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:
Feb. 28, 2023

Filed:

Feb. 22, 2017
Applicant:

Waveone Inc., Mountain View, CA (US);

Inventors:

Oren Rippel, Mountain View, CA (US);

Lubomir Bourdev, Mountain View, CA (US);

Assignee:

WaveOne Inc., Palo Alto, CA (US);

Attorney:
Primary Examiner:
Int. Cl.
CPC ...
G06N 3/08 (2006.01); G06N 3/04 (2006.01); G06N 20/00 (2019.01); G06K 9/62 (2022.01); G06V 10/44 (2022.01); G06V 10/75 (2022.01); G06V 20/40 (2022.01); G06V 20/52 (2022.01); G06V 30/10 (2022.01); G06V 30/194 (2022.01); G06V 40/16 (2022.01); H04N 19/126 (2014.01); H04N 19/167 (2014.01); H04N 19/172 (2014.01); H04N 19/196 (2014.01); H04N 19/91 (2014.01); H04N 19/44 (2014.01); G06T 5/00 (2006.01); H04N 19/13 (2014.01); H04N 19/149 (2014.01); G06N 3/084 (2023.01); H04N 19/18 (2014.01); H04N 19/48 (2014.01); H04N 19/154 (2014.01); H04N 19/33 (2014.01);
U.S. Cl.
CPC ...
G06N 3/08 (2013.01); G06N 3/04 (2013.01); G06N 20/00 (2019.01); G06K 9/6232 (2013.01); G06K 9/6256 (2013.01); G06K 9/6263 (2013.01); G06K 9/6274 (2013.01); G06N 3/0454 (2013.01); G06N 3/084 (2013.01); G06T 5/002 (2013.01); G06V 10/449 (2022.01); G06V 10/454 (2022.01); G06V 10/758 (2022.01); G06V 20/46 (2022.01); G06V 20/52 (2022.01); G06V 30/10 (2022.01); G06V 30/194 (2022.01); G06V 40/172 (2022.01); H04N 19/126 (2014.11); H04N 19/13 (2014.11); H04N 19/149 (2014.11); H04N 19/154 (2014.11); H04N 19/167 (2014.11); H04N 19/172 (2014.11); H04N 19/18 (2014.11); H04N 19/197 (2014.11); H04N 19/33 (2014.11); H04N 19/44 (2014.11); H04N 19/48 (2014.11); H04N 19/91 (2014.11);
Abstract

A deep learning based compression (DLBC) system trains multiple models that, when deployed, generates a compressed binary encoding of an input image that achieves a reconstruction quality and a target compression ratio. The applied models effectively identifies structures of an input image, quantizes the input image to a target bit precision, and compresses the binary code of the input image via adaptive arithmetic coding to a target codelength. During training, the DLBC system reconstructs the input image from the compressed binary encoding and determines the loss in quality from the encoding process. Thus, the models can be continually trained to, when applied to an input image, minimize the loss in reconstruction quality that arises due to the encoding process while also achieving the target compression ratio.


Find Patent Forward Citations

Loading…