.png){kind=icon}
.png){kind=icon}
Return
Improved multi-scale dynamic feature encoding network for image demoireing
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patcog.2021.107970.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The popularity of smartphones with digital cameras makes photographing using smartphones an important daily activity. Moirepatterns can easily appear when shooting objects with rich textures, such as computer screens, and will severely degrade the image quality. Image demoireing is an important image restoration task that aims to remove moirepatterns and reveal the underlying clean image. Two key properties of moirepatterns-the widely distributed frequency spectrum and the dynamic nature of moiretextures-challenge the image demoireing task. In this paper, we propose an improved Multi-scale convolutional network with Dynamic feature encoding for image DeMoireing (MDDM+). We design two schemes in our network to respectively attack the broad frequency spectrum and the dynamic texture of moire: a multi-scale structure to process images at different spatial resolutions and a dynamic feature encoding module to encode the texture dynamically. To capture more moireand texture information from different frequencies, we further propose a novel L-1 wavelet loss used to train our model. Extensive experiments on two benchmarks show that our proposed image demoireing network can outperform the state of the arts in terms of fidelity as well as perception. (C) 2021ElsevierLtd. Allrightsreserved.
Keywords:
Image demoireing
Screen shot images
Moirepattern
Dynamic feature encoding
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
7.6
Papers:
1.3W
Citations:
4.5W
Organization
No organization information available