公开(公告)号：US20240273926A1

公开(公告)日：2024-08-15

申请号：US18647261

申请日：2024-04-26

Applicant: NVIDIA CORPORATION

Inventor： Kang Wang ,  Yue Wu ,  Minwoo Park ,  Gang Pan

IPC: G06V20/64 ,  B60G17/0165 ,  B60K31/00 ,  B60W60/00 ,  G01S17/89 ,  G01S17/931 ,  G06F18/214 ,  G06V20/58

CPC classification number: G06V20/64 ,  G01S17/89 ,  G01S17/931 ,  G06F18/214 ,  G06V20/58 ,  B60G17/0165 ,  B60K31/00 ,  B60W60/001 ,  B60W2420/408

Abstract: In various examples, to support training a deep neural network (DNN) to predict a dense representation of a 3D surface structure of interest, a training dataset is generated from real-world data. For example, one or more vehicles may collect image data and LiDAR data while navigating through a real-world environment. To generate input training data, 3D surface structure estimation may be performed on captured image data to generate a sparse representation of a 3D surface structure of interest (e.g., a 3D road surface). To generate corresponding ground truth training data, captured LiDAR data may be smoothed, subject to outlier removal, subject to triangulation to filling missing values, accumulated from multiple LiDAR sensors, aligned with corresponding frames of image data, and/or annotated to identify 3D points on the 3D surface of interest, and the identified 3D points may be projected to generate a dense representation of the 3D surface structure.

公开(公告)号：US12008822B2

公开(公告)日：2024-06-11

申请号：US17452752

申请日：2021-10-28

Applicant: NVIDIA Corporation

Inventor： Kang Wang ,  Yue Wu ,  Minwoo Park ,  Gang Pan

IPC: G06V20/64 ,  G01S17/89 ,  G01S17/931 ,  G06F18/214 ,  G06V20/58 ,  B60G17/0165 ,  B60K31/00 ,  B60W60/00

CPC classification number: G06V20/64 ,  G01S17/89 ,  G01S17/931 ,  G06F18/214 ,  G06V20/58 ,  B60G17/0165 ,  B60K31/00 ,  B60W60/001 ,  B60W2420/408

Abstract: In various examples, to support training a deep neural network (DNN) to predict a dense representation of a 3D surface structure of interest, a training dataset is generated from real-world data. For example, one or more vehicles may collect image data and LiDAR data while navigating through a real-world environment. To generate input training data, 3D surface structure estimation may be performed on captured image data to generate a sparse representation of a 3D surface structure of interest (e.g., a 3D road surface). To generate corresponding ground truth training data, captured LiDAR data may be smoothed, subject to outlier removal, subject to triangulation to filling missing values, accumulated from multiple LiDAR sensors, aligned with corresponding frames of image data, and/or annotated to identify 3D points on the 3D surface of interest, and the identified 3D points may be projected to generate a dense representation of the 3D surface structure.

公开(公告)号：US11967022B2

公开(公告)日：2024-04-23

申请号：US17452751

申请日：2021-10-28

Applicant: NVIDIA Corporation

Inventor： Kang Wang ,  Yue Wu ,  Minwoo Park ,  Gang Pan

IPC: G06T17/05 ,  B60W30/09 ,  B60W30/14 ,  B60W40/06 ,  B60W50/06 ,  B60W60/00 ,  G06F18/214 ,  G06V20/05 ,  G06V20/58

CPC classification number: G06T17/05 ,  B60W30/09 ,  B60W30/143 ,  B60W40/06 ,  B60W50/06 ,  B60W60/001 ,  G06F18/214 ,  G06V20/05 ,  B60W2420/42 ,  B60W2420/52 ,  B60W2552/15

Abstract: In various examples, to support training a deep neural network (DNN) to predict a dense representation of a 3D surface structure of interest, a training dataset is generated using a parametric mathematical modeling. A variety of synthetic 3D road surfaces may be generated by modeling a 3D road surface using varied parameters to simulate changes in road direction and lateral surface slope. In an example embodiment, a synthetic 3D road surface may be created by modeling a longitudinal 3D curve and expanding the longitudinal 3D curve to a 3D surface, and the resulting synthetic 3D surface may be sampled to form a synthetic ground truth projection image (e.g., a 2D height map). To generate corresponding input training data, a known pattern that represents which pixels may remain unobserved during 3D structure estimation may be generated and applied to a ground truth projection image to simulate a corresponding sparse projection image.

公开(公告)号：US20250091607A1

公开(公告)日：2025-03-20

申请号：US18971085

申请日：2024-12-06

Applicant: NVIDIA CORPORATION

Inventor： Kang Wang ,  Yue Wu ,  Minwoo Park ,  Gang Pan

IPC: B60W60/00 ,  B60W40/09 ,  G06N3/02

Abstract: In various examples, a 3D surface structure such as the 3D surface structure of a road (3D road surface) may be observed and estimated to generate a 3D point cloud or other representation of the 3D surface structure. Since the estimated representation may be sparse, a deep neural network (DNN) may be used to predict values for a dense representation of the 3D surface structure from the sparse representation. For example, a sparse 3D point cloud may be projected to form a sparse projection image (e.g., a sparse 2D height map), which may be fed into the DNN to predict a dense projection image (e.g., a dense 2D height map). The predicted dense representation of the 3D surface structure may be provided to an autonomous vehicle drive stack to enable safe and comfortable planning and control of the autonomous vehicle.

公开(公告)号：US12145617B2

公开(公告)日：2024-11-19

申请号：US17452744

申请日：2021-10-28

Applicant: NVIDIA Corporation

Inventor： Kang Wang ,  Yue Wu ,  Minwoo Park ,  Gang Pan

IPC: B60W60/00 ,  B60W40/06 ,  B60W40/105 ,  G06V20/58

Abstract: In various examples, a 3D surface structure such as the 3D surface structure of a road (3D road surface) may be observed and estimated to generate a 3D point cloud or other representation of the 3D surface structure. Since the representation may be sparse, one or more densification techniques may be applied to densify the representation of the 3D surface structure. For example, the relationship between sparse and dense projection images (e.g., 2D height maps) may be modeled with a Markov random field, and Maximum a Posterior (MAP) inference may be performed using a corresponding joint probability distribution to estimate the most likely dense values given the sparse values. The resulting dense representation of the 3D surface structure may be provided to an autonomous vehicle drive stack to enable safe and comfortable planning and control of the autonomous vehicle.

公开(公告)号：US11922571B2

公开(公告)日：2024-03-05

申请号：US17452751

申请日：2021-10-28

Applicant: NVIDIA Corporation

Inventor： Kang Wang ,  Yue Wu ,  Minwoo Park ,  Gang Pan

IPC: G06T17/05 ,  B60W30/09 ,  B60W30/14 ,  B60W40/06 ,  B60W50/06 ,  B60W60/00 ,  G06F18/214 ,  G06V20/05 ,  G06V20/58

CPC classification number: G06T17/05 ,  B60W30/09 ,  B60W30/143 ,  B60W40/06 ,  B60W50/06 ,  B60W60/001 ,  G06F18/214 ,  G06V20/05 ,  B60W2420/42 ,  B60W2420/52 ,  B60W2552/15

Abstract: In various examples, to support training a deep neural network (DNN) to predict a dense representation of a 3D surface structure of interest, a training dataset is generated using a parametric mathematical modeling. A variety of synthetic 3D road surfaces may be generated by modeling a 3D road surface using varied parameters to simulate changes in road direction and lateral surface slope. In an example embodiment, a synthetic 3D road surface may be created by modeling a longitudinal 3D curve and expanding the longitudinal 3D curve to a 3D surface, and the resulting synthetic 3D surface may be sampled to form a synthetic ground truth projection image (e.g., a 2D height map). To generate corresponding input training data, a known pattern that represents which pixels may remain unobserved during 3D structure estimation may be generated and applied to a ground truth projection image to simulate a corresponding sparse projection image.