公开(公告)号：US12051206B2

公开(公告)日：2024-07-30

申请号：US16938706

申请日：2020-07-24

Applicant: NVIDIA Corporation

Inventor： Ke Chen ,  Nikolai Smolyanskiy ,  Alexey Kamenev ,  Ryan Oldja ,  Tilman Wekel ,  David Nister ,  Joachim Pehserl ,  Ibrahim Eden ,  Sangmin Oh ,  Ruchi Bhargava

IPC: G06T7/00 ,  G05D1/00 ,  G06F18/00 ,  G06F18/22 ,  G06F18/23 ,  G06T5/50 ,  G06T7/10 ,  G06T7/11 ,  G06V10/82 ,  G06V20/56 ,  G06V20/58 ,  G06V10/44

CPC classification number: G06T7/11 ,  G05D1/0088 ,  G06F18/22 ,  G06F18/23 ,  G06T5/50 ,  G06T7/10 ,  G06V10/82 ,  G06V20/56 ,  G06V20/58 ,  G06T2207/10028 ,  G06T2207/20084 ,  G06T2207/30252 ,  G06V10/454

Abstract: A deep neural network(s) (DNN) may be used to perform panoptic segmentation by performing pixel-level class and instance segmentation of a scene using a single pass of the DNN. Generally, one or more images and/or other sensor data may be stitched together, stacked, and/or combined, and fed into a DNN that includes a common trunk and several heads that predict different outputs. The DNN may include a class confidence head that predicts a confidence map representing pixels that belong to particular classes, an instance regression head that predicts object instance data for detected objects, an instance clustering head that predicts a confidence map of pixels that belong to particular instances, and/or a depth head that predicts range values. These outputs may be decoded to identify bounding shapes, class labels, instance labels, and/or range values for detected objects, and used to enable safe path planning and control of an autonomous vehicle.

公开(公告)号：US11885907B2

公开(公告)日：2024-01-30

申请号：US16836583

申请日：2020-03-31

Applicant: NVIDIA Corporation

Inventor： Alexander Popov ,  Nikolai Smolyanskiy ,  Ryan Oldja ,  Shane Murray ,  Tilman Wekel ,  David Nister ,  Joachim Pehserl ,  Ruchi Bhargava ,  Sangmin Oh

IPC: G01S7/295 ,  G06T7/246 ,  G06T7/73 ,  G01S7/41 ,  G01S13/931 ,  G06N3/08 ,  G06V10/764 ,  G06V10/82 ,  G06V20/58 ,  G06V20/64

CPC classification number: G01S7/2955 ,  G01S7/414 ,  G01S7/417 ,  G01S13/931 ,  G06N3/08 ,  G06T7/246 ,  G06T7/73 ,  G06V10/764 ,  G06V10/82 ,  G06V20/58 ,  G06V20/64 ,  G06T2207/10044 ,  G06T2207/20084 ,  G06T2207/30261

Abstract: In various examples, a deep neural network(s) (e.g., a convolutional neural network) may be trained to detect moving and stationary obstacles from RADAR data of a three dimensional (3D) space, in both highway and urban scenarios. RADAR detections may be accumulated, ego-motion-compensated, orthographically projected, and fed into a neural network(s). The neural network(s) may include a common trunk with a feature extractor and several heads that predict different outputs such as a class confidence head that predicts a confidence map and an instance regression head that predicts object instance data for detected objects. The outputs may be decoded, filtered, and/or clustered to form bounding shapes identifying the location, size, and/or orientation of detected object instances. The detected object instances may be provided to an autonomous vehicle drive stack to enable safe planning and control of the autonomous vehicle.

公开(公告)号：US20230049567A1

公开(公告)日：2023-02-16

申请号：US17976581

申请日：2022-10-28

Applicant: NVIDIA Corporation

Inventor： Alexander Popov ,  Nikolai Smolyanskiy ,  Ryan Oldja ,  Shane Murray ,  Tilman Wekel ,  David Nister ,  Joachim Pehserl ,  Ruchi Bhargava ,  Sangmin Oh

IPC: G01S7/41 ,  G01S13/86 ,  G06N3/04 ,  G06N3/08 ,  G01S13/89

Abstract: In various examples, a deep neural network(s) (e.g., a convolutional neural network) may be trained to detect moving and stationary obstacles from RADAR data of a three dimensional (3D) space. In some embodiments, ground truth training data for the neural network(s) may be generated from LIDAR data. More specifically, a scene may be observed with RADAR and LIDAR sensors to collect RADAR data and LIDAR data for a particular time slice. The RADAR data may be used for input training data, and the LIDAR data associated with the same or closest time slice as the RADAR data may be annotated with ground truth labels identifying objects to be detected. The LIDAR labels may be propagated to the RADAR data, and LIDAR labels containing less than some threshold number of RADAR detections may be omitted. The (remaining) LIDAR labels may be used to generate ground truth data.

公开(公告)号：US20220415059A1

公开(公告)日：2022-12-29

申请号：US17895940

申请日：2022-08-25

Applicant: NVIDIA Corporation

Inventor： Nikolai Smolyanskiy ,  Ryan Oldja ,  Ke Chen ,  Alexander Popov ,  Joachim Pehserl ,  Ibrahim Eden ,  Tilman Wekel ,  David Wehr ,  Ruchi Bhargava ,  David Nister

IPC: G06V20/58 ,  G01S17/931 ,  B60W60/00 ,  G01S17/89 ,  G05D1/00 ,  G06N3/04 ,  G06T19/00

Abstract: A deep neural network(s) (DNN) may be used to detect objects from sensor data of a three dimensional (3D) environment. For example, a multi-view perception DNN may include multiple constituent DNNs or stages chained together that sequentially process different views of the 3D environment. An example DNN may include a first stage that performs class segmentation in a first view (e.g., perspective view) and a second stage that performs class segmentation and/or regresses instance geometry in a second view (e.g., top-down). The DNN outputs may be processed to generate 2D and/or 3D bounding boxes and class labels for detected objects in the 3D environment. As such, the techniques described herein may be used to detect and classify animate objects and/or parts of an environment, and these detections and classifications may be provided to an autonomous vehicle drive stack to enable safe planning and control of the autonomous vehicle.

公开(公告)号：US20210342609A1

公开(公告)日：2021-11-04

申请号：US17377064

申请日：2021-07-15

Applicant: NVIDIA Corporation

Inventor： Nikolai Smolyanskiy ,  Ryan Oldja ,  Ke Chen ,  Alexander Popov ,  Joachim Pehserl ,  Ibrahim Eden ,  Tilman Wekel ,  David Wehr ,  Ruchi Bhargava ,  David Nister

IPC: G06K9/00 ,  G01S17/931 ,  B60W60/00 ,  G01S17/89 ,  G05D1/00 ,  G06N3/04 ,  G06T19/00

Abstract: A deep neural network(s) (DNN) may be used to detect objects from sensor data of a three dimensional (3D) environment. For example, a multi-view perception DNN may include multiple constituent DNNs or stages chained together that sequentially process different views of the 3D environment. An example DNN may include a first stage that performs class segmentation in a first view (e.g., perspective view) and a second stage that performs class segmentation and/or regresses instance geometry in a second view (e.g., top-down). The DNN outputs may be processed to generate 2D and/or 3D bounding boxes and class labels for detected objects in the 3D environment. As such, the techniques described herein may be used to detect and classify animate objects and/or parts of an environment, and these detections and classifications may be provided to an autonomous vehicle drive stack to enable safe planning and control of the autonomous vehicle.

公开(公告)号：US12164059B2

公开(公告)日：2024-12-10

申请号：US17377064

申请日：2021-07-15

Applicant: NVIDIA Corporation

Inventor： Nikolai Smolyanskiy ,  Ryan Oldja ,  Ke Chen ,  Alexander Popov ,  Joachim Pehserl ,  Ibrahim Eden ,  Tilman Wekel ,  David Wehr ,  Ruchi Bhargava ,  David Nister

IPC: G01S7/48 ,  B60W60/00 ,  G01S17/89 ,  G01S17/931 ,  G05D1/00 ,  G06N3/045 ,  G06T19/00 ,  G06V10/10 ,  G06V10/25 ,  G06V10/26 ,  G06V10/44 ,  G06V10/764 ,  G06V10/774 ,  G06V10/80 ,  G06V10/82 ,  G06V20/56 ,  G06V20/58

Abstract: A deep neural network(s) (DNN) may be used to detect objects from sensor data of a three dimensional (3D) environment. For example, a multi-view perception DNN may include multiple constituent DNNs or stages chained together that sequentially process different views of the 3D environment. An example DNN may include a first stage that performs class segmentation in a first view (e.g., perspective view) and a second stage that performs class segmentation and/or regresses instance geometry in a second view (e.g., top-down). The DNN outputs may be processed to generate 2D and/or 3D bounding boxes and class labels for detected objects in the 3D environment. As such, the techniques described herein may be used to detect and classify animate objects and/or parts of an environment, and these detections and classifications may be provided to an autonomous vehicle drive stack to enable safe planning and control of the autonomous vehicle.

公开(公告)号：US12080078B2

公开(公告)日：2024-09-03

申请号：US17895940

申请日：2022-08-25

Applicant: NVIDIA Corporation

Inventor： Nikolai Smolyanskiy ,  Ryan Oldja ,  Ke Chen ,  Alexander Popov ,  Joachim Pehserl ,  Ibrahim Eden ,  Tilman Wekel ,  David Wehr ,  Ruchi Bhargava ,  David Nister

IPC: G06V20/58 ,  B60W60/00 ,  G01S17/89 ,  G01S17/931 ,  G05D1/00 ,  G06N3/045 ,  G06T19/00

CPC classification number: G06V20/584 ,  B60W60/0011 ,  B60W60/0016 ,  B60W60/0027 ,  G01S17/89 ,  G01S17/931 ,  G05D1/0088 ,  G06N3/045 ,  G06T19/006 ,  G06V20/58 ,  B60W2420/403 ,  G06T2207/10028 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30261

Abstract: A deep neural network(s) (DNN) may be used to detect objects from sensor data of a three dimensional (3D) environment. For example, a multi-view perception DNN may include multiple constituent DNNs or stages chained together that sequentially process different views of the 3D environment. An example DNN may include a first stage that performs class segmentation in a first view (e.g., perspective view) and a second stage that performs class segmentation and/or regresses instance geometry in a second view (e.g., top-down). The DNN outputs may be processed to generate 2D and/or 3D bounding boxes and class labels for detected objects in the 3D environment. As such, the techniques described herein may be used to detect and classify animate objects and/or parts of an environment, and these detections and classifications may be provided to an autonomous vehicle drive stack to enable safe planning and control of the autonomous vehicle.

公开(公告)号：US12072443B2

公开(公告)日：2024-08-27

申请号：US17377053

申请日：2021-07-15

Applicant: NVIDIA Corporation

Inventor： Nikolai Smolyanskiy ,  Ryan Oldja ,  Ke Chen ,  Alexander Popov ,  Joachim Pehserl ,  Ibrahim Eden ,  Tilman Wekel ,  David Wehr ,  Ruchi Bhargava ,  David Nister

IPC: G01S7/48 ,  B60W60/00 ,  G01S17/89 ,  G01S17/931 ,  G05D1/00 ,  G06N3/045 ,  G06T19/00 ,  G06V10/25 ,  G06V10/26 ,  G06V10/44 ,  G06V10/764 ,  G06V10/774 ,  G06V10/80 ,  G06V10/82 ,  G06V20/56 ,  G06V20/58 ,  G06V10/10

CPC classification number: G01S7/4802 ,  B60W60/0011 ,  B60W60/0016 ,  B60W60/0027 ,  G01S17/89 ,  G01S17/931 ,  G05D1/0088 ,  G06N3/045 ,  G06T19/006 ,  G06V10/25 ,  G06V10/26 ,  G06V10/454 ,  G06V10/764 ,  G06V10/774 ,  G06V10/803 ,  G06V10/82 ,  G06V20/56 ,  G06V20/58 ,  G06V20/584 ,  B60W2420/403 ,  G06T2207/10028 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30261 ,  G06V10/16

Abstract: A deep neural network(s) (DNN) may be used to detect objects from sensor data of a three dimensional (3D) environment. For example, a multi-view perception DNN may include multiple constituent DNNs or stages chained together that sequentially process different views of the 3D environment. An example DNN may include a first stage that performs class segmentation in a first view (e.g., perspective view) and a second stage that performs class segmentation and/or regresses instance geometry in a second view (e.g., top-down). The DNN outputs may be processed to generate 2D and/or 3D bounding boxes and class labels for detected objects in the 3D environment. As such, the techniques described herein may be used to detect and classify animate objects and/or parts of an environment, and these detections and classifications may be provided to an autonomous vehicle drive stack to enable safe planning and control of the autonomous vehicle.

公开(公告)号：US20240273919A1

公开(公告)日：2024-08-15

申请号：US18647415

申请日：2024-04-26

Applicant: NVIDIA CORPORATION

Inventor： Nikolai Smolyanskiy ,  Ryan Oldja ,  Ke Chen ,  Alexander Popov ,  Joachim Pehserl ,  Ibrahim Eden ,  Tilman Wekel ,  David Wehr ,  Ruchi Bhargava ,  David Nister

IPC: G06V20/58 ,  B60W60/00 ,  G01S17/89 ,  G01S17/931 ,  G06N3/045 ,  G06T19/00

CPC classification number: G06V20/584 ,  B60W60/0011 ,  B60W60/0016 ,  B60W60/0027 ,  G01S17/89 ,  G01S17/931 ,  G06N3/045 ,  G06T19/006 ,  G06V20/58 ,  B60W2420/403 ,  G06T2207/10028 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30261

Abstract: A deep neural network(s) (DNN) may be used to detect objects from sensor data of a three dimensional (3D) environment. For example, a multi-view perception DNN may include multiple constituent DNNs or stages chained together that sequentially process different views of the 3D environment. An example DNN may include a first stage that performs class segmentation in a first view (e.g., perspective view) and a second stage that performs class segmentation and/or regresses instance geometry in a second view (e.g., top-down). The DNN outputs may be processed to generate 2D and/or 3D bounding boxes and class labels for detected objects in the 3D environment. As such, the techniques described herein may be used to detect and classify animate objects and/or parts of an environment, and these detections and classifications may be provided to an autonomous vehicle drive stack to enable safe planning and control of the autonomous vehicle.

公开(公告)号：US11960026B2

公开(公告)日：2024-04-16

申请号：US17976581

申请日：2022-10-28

Applicant: NVIDIA Corporation

Inventor： Alexander Popov ,  Nikolai Smolyanskiy ,  Ryan Oldja ,  Shane Murray ,  Tilman Wekel ,  David Nister ,  Joachim Pehserl ,  Ruchi Bhargava ,  Sangmin Oh

IPC: G01S7/41 ,  G01S13/86 ,  G01S13/89 ,  G06N3/04 ,  G06N3/08

CPC classification number: G01S7/417 ,  G01S13/865 ,  G01S13/89 ,  G06N3/04 ,  G06N3/08

Abstract: In various examples, a deep neural network(s) (e.g., a convolutional neural network) may be trained to detect moving and stationary obstacles from RADAR data of a three dimensional (3D) space. In some embodiments, ground truth training data for the neural network(s) may be generated from LIDAR data. More specifically, a scene may be observed with RADAR and LIDAR sensors to collect RADAR data and LIDAR data for a particular time slice. The RADAR data may be used for input training data, and the LIDAR data associated with the same or closest time slice as the RADAR data may be annotated with ground truth labels identifying objects to be detected. The LIDAR labels may be propagated to the RADAR data, and LIDAR labels containing less than some threshold number of RADAR detections may be omitted. The (remaining) LIDAR labels may be used to generate ground truth data.