Guardado en:
Detalles Bibliográficos
Autores principales: Wen, Yang, Lai, Anyu, Qian, Bo, Wang, Hao, Shi, Wuzhen, Cao, Wenming
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2409.06334
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866929494257827840
author Wen, Yang
Lai, Anyu
Qian, Bo
Wang, Hao
Shi, Wuzhen
Cao, Wenming
author_facet Wen, Yang
Lai, Anyu
Qian, Bo
Wang, Hao
Shi, Wuzhen
Cao, Wenming
contents Currently, the mainstream restoration tasks under adverse weather conditions have predominantly focused on single-weather scenarios. However, in reality, multiple weather conditions always coexist and their degree of mixing is usually unknown. Under such complex and diverse weather conditions, single-weather restoration models struggle to meet practical demands. This is particularly critical in fields such as autonomous driving, where there is an urgent need for a model capable of effectively handling mixed weather conditions and enhancing image quality in an automated manner. In this paper, we propose a Task Sequence Generator module that, in conjunction with the Task Intra-patch Block, effectively extracts task-specific features embedded in degraded images. The Task Intra-patch Block introduces an external learnable sequence that aids the network in capturing task-specific information. Additionally, we employ a histogram-based transformer module as the backbone of our network, enabling the capture of both global and local dynamic range features. Our proposed model achieves state-of-the-art performance on public datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2409_06334
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Multi-Weather Image Restoration via Histogram-Based Transformer Feature Enhancement
Wen, Yang
Lai, Anyu
Qian, Bo
Wang, Hao
Shi, Wuzhen
Cao, Wenming
Computer Vision and Pattern Recognition
Currently, the mainstream restoration tasks under adverse weather conditions have predominantly focused on single-weather scenarios. However, in reality, multiple weather conditions always coexist and their degree of mixing is usually unknown. Under such complex and diverse weather conditions, single-weather restoration models struggle to meet practical demands. This is particularly critical in fields such as autonomous driving, where there is an urgent need for a model capable of effectively handling mixed weather conditions and enhancing image quality in an automated manner. In this paper, we propose a Task Sequence Generator module that, in conjunction with the Task Intra-patch Block, effectively extracts task-specific features embedded in degraded images. The Task Intra-patch Block introduces an external learnable sequence that aids the network in capturing task-specific information. Additionally, we employ a histogram-based transformer module as the backbone of our network, enabling the capture of both global and local dynamic range features. Our proposed model achieves state-of-the-art performance on public datasets.
title Multi-Weather Image Restoration via Histogram-Based Transformer Feature Enhancement
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2409.06334