Asymmetric Mask Scheme for Self-Supervised Real Image Denoising

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liao, Xiangyu, Zheng, Tianheng, Zhong, Jiayu, Zhang, Pingping, Ren, Chao
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916321994735616
author Liao, Xiangyu
Zheng, Tianheng
Zhong, Jiayu
Zhang, Pingping
Ren, Chao
author_facet Liao, Xiangyu
Zheng, Tianheng
Zhong, Jiayu
Zhang, Pingping
Ren, Chao
contents In recent years, self-supervised denoising methods have gained significant success and become critically important in the field of image restoration. Among them, the blind spot network based methods are the most typical type and have attracted the attentions of a large number of researchers. Although the introduction of blind spot operations can prevent identity mapping from noise to noise, it imposes stringent requirements on the receptive fields in the network design, thereby limiting overall performance. To address this challenge, we propose a single mask scheme for self-supervised denoising training, which eliminates the need for blind spot operation and thereby removes constraints on the network structure design. Furthermore, to achieve denoising across entire image during inference, we propose a multi-mask scheme. Our method, featuring the asymmetric mask scheme in training and inference, achieves state-of-the-art performance on existing real noisy image datasets. All the source code will be made available to the public.
format Preprint
id arxiv_https___arxiv_org_abs_2407_06514
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Asymmetric Mask Scheme for Self-Supervised Real Image Denoising
Liao, Xiangyu
Zheng, Tianheng
Zhong, Jiayu
Zhang, Pingping
Ren, Chao
Image and Video Processing
Computer Vision and Pattern Recognition
In recent years, self-supervised denoising methods have gained significant success and become critically important in the field of image restoration. Among them, the blind spot network based methods are the most typical type and have attracted the attentions of a large number of researchers. Although the introduction of blind spot operations can prevent identity mapping from noise to noise, it imposes stringent requirements on the receptive fields in the network design, thereby limiting overall performance. To address this challenge, we propose a single mask scheme for self-supervised denoising training, which eliminates the need for blind spot operation and thereby removes constraints on the network structure design. Furthermore, to achieve denoising across entire image during inference, we propose a multi-mask scheme. Our method, featuring the asymmetric mask scheme in training and inference, achieves state-of-the-art performance on existing real noisy image datasets. All the source code will be made available to the public.
title Asymmetric Mask Scheme for Self-Supervised Real Image Denoising
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.06514