Title:
IMAGE PROCESSING METHOD AND APPARATUS, AND COMPUTER DEVICE, PROGRAM AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2022/242448
Kind Code:
A1
Abstract:
An image processing method and apparatus, and a computer device, a program and a storage medium, which relate to the field of image processing. The method comprises: acquiring an original image sequence, wherein the original image sequence includes at least three original image frames (101); performing image preprocessing on the original image sequence, so as to obtain a feature map sequence and a confidence map sequence, which correspond to the original image sequence, wherein the feature map sequence is obtained by performing feature extraction on each original image frame, the confidence map sequence includes a confidence map corresponding to each original image frame, and the confidence map is used for representing the degree of confidence of each pixel point in the original image frame during a feature fusion process (102); performing feature fusion on the feature map sequence on the basis of the confidence map sequence, so as to obtain a target fused feature map, which corresponds to a target original image frame in the original image sequence (103); and reconstructing the target original image frame on the basis of the target fused feature map, so as to obtain a target reconstructed image frame (104). Pixel-level credibility supervision is performed on features during a feature fusion process, and the fusion of image features with high credibility is guided, such that the image quality of a reconstructed image is improved.
Inventors:
ZHANG XINYI (CN)
ZHU JUNWEI (CN)
CHU WENQING (CN)
TAI YING (CN)
WANG CHENGJIE (CN)
LI JILIN (CN)
HUANG FEIYUE (CN)
ZHU JUNWEI (CN)
CHU WENQING (CN)
TAI YING (CN)
WANG CHENGJIE (CN)
LI JILIN (CN)
HUANG FEIYUE (CN)
Application Number:
PCT/CN2022/090153
Publication Date:
November 24, 2022
Filing Date:
April 29, 2022
Export Citation:
Assignee:
TENCENT TECH SHENZHEN CO LTD (CN)
International Classes:
G06K9/00
Foreign References:
| CN112990171A | 2021-06-18 | |||
| CN110070511A | 2019-07-30 | |||
| CN104240186A | 2014-12-24 | |||
| CN111639719A | 2020-09-08 | |||
| CN111047516A | 2020-04-21 | |||
| US20200402243A1 | 2020-12-24 | |||
| US20060291750A1 | 2006-12-28 |
Other References:
WANG XINTAO; CHAN KELVIN C.K.; YU KE; DONG CHAO; LOY CHEN CHANGE: "EDVR: Video Restoration With Enhanced Deformable Convolutional Networks", 2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW), IEEE, 16 June 2019 (2019-06-16), pages 1954 - 1963, XP033747042, DOI: 10.1109/CVPRW.2019.00247
JAEYEON KANG; YOUNGHYUN JO; SEOUNG WUG OH; PETER VAJDA; SEON JOO KIM: "Deep Space-Time Video Upsampling Networks", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 10 August 2020 (2020-08-10), 201 Olin Library Cornell University Ithaca, NY 14853 , XP081735360
JAEYEON KANG; YOUNGHYUN JO; SEOUNG WUG OH; PETER VAJDA; SEON JOO KIM: "Deep Space-Time Video Upsampling Networks", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 10 August 2020 (2020-08-10), 201 Olin Library Cornell University Ithaca, NY 14853 , XP081735360
Attorney, Agent or Firm:
DEQI INTELLECTUAL PROPERTY LAW CORPORATION (CN)
Download PDF: