-
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
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
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
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:US20210063199A1
公开(公告)日:2021-03-04
申请号:US17008100
申请日:2020-08-31
Applicant: NVIDIA Corporation
Inventor: Amir Akbarzadeh , David Nister , Ruchi Bhargava , Birgit Henke , Ivana Stojanovic , Yu Sheng
Abstract: An end-to-end system for data generation, map creation using the generated data, and localization to the created map is disclosed. Mapstreams—or streams of sensor data, perception outputs from deep neural networks (DNNs), and/or relative trajectory data—corresponding to any number of drives by any number of vehicles may be generated and uploaded to the cloud. The mapstreams may be used to generate map data—and ultimately a fused high definition (HD) map—that represents data generated over a plurality of drives. When localizing to the fused HD map, individual localization results may be generated based on comparisons of real-time data from a sensor modality to map data corresponding to the same sensor modality. This process may be repeated for any number of sensor modalities and the results may be fused together to determine a final fused localization result.
-
公开(公告)号:US20240410981A1
公开(公告)日:2024-12-12
申请号:US18810728
申请日:2024-08-21
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 , G05D1/81 , 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:US11788861B2
公开(公告)日:2023-10-17
申请号:US17008074
申请日:2020-08-31
Applicant: NVIDIA Corporation
Inventor: David Nister , Ruchi Bhargava , Vaibhav Thukral , Michael Grabner , Ibrahim Eden , Jeffrey Liu
CPC classification number: G01C21/3841 , G01C21/1652 , G01C21/3811 , G01C21/3867 , G01C21/3878 , G01C21/3896 , G06N3/02
Abstract: An end-to-end system for data generation, map creation using the generated data, and localization to the created map is disclosed. Mapstreams—or streams of sensor data, perception outputs from deep neural networks (DNNs), and/or relative trajectory data—corresponding to any number of drives by any number of vehicles may be generated and uploaded to the cloud. The mapstreams may be used to generate map data—and ultimately a fused high definition (HD) map—that represents data generated over a plurality of drives. When localizing to the fused HD map, individual localization results may be generated based on comparisons of real-time data from a sensor modality to map data corresponding to the same sensor modality. This process may be repeated for any number of sensor modalities and the results may be fused together to determine a final fused localization result.
-
公开(公告)号:US20210156963A1
公开(公告)日:2021-05-27
申请号:US16836618
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
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:US20210150230A1
公开(公告)日:2021-05-20
申请号:US16915346
申请日:2020-06-29
Applicant: NVIDIA Corporation
Inventor: Nikolai Smolyanskiy , Ryan Oldja , Ke Chen , Alexander Popov , Joachim Pehserl , Ibrahim Eden , Tilman Wekel , David Wehr , Ruchi Bhargava , David Nister
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.
-
-
-
-
-
-
-
-
-