公开(公告)号：US20190266418A1

公开(公告)日：2019-08-29

申请号：US16286329

申请日：2019-02-26

Applicant: NVIDIA Corporation

Inventor： Yifang Xu ,  Xin Liu ,  Chia-Chih Chen ,  Carolina Parada ,  Davide Onofrio ,  Minwoo Park ,  Mehdi Sajjadi Mohammadabadi ,  Vijay Chintalapudi ,  Ozan Tonkal ,  John Zedlewski ,  Pekka Janis ,  Jan Nikolaus Fritsch ,  Gordon Grigor ,  Zuoguan Wang ,  I-Kuei Chen ,  Miguel Sainz

IPC: G06K9/00 ,  G06K9/32 ,  G05D1/02 ,  G05D1/00 ,  G06N3/08 ,  G06T7/10

Abstract: In various examples, sensor data representative of an image of a field of view of a vehicle sensor may be received and the sensor data may be applied to a machine learning model. The machine learning model may compute a segmentation mask representative of portions of the image corresponding to lane markings of the driving surface of the vehicle. Analysis of the segmentation mask may be performed to determine lane marking types, and lane boundaries may be generated by performing curve fitting on the lane markings corresponding to each of the lane marking types. The data representative of the lane boundaries may then be sent to a component of the vehicle for use in navigating the vehicle through the driving surface.

公开(公告)号：US20250104186A1

公开(公告)日：2025-03-27

申请号：US17514448

申请日：2021-10-29

Applicant: Nvidia Corporation

Inventor： Loudon Cohen ,  Gregory Massal ,  Pekka Janis

IPC: G06T3/40 ,  G06N3/04 ,  G06T5/00 ,  G06T5/50

Abstract: Apparatuses, systems, and techniques are presented to reconstruct one or more images. In at least one embodiment, one or more neural networks are used to upsample one or more images based, at least in part, on one or more brightness values.

公开(公告)号：US20240403640A1

公开(公告)日：2024-12-05

申请号：US18799229

申请日：2024-08-09

Applicant: NVIDIA Corporation

Inventor： Yilin Yang ,  Bala Siva Sashank Jujjavarapu ,  Pekka Janis ,  Zhaoting Ye ,  Sangmin Oh ,  Minwoo Park ,  Daniel Herrera Castro ,  Tommi Koivisto ,  David Nister

IPC: G06N3/08 ,  B60W30/14 ,  B60W60/00 ,  G06F18/214 ,  G06V10/762 ,  G06V20/56

Abstract: In various examples, a deep neural network (DNN) is trained to accurately predict, in deployment, distances to objects and obstacles using image data alone. The DNN may be trained with ground truth data that is generated and encoded using sensor data from any number of depth predicting sensors, such as, without limitation, RADAR sensors, LIDAR sensors, and/or SONAR sensors. Camera adaptation algorithms may be used in various embodiments to adapt the DNN for use with image data generated by cameras with varying parameters—such as varying fields of view. In some examples, a post-processing safety bounds operation may be executed on the predictions of the DNN to ensure that the predictions fall within a safety-permissible range.

公开(公告)号：US12072442B2

公开(公告)日：2024-08-27

申请号：US17456045

申请日：2021-11-22

Applicant: NVIDIA Corporation

Inventor： Tommi Koivisto ,  Pekka Janis ,  Tero Kuosmanen ,  Timo Roman ,  Sriya Sarathy ,  William Zhang ,  Nizar Assaf ,  Colin Tracey

IPC: G06V10/46 ,  B60W50/00 ,  G01S7/41 ,  G05D1/00 ,  G06F16/35 ,  G06F18/21 ,  G06F18/214 ,  G06F18/23 ,  G06F18/2413 ,  G06N3/044 ,  G06N3/045 ,  G06N3/084 ,  G06N20/00 ,  G06V10/20 ,  G06V10/44 ,  G06V10/762 ,  G06V10/764 ,  G06V10/77 ,  G06V10/774 ,  G06V20/58 ,  G01S7/48 ,  G01S13/86 ,  G01S13/931 ,  G01S17/931 ,  G06N3/047 ,  G06N3/048

CPC classification number: G01S7/417 ,  B60W50/00 ,  G05D1/0246 ,  G06F16/35 ,  G06F18/214 ,  G06F18/217 ,  G06F18/23 ,  G06F18/2414 ,  G06N3/044 ,  G06N3/045 ,  G06N3/084 ,  G06N20/00 ,  G06V10/255 ,  G06V10/454 ,  G06V10/46 ,  G06V10/762 ,  G06V10/764 ,  G06V10/7715 ,  G06V10/774 ,  G06V20/58 ,  G06V20/584 ,  G01S7/412 ,  G01S7/4802 ,  G01S13/867 ,  G01S2013/9318 ,  G01S2013/9323 ,  G01S17/931 ,  G06N3/047 ,  G06N3/048

Abstract: In various examples, detected object data representative of locations of detected objects in a field of view may be determined. One or more clusters of the detected objects may be generated based at least in part on the locations and features of the cluster may be determined for use as inputs to a machine learning model(s). A confidence score, computed by the machine learning model(s) based at least in part on the inputs, may be received, where the confidence score may be representative of a probability that the cluster corresponds to an object depicted at least partially in the field of view. Further examples provide approaches for determining ground truth data for training object detectors, such as for determining coverage values for ground truth objects using associated shapes, and for determining soft coverage values for ground truth objects.

公开(公告)号：US11790230B2

公开(公告)日：2023-10-17

申请号：US17723195

申请日：2022-04-18

Applicant: NVIDIA Corporation

Inventor： Yilin Yang ,  Bala Siva Sashank Jujjavarapu ,  Pekka Janis ,  Zhaoting Ye ,  Sangmin Oh ,  Minwoo Park ,  Daniel Herrera Castro ,  Tommi Koivisto ,  David Nister

IPC: G06K9/00 ,  G06N3/08 ,  B60W30/14 ,  B60W60/00 ,  G06V20/56 ,  G06F18/214 ,  G06V10/762

CPC classification number: G06N3/08 ,  B60W30/14 ,  B60W60/0011 ,  G06F18/2155 ,  G06V10/763 ,  G06V20/56

Abstract: In various examples, a deep neural network (DNN) is trained to accurately predict, in deployment, distances to objects and obstacles using image data alone. The DNN may be trained with ground truth data that is generated and encoded using sensor data from any number of depth predicting sensors, such as, without limitation, RADAR sensors, LIDAR sensors, and/or SONAR sensors. Camera adaptation algorithms may be used in various embodiments to adapt the DNN for use with image data generated by cameras with varying parameters—such as varying fields of view. In some examples, a post-processing safety bounds operation may be executed on the predictions of the DNN to ensure that the predictions fall within a safety-permissible range.

公开(公告)号：US20230267701A1

公开(公告)日：2023-08-24

申请号：US18309882

申请日：2023-05-01

Applicant: NVIDIA Corporation

Inventor： Yifang Xu ,  Xin Liu ,  Chia-Chin Chen ,  Carolina Parada ,  Davide Onofrio ,  Minwoo Park ,  Mehdi Sajjadi Mohammadabadi ,  Vijay Chintalapudi ,  Ozan Tonkal ,  John Zedlewski ,  Pekka Janis ,  Jan Nikolaus Fritsch ,  Gordon Grigor ,  Zuoguan Wang ,  I-Kuei Chen ,  Miguel Sainz

IPC: G06V10/44 ,  G06T7/10 ,  G05D1/00 ,  G06N3/084 ,  G05D1/02 ,  G06V20/56 ,  G06V10/46 ,  G06V20/40 ,  G06F18/2413 ,  G06V10/764 ,  G06V10/82

CPC classification number: G06V10/44 ,  G06T7/10 ,  G05D1/0088 ,  G06N3/084 ,  G05D1/0221 ,  G06V20/588 ,  G06V10/46 ,  G06V10/457 ,  G06V20/41 ,  G06F18/24143 ,  G06V10/764 ,  G06V10/82 ,  G06V10/471

Abstract: In various examples, sensor data representative of an image of a field of view of a vehicle sensor may be received and the sensor data may be applied to a machine learning model. The machine learning model may compute a segmentation mask representative of portions of the image corresponding to lane markings of the driving surface of the vehicle. Analysis of the segmentation mask may be performed to determine lane marking types, and lane boundaries may be generated by performing curve fitting on the lane markings corresponding to each of the lane marking types. The data representative of the lane boundaries may then be sent to a component of the vehicle for use in navigating the vehicle through the driving surface.

公开(公告)号：US20230122119A1

公开(公告)日：2023-04-20

申请号：US18067176

申请日：2022-12-16

Applicant: NVIDIA Corporation

Inventor： Yue Wu ,  Pekka Janis ,  Xin Tong ,  Cheng-Chieh Yang ,  Minwoo Park ,  David Nister

IPC: G05D1/02 ,  G06N3/04 ,  G06T7/20 ,  G05D1/00 ,  G06V20/10 ,  G06V20/58 ,  G06F18/214

Abstract: In various examples, a sequential deep neural network (DNN) may be trained using ground truth data generated by correlating (e.g., by cross-sensor fusion) sensor data with image data representative of a sequences of images. In deployment, the sequential DNN may leverage the sensor correlation to compute various predictions using image data alone. The predictions may include velocities, in world space, of objects in fields of view of an ego-vehicle, current and future locations of the objects in image space, and/or a time-to-collision (TTC) between the objects and the ego-vehicle. These predictions may be used as part of a perception system for understanding and reacting to a current physical environment of the ego-vehicle.

公开(公告)号：US20220222778A1

公开(公告)日：2022-07-14

申请号：US17710643

申请日：2022-03-31

Applicant: NVIDIA Corporation

Inventor： Shiqiu Liu ,  Robert Thomas Pottorff ,  Guilin Liu ,  Karan Sapra ,  Jon Barker ,  David Tarjan ,  Pekka Janis ,  Edvard Olav Valter Fagerholm ,  Lei Yang ,  Kevin Jonathan Shih ,  Marco Salvi ,  Timo Roman ,  Andrew Tao ,  Bryan Christopher Catanzaro

IPC: G06T3/40 ,  G06T5/20 ,  G06T5/50 ,  G06T5/00

Abstract: Apparatuses, systems, and techniques are presented to generate images. In at least one embodiment, one or more neural networks are used to generate one or more images using one or more pixel weights determined based, at least in part, on one or more sub-pixel offset values.

公开(公告)号：US11182916B2

公开(公告)日：2021-11-23

申请号：US16728598

申请日：2019-12-27

Applicant: NVIDIA Corporation

Inventor： Yilin Yang ,  Bala Siva Sashank Jujjavarapu ,  Pekka Janis ,  Zhaoting Ye ,  Sangmin Oh ,  Minwoo Park ,  Daniel Herrera Castro ,  Tommi Koivisto ,  David Nister

IPC: G06K9/00 ,  G06T7/536

Abstract: In various examples, a deep neural network (DNN) is trained to accurately predict, in deployment, distances to objects and obstacles using image data alone. The DNN may be trained with ground truth data that is generated and encoded using sensor data from any number of depth predicting sensors, such as, without limitation, RADAR sensors, LIDAR sensors, and/or SONAR sensors. Camera adaptation algorithms may be used in various embodiments to adapt the DNN for use with image data generated by cameras with varying parameters—such as varying fields of view. In some examples, a post-processing safety bounds operation may be executed on the predictions of the DNN to ensure that the predictions fall within a safety-permissible range.

公开(公告)号：US20190258878A1

公开(公告)日：2019-08-22

申请号：US16277895

申请日：2019-02-15

Applicant: NVIDIA Corporation

Inventor： Tommi Koivisto ,  Pekka Janis ,  Tero Kuosmanen ,  Timo Roman ,  Sriya Sarathy ,  William Zhang ,  Nizar Assaf ,  Colin Tracey

IPC: G06K9/00 ,  G06K9/46 ,  G06K9/62

Abstract: In various examples, detected object data representative of locations of detected objects in a field of view may be determined. One or more clusters of the detected objects may be generated based at least in part on the locations and features of the cluster may be determined for use as inputs to a machine learning model(s). A confidence score, computed by the machine learning model(s) based at least in part on the inputs, may be received, where the confidence score may be representative of a probability that the cluster corresponds to an object depicted at least partially in the field of view. Further examples provide approaches for determining ground truth data for training object detectors, such as for determining coverage values for ground truth objects using associated shapes, and for determining soft coverage values for ground truth objects.