Jian M Fang

US Patent:

20210374468, Dec 2, 2021

Filed:

May 26, 2021

Appl. No.:

17/330832

Inventors:

- Princeton NJ, US
Ting Wang - West Windsor NJ, US
Xiang Xu - Mountain View CA, US
Francesco Pittaluga - Los Angeles CA, US
Gaurav Sharma - Newark CA, US
Yi-Hsuan Tsai - Santa Clara CA, US
Masoud Faraki - San Jose CA, US
Yuheng Chen - South Brunswick NJ, US
Yue Tian - Princeton NJ, US
Ming-Fang Huang - Princeton NJ, US
Jian Fang - Princeton NJ, US

International Classification:

G06K 9/62
G06T 3/00

Abstract:

Methods and systems for training a neural network include generate an image of a mask. A copy of an image is generated from an original set of training data. The copy is altered to add the image of a mask to a face detected within the copy. An augmented set of training data is generated that includes the original set of training data and the altered copy. A neural network model is trained to recognize masked faces using the augmented set of training data.

US Patent:

20190234832, Aug 1, 2019

Filed:

Jan 26, 2019

Appl. No.:

16/258588

Inventors:

- Princeton NJ, US
Jian Fang - West Windsor NJ, US
Ting Wang - West Windsor NJ, US

Assignee:

NEC LABORATORIES AMERICA, INC - Princeton NJ

International Classification:

G01M 11/00
G01H 9/00

Abstract:

Aspects of the present disclosure describe a multilayer coding method for physical value measurements in which a multi-pulse pattern is generated through several layers, each of which is encoded with a different cyclic codeword. The total coding gain is the product of the gains of each layer advantageously allowing a long cyclic codeword to be replaced by combining several short cyclic codewords. Of particular advantage, methods according to the present disclosure provide better signal-to-noise characteristics, longer sensing distances and lower computational complexity particularly suitable for high-performance and cost-effective distributed optical fiber sensing (DOFS).