David A Wehr

US Patent:

20210342608, Nov 4, 2021

Filed:

Jul 15, 2021

Appl. No.:

17/377053

Inventors:

- Santa Clara CA, US
Ryan Oldja - Redmond WA, US
Ke Chen - Sunnyvale CA, US
Alexander Popov - Kirkland WA, US
Joachim Pehserl - Lynnwood WA, US
Ibrahim Eden - Redmond WA, US
Tilman Wekel - Sunnyvale CA, US
David Wehr - Redmond WA, US
Ruchi Bhargava - Redmond WA, US
David Nister - Bellevue WA, US

International Classification:

G06K 9/00
G01S 17/931
B60W 60/00
G01S 17/89
G05D 1/00
G06N 3/04
G06T 19/00

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.

US Patent:

20210342609, Nov 4, 2021

Filed:

Jul 15, 2021

Appl. No.:

17/377064

Inventors:

- Santa Clara CA, US
Ryan Oldja - Redmond WA, US
Ke Chen - Sunnyvale CA, US
Alexander Popov - Kirkland WA, US
Joachim Pehserl - Lynnwood WA, US
Ibrahim Eden - Redmond WA, US
Tilman Wekel - Sunnyvale CA, US
David Wehr - Redmond WA, US
Ruchi Bhargava - Redmond WA, US
David Nister - Bellevue WA, US

International Classification:

G06K 9/00
G01S 17/931
B60W 60/00
G01S 17/89
G05D 1/00
G06N 3/04
G06T 19/00

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.

US Patent:

20210150230, May 20, 2021

Filed:

Jun 29, 2020

Appl. No.:

16/915346

Inventors:

- Santa Clara CA, US
Ryan Oldja - Redmond WA, US
Ke Chen - Sunnyvale CA, US
Alexander Popov - Kirkland WA, US
Joachim Pehserl - Lynnwood WA, US
Ibrahim Eden - Redmond WA, US
Tilman Wekel - Sunnyvale CA, US
David Wehr - Redmond WA, US
Ruchi Bhargava - Redmond WA, US
David Nister - Bellevue WA, US

International Classification:

G06K 9/00
B60W 60/00
G05D 1/00
G06N 3/04
G06T 19/00
G01S 17/89
G01S 17/931

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.