Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
IMAGE PROCESSING APPARATUS AND OPERATION METHOD THEREOF
Document Type and Number:
WIPO Patent Application WO/2023/101276
Kind Code:
A1
Abstract:
An image processing apparatus for processing an image by using one or more convolution neural networks, according to an embodiment, may comprise a memory that stores one or more instructions, and at least one processor, by executing the one or more instructions stored in the memory: acquires first feature data by performing a convolution operation between input data, which has been acquired from a first image, and a first kernel; segments, into first groups, a plurality of channels included in the first feature data; acquires pieces of second feature data by performing a convolution operation between pieces of the first feature data respectively corresponding to the first groups and second kernels respectively corresponding to the first groups; acquires pieces of shuffling data by shuffling the pieces of second feature data; acquires output data by performing a convolution operation between data obtained by summing channels included in the pieces of shuffling data and a third kernel; and generates a second image on the basis of the output data.

Inventors:
AHN ILJUN (KR)
KANG SOOMIN (KR)
PARK JAEYEON (KR)
SONG YOUNGCHAN (KR)
SHIN HANUL (KR)
LEE TAMMY (KR)
Application Number:
PCT/KR2022/018204
Publication Date:
June 08, 2023
Filing Date:
November 17, 2022
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
SAMSUNG ELECTRONICS CO LTD (KR)
International Classes:
G06T3/40; G06T1/20; G06T5/00; G06T5/10; G06T5/20; G06V10/74
Foreign References:
KR20210140757A2021-11-23
JP2021103441A2021-07-15
KR20210019537A2021-02-22
KR102305470B12021-09-28
Other References:
XIANGYU ZHANG, ZHOU XINYU, LIN MENGXIAO, SUN JIAN: "ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices", ARXIV:1707.01083V2, 7 December 2017 (2017-12-07), XP055696820, Retrieved from the Internet [retrieved on 20200519], DOI: 10.1109/CVPR.2018.00716
Attorney, Agent or Firm:
Y.P.LEE, MOCK & PARTNERS (KR)
Download PDF: