Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
IMAGE PROCESSING APPARATUS AND METHOD USING MULTI-CHANNEL FEATURE MAP
Document Type and Number:
WIPO Patent Application WO/2018/230832
Kind Code:
A1
Abstract:
A convolution neural network (CNN)-based image processing method and apparatus are provided. The CNN-based image processing method includes identifying whether values of pixels of each of feature maps having a plurality of channels at a first layer are zero, and storing information regarding a result of identifying whether the values of the pixels are zero; writing image feature information of the feature maps at the first layer to an external memory; reading information regarding pixels having values which are not zero among the written image feature information from the external memory based on the information regarding the result of identifying whether the values of the pixels are zero; and performing a feature map operation at a second layer using the read image feature information of the feature maps.

Inventors:
LEE WON-JAE (KR)
CHOI YONG-SEOK (KR)
KIM MIN-SOO (KR)
Application Number:
PCT/KR2018/004610
Publication Date:
December 20, 2018
Filing Date:
April 20, 2018
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
SAMSUNG ELECTRONICS CO LTD (KR)
International Classes:
G06T1/20; G06N3/02; G06T1/60; G06T5/00; G06T5/40
Domestic Patent References:
WO2016141282A12016-09-09
Foreign References:
EP3104309A22016-12-14
US20160358070A12016-12-08
US9524450B22016-12-20
US20170124432A12017-05-04
US20160358069A12016-12-08
US20160358070A12016-12-08
EP3104309A22016-12-14
Other References:
CHEN YU-HSIN ET AL.: "IEEE Journal of Solid-State Circuit", vol. 52, January 2017, IEEE SERVICE CENTER, article "Eyeriss: An Energy-Efficient Reconfigurable Accelerator for Deep Convolutional Neural Networks"
See also references of EP 3568828A4
Attorney, Agent or Firm:
Y.P.LEE, MOCK & PARTNERS (KR)
Download PDF: