Jun Zhou

US Patent:

20220410841, Dec 29, 2022

Filed:

Aug 25, 2022

Appl. No.:

17/895144

Inventors:

- San Francisco CA, US
Robin Yang - San Francisco CA, US
Jiajian Zhou - San Francisco CA, US
Jackie Shannon - San Francisco CA, US
Jun Zhou - San Francisco CA, US
Divya Thakur - San Francisco CA, US
Alex Van Dorn - San Francisco CA, US

International Classification:

B60R 25/25
G06Q 10/02
G06V 40/16
G06Q 50/30

Abstract:

The subject disclosure relates to ways to authenticate a rider/user of an autonomous vehicle (AV) using biometric data, such as facial recognition. A process of the disclosed technology can facilitate the automatic unlocking of an AV by performing steps that include: receiving a dispatch request associated with a user identifier (ID), receiving a recognition model that corresponds with the user ID, and receiving an image stream including images of pedestrian faces. In some aspects, the process can further include steps for: providing the images to the recognition model, and determining if a user represented in the images corresponds with the user ID. Systems and machine-readable media are also provided.

US Patent:

20200034988, Jan 30, 2020

Filed:

Jul 29, 2019

Appl. No.:

16/524628

Inventors:

- Cayman Islands, KY
Jun Zhou - Mountain View CA, US

International Classification:

G06T 7/80
G06T 7/70
H04N 17/00
H04N 5/247
G01S 17/02
B60R 11/04

Abstract:

Provided herein are methods for calibrating a camera. The method may include capturing an image that includes at least a traffic sign. The location of the traffic sign using the image may then be determined, which may include taking as input the location and direction of the vehicle and the location and the direction of the camera relative to the vehicle. The method may also include obtaining an actual location of the traffic sign. The camera may be determined to require recalibration if the determined location is different from the actual location.

US Patent:

20180253647, Sep 6, 2018

Filed:

Mar 6, 2017

Appl. No.:

15/451345

Inventors:

- Sunnyvale CA, US
XIAO BO - Sunnyvale CA, US
JUN ZHOU - Sunnyvale CA, US
WEIDE ZHANG - Santa Clara CA, US
TONY HAN - Sunnyvale CA, US

International Classification:

G06N 3/08
G06N 3/04
G06F 17/16

Abstract:

In one embodiment, a system to accelerate batch-normalized convolutional neural network (CNN) models is disclosed. The system extracts a plurality of first groups of layers from a first CNN model, each group of the first groups having a first convolutional layer and a first batch-norm layer. For each group of the plurality of first groups, the system calculates a first scale vector and a first shift vector based on the first batch-norm layer, and generates a second convolutional layer representing the corresponding group of the plurality of first groups based on the first convolutional layer and the first scale and the first shift vectors. The system generates an accelerated CNN model based on the second convolutional layer corresponding to the plurality of the first groups, such that the accelerated CNN model is utilized subsequently to classify an object perceived by an autonomous driving vehicle (ADV).

US Patent:

20170257432, Sep 7, 2017

Filed:

May 22, 2017

Appl. No.:

15/601993

Inventors:

- San Jose CA, US
VENKAT NARAYAN SRINIVASAN - Fremont CA, US
JUN ZHOU - Mountain View CA, US

International Classification:

H04L 29/08
G06F 9/50

Abstract:

Embodiments disclosed facilitate distributed orchestration and deployment of a single instance of a distributed computing application over a plurality of clouds and container clusters, including container clusters provided through a Container as a Service (CaaS) offering. In some embodiments, system and pattern constructs associated with a hybrid distributed multi-tier application may be used to obtain an infrastructure independent representation of the distributed multi-tier application. The infrastructure independent representation may comprise a representation of an underlying pattern of resource utilization of the application. Further, the underlying pattern of resource utilization of the application may be neither cloud specific nor container cluster specific. In some embodiments, a single instance of the hybrid distributed multi-tier application may be deployed on a plurality of cloud infrastructures and on at least one container cluster, based, in part, on the cloud-infrastructure independent representation of the application.