arrow
返回

Multi-Task Image Restoration Algorithm Under Different Weather Influence Factors

delete2024-01-01
delete0
delete
OA
AI
L
Lingwen Meng *
M
Mingyong Xin
J
Junwei Zhang
F
Fu Zou
Q
Qingqing Zhao
DOI:10.1109/ACCESS.2024.3520484delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Bad weather, such as rain, snow, and fog, will reduce the quality of the image acquired, and it will affect the performance of many related visual fields. The existing researches on image restoration under severe weather either focus on the restoration under a certain kind of weather, which cannot be generalized under different weather, or need to introduce additional model structures, which increases the burden of practical applications. Therefore, we explores an integrated image restoration framework, which can restore images under different adverse weather effects. Specifically, the model is trained in two stages. In the pre-training stage, a basic general network that can deal with multiple weather is trained by supervised learning. In the fine-tuning stage, soft prompts are introduced to stimulate the model's ability to cope with different weather, so as to enhance the generalization ability. Specifically, the hidden information of the prompts is explored by low-rank decomposition, and the contras loss is added to prompt the prompts to converge on similar tasks. Furthermore, we address the distribution shift problem by aligning out-of-distribution (OOD) test sample statistics with those of the source data using test-time prompt tuning. Finally, we use the evaluation metrics, PSNR and SSIM, to evaluate the proposed method under four tasks: rain removal, snow removal, fog removal, and raindrop removal. The results demonstrate that the proposed method achieves superior performance compared to the existing state-of-the-art.
Keyword:
Transformer architecture
pretrain-finetuning
soft prompts
contrast learning
distribution alignment strategy
Transformer architecture
pretrain-finetuning
soft prompts
contrast learning
distribution alignment strategy

期刊

IEEE Access 封面图
IEEE Access
IF:
3.6
论文数:
9.8W
被引数:
29.4W

机构

C
China Southern Power Grid
学者数:
3.4K
论文数: 2.4K
被引数: 8
引用论文

引用论文

Splenic Doppler Impedance Indices Estimate Splenic Congestion in Patients With Right-Sided or Congestive Heart Failure
err2012-01-01
err0
PREAI
errMassimo Bolognesi; Cristina Quaglio; Giancarlo Bombonato; Silvia Gaiani; Paola Pesce; Paola Bizzotto; Enrico Favaretto; Angelo Gatta; David Sacerdoti
err分享
err收藏
Emerging Adults’ Psychopathology in Seven Countries: The Impact of Identity-Related Risk Factors
err2018-09-18
err0
PREAI
errMalte Persike; Inge Seiffge-Krenke; Figen Cok; Karolina Głogowska; Vassilis Pavlopoulos; Spyridon Tantaros; Cyrille Perchec; Iffat Rohail; Juan Carlos Saravia
err分享
err收藏
Deep Dense Multi-Scale Network for Snow Removal Using Semantic and Depth Priors
err2021-01-01
err71
errOAAI
errZhang, Kaihao; Li, Rongqing; Yu, Yanjiang; Luo, Wenhan; Li, Changsheng
err分享
err收藏
DesnowNet: Context-Aware Deep Network for Snow Removal
err2018-06-01
err167
errOAAI
errLiu, Yun-Fu; Jaw, Da-Wei; Huang, Shih-Chia; Hwang, Jenq-Neng
err分享
err收藏
Methionine imbalance and toxicity in calves.
err2000-01-01
err0
PREAI
errM Abe; H Okada; D Matsumura; H Sato; M Funaba; T Iriki
err分享
err收藏
The Elliptic Law
err2014-10-15
err0
PREAI
errHoi H. Nguyen; Sean O’Rourke
err分享
err收藏
学者 查看更多内容