Jie Chen

US Patent:

20100246203, Sep 30, 2010

Filed:

Mar 27, 2009

Appl. No.:

12/412709

Inventors:

Jie Chen - Novi MI, US
Ben Wang - Northville MI, US
Luc Guy Louis Lacroix - West Bloomfield MI, US
Amine Taleb-Bendiab - Ann Arbor MI, US

Assignee:

NORTH AMERICAN LIGHTING, INC. - Farmington Hills MI

International Classification:

B60Q 1/00

US Classification:

362538

Abstract:

A system and method for exterior projection lighting of a vehicle include a generally ellipsoidal faceted reflector disposed on a rear side of a projection lens. First and second focal points of each facet define corresponding reflector focal regions. A semi-conducting light source positioned at an acute angle, such as a single light emitting diode (LED) without a primary optic, emits light toward the reflector from within the first reflector focal region. Reflected light passes through the projection lens to illuminate a desired beam pattern. A curved shade disposed between the projection lens and the light source has at least a portion of a top edge disposed within the second reflector focal region. The shade blocks a portion of reflected light from extending below the optical axis.

US Patent:

20210081673, Mar 18, 2021

Filed:

Sep 9, 2020

Appl. No.:

17/016260

Inventors:

- Princeton NJ, US
Asim KADAV - Jersey City NJ, US
Jie CHEN - Bellevue WA, US

Assignee:

NEC LABORATORIES AMERICA, INC - Princeton NJ

International Classification:

G06K 9/00

Abstract:

Aspects of the present disclosure describe systems, methods, and structures that provide action recognition with high-order interaction with spatio-temporal object tracking. Image and object features are organized into into tracks, which advantageously facilitates many possible learnable embeddings and intra/inter-track interaction(s). Operationally, our systems, method, and structures according to the present disclosure employ an efficient high-order interaction model to learn embeddings and intra/inter object track interaction across the space and time for AR. Each frame is detected by an object detector to locate visual objects. Those objects are linked through time to form object tracks. The object tracks are then organized and combined with the embeddings as the input to our model. The model is trained to generate representative embeddings and discriminative video features through high-order interaction which is formulated as an efficient matrix operation without iterative processing delay.