An Li Lin

US Patent:

20090206564, Aug 20, 2009

Filed:

Jul 22, 2008

Appl. No.:

12/177781

Inventors:

An Hao Lin - San Mateo CA, US

International Classification:

A63C 5/03
A63C 5/075

US Classification:

280 1422, 280 25

Abstract:

A snow glider comprises a longitudinally-extending runner for supporting a rider on ridden snow and having a generally central, longitudinally-extending slot. The glider further comprises an elevated rider-supporting deck and chatter-absorbing means straddling the slot for supporting the deck above the runner. Preferably, the bottom surface of the runner is preferably generally convex across its width, with curved outer edges that provide directional control to the rider when turning, and at least one longitudinally-extending, generally straight longitudinally-extending inner edge adjacent the periphery of the slot for greater directional control when the rider is going straight. To enhance controllability, the deck of the preferred embodiment is preferably coupled to the runner near the runner's outer periphery so that the force exerted by the rider is transmitted from his/her feet directly to the outer edges of the runner. To further reduce frictional contact with the snow, the bottom surface of the runner is preferably dimpled.

US Patent:

20190065861, Feb 28, 2019

Filed:

Aug 22, 2018

Appl. No.:

16/109533

Inventors:

- Pittsburgh PA, US
An Pang Lin - Mountain View CA, US
Shreyas Venugopalan - Pittsburgh PA, US
Ajmal Thanikkal - Pittsburgh PA, US
Gavriel Adler - Brookline MA, US
Kyle Neblett - Brookline MA, US
Christopher Palmer - Pittsburgh PA, US

International Classification:

G06K 9/00
G06F 17/30

Abstract:

A method for processing arrival or removal of packages within the field of view of a video camera includes providing a database for recording packages placed in the field of view. Based on real-time analysis of successive image frames in the camera, a human person's entry and exit from the field of view of the camera is also detected. Delivery or removal of objects is recorded in the database. In one embodiment, the method also determines whether or not a newly arrived package is placed alongside or on top of an existing package.

US Patent:

20190057588, Feb 21, 2019

Filed:

Aug 17, 2018

Appl. No.:

16/104668

Inventors:

- Pittsburgh PA, US
An Pang Lin - Mountain View CA, US
Shreyas Venugopalan - Pittsburgh PA, US
Ajmal Thanikkal - Pittsburgh PA, US
Gavriel Adler - Brookline MA, US
Kyle Neblett - Brookline MA, US

International Classification:

G08B 13/196
G06K 9/00
G06T 7/11
G06F 15/18

Abstract:

A video monitoring method that involves determining motion changes in a set of video frames to find potential objects is described. One or more bounding boxes are defined around the potential objects. These bounding boxes are spatially and temporally filtered to eliminate potential object candidates, with only potential objects in the bounding boxes remaining after filtering being classified or identified.

US Patent:

20180053091, Feb 22, 2018

Filed:

Aug 17, 2017

Appl. No.:

15/679926

Inventors:

- Fremont CA, US
An Pang LIN - Mountain View CA, US
Shreyas VENUGOPALAN - Pittsburgh PA, US
Ajmal THANIKKAL - Pittsburgh PA, US
John MATTY - Gibsonia PA, US
Gavriel ADLER - Brookline MA, US
Kyle NEBLETT - Seattle WA, US

Assignee:

HAWXEYE, INC. - Fremont CA

International Classification:

G06N 3/08
G06N 3/04

Abstract:

Embodiments of the present disclosure include a non-transitory computer-readable medium with computer-executable instructions stored thereon executed by one or more processors to perform a method to select and implement a neural network for an embedded system. The method includes selecting a neural network from a library of neural networks based on one or more parameters of the embedded system, the one or more parameters constraining the selection of the neural network. The method also includes training the neural network using a dataset. The method further includes compressing the neural network for implementation on the embedded system, wherein compressing the neural network comprises adjusting at least one float of the neural network.