LOOK-UP CONVOLUTIONAL LAYER IN CONVOLUTIONAL NEURAL NETWORK

Title:

LOOK-UP CONVOLUTIONAL LAYER IN CONVOLUTIONAL NEURAL NETWORK

Document Type and Number:

WIPO Patent Application WO/2017/206156

Kind Code:

A1

Abstract:

Embodiments provide for a processor including logic to accelerate convolutional neural network processing, the processor including first logic to apply a convolutional layer to an image to generate a first convolution result and second logic to apply a look-up convolutional layer to the first convolution result to generate a second convolution result, the second convolution result associated with a location of the first convolution result within a global filter kernel.

Inventors:

MA LIWEI (CN)
SONG JIQIANG (CN)

Application Number:

PCT/CN2016/084621

Publication Date:

December 07, 2017

Filing Date:

June 03, 2016

Export Citation:

Click for automatic bibliography generation Help

Assignee:

INTEL CORP (US)
MA LIWEI (CN)
SONG JIQIANG (CN)

International Classes:

G06K9/66

Foreign References:

US20160148078A1	2016-05-26
CN104573731A	2015-04-29
CN104616032A	2015-05-13
US20070086655A1	2007-04-19

Other References:

JI, SHUIWANG ET AL.: "3D Convolutional Neural Networks for Human Action Recognition", IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, vol. 35, no. 1, 1 January 2013 (2013-01-01), pages 221 - 231, XP011490774, Retrieved from the Internet
ALEX KENDALL ET AL.: "PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization", ARXIV, 1 December 2015 (2015-12-01), pages 2938 - 2946, XP055568190, DOI: 10.1109/ICCV.2015.336

Attorney, Agent or Firm:

NTD PATENT AND TRADEMARK AGENCY LIMITED (CN)

Download PDF:

View/Download PDF PDF Help

Previous Patent: WEAVING METHOD FOR NASAL SINUS STENT AND STENT OBTAINED THEREOF

Next Patent: SYSTEMS, METHODS, AND COMPUTER READABLE MEDIA FOR ENCHANCED SIMULATION OF DRILLING DYNAMICS