA Corporation

Inventor： Davide Marco Onofrio ,  Hae-Jong Seo ,  David Nister ,  Minwoo Park ,  Neda Cvijetic

IPC: G05D1/02 ,  G06K9/00 ,  G06N3/08 ,  G05D1/00 ,  G06K9/62

Abstract: In various examples, a path perception ensemble is used to produce a more accurate and reliable understanding of a driving surface and/or a path there through. For example, an analysis of a plurality of path perception inputs provides testability and reliability for accurate and redundant lane mapping and/or path planning in real-time or near real-time. By incorporating a plurality of separate path perception computations, a means of metricizing path perception correctness, quality, and reliability is provided by analyzing whether and how much the individual path perception signals agree or disagree. By implementing this approach—where individual path perception inputs fail in almost independent ways—a system failure is less statistically likely. In addition, with diversity and redundancy in path perception, comfortable lane keeping on high curvature roads, under severe road conditions, and/or at complex intersections, as well as autonomous negotiation of turns at intersections, may be enabled.

公开(公告)号：US20200090322A1

公开(公告)日：2020-03-19

申请号：US16570187

申请日：2019-09-13

Applicant: NVIDIA Corporation

Inventor： Hae-Jong Seo ,  Abhishek Bajpayee ,  David Nister ,  Minwoo Park ,  Neda Cvijetic

IPC: G06T7/00 ,  G06N3/04 ,  G06N3/08 ,  G06K9/62 ,  G06K9/03

Abstract: In various examples, a deep neural network (DNN) is trained for sensor blindness detection using a region and context-based approach. Using sensor data, the DNN may compute locations of blindness or compromised visibility regions as well as associated blindness classifications and/or blindness attributes associated therewith. In addition, the DNN may predict a usability of each instance of the sensor data for performing one or more operations—such as operations associated with semi-autonomous or autonomous driving. The combination of the outputs of the DNN may be used to filter out instances of the sensor data—or to filter out portions of instances of the sensor data determined to be compromised—that may lead to inaccurate or ineffective results for the one or more operations of the system.