The patent badge is an abbreviated version of the USPTO patent document. The patent badge does contain a link to the full patent document.

The patent badge is an abbreviated version of the USPTO patent document. The patent badge covers the following: Patent number, Date patent was issued, Date patent was filed, Title of the patent, Applicant, Inventor, Assignee, Attorney firm, Primary examiner, Assistant examiner, CPCs, and Abstract. The patent badge does contain a link to the full patent document (in Adobe Acrobat format, aka pdf). To download or print any patent click here.

Date of Patent:
Aug. 14, 2018

Filed:

Feb. 21, 2017
Applicant:

A9.com, Inc., Palo Alto, CA (US);

Inventors:

Arnab Sanat Kumar Dhua, Mountain View, CA (US);

Ming Du, Fremont, CA (US);

Aishwarya Natesh, Mountain View, CA (US);

Assignee:

A9.com, Inc., Palo Alto, CA (US);

Attorney:
Primary Examiner:
Int. Cl.
CPC ...
G06K 9/60 (2006.01); G06K 9/66 (2006.01); G06K 9/46 (2006.01); G06K 9/62 (2006.01); G06T 11/60 (2006.01); G06T 7/194 (2017.01); G06T 7/90 (2017.01); G06T 7/73 (2017.01); G06N 3/08 (2006.01);
U.S. Cl.
CPC ...
G06K 9/66 (2013.01); G06K 9/4628 (2013.01); G06K 9/628 (2013.01); G06K 9/6256 (2013.01); G06N 3/08 (2013.01); G06T 7/194 (2017.01); G06T 7/73 (2017.01); G06T 7/90 (2017.01); G06T 11/60 (2013.01); G06K 2009/4666 (2013.01); G06T 2207/10024 (2013.01); G06T 2207/20036 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/20221 (2013.01);
Abstract

Training images can be synthesized in order to obtain enough data to train a convolutional neural network to recognize various classes of a type of item. Images can be synthesized by blending images of items labeled using those classes into selected background images. Catalog images can represent items against a solid background, which can be identified using connected components or other such approaches. Removing the background using such approaches can result in edge artifacts proximate the item region. To improve the results, one or more operations are performed, such as a morphological erosion operation followed by an opening operation. The isolated item portion then can be blended into a randomly selected background region in order to generate a synthesized training image. The training images can be used with real world images to train the neural network.


Find Patent Forward Citations

Loading…