Lightweight Multi-Scale Context Aggregation Deraining Network With Artifact-Attenuating Pooling and Activation Functions

Single image deraining is a fundamental pre-processing step in many computer vision applications for improving the visual effect and system performance of the downstream high-level tasks in adverse weather conditions. This study proposes a novel multi-scale context aggregation network, to effectivel...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Kohei Yamamichi, Xian-Hua Han
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Deep residual block multi-scale progressive aggregation context hallucinate block artifact-attenuating pooling and activation image deraining Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/e9d13871567345b88f253a9e301491ef
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:e9d13871567345b88f253a9e301491ef
record_format	dspace
spelling	oai:doaj.org-article:e9d13871567345b88f253a9e301491ef2021-11-09T00:02:16ZLightweight Multi-Scale Context Aggregation Deraining Network With Artifact-Attenuating Pooling and Activation Functions2169-353610.1109/ACCESS.2021.3122450https://doaj.org/article/e9d13871567345b88f253a9e301491ef2021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9585107/https://doaj.org/toc/2169-3536Single image deraining is a fundamental pre-processing step in many computer vision applications for improving the visual effect and system performance of the downstream high-level tasks in adverse weather conditions. This study proposes a novel multi-scale context aggregation network, to effectively solve the single image deraining problem. Specifically, we exploit a lightweight residual structure subnet as the baseline architecture to extract fine and detailed texture context at the original scale and further incorporate a multi-scale progressive aggregation module (MPAM) to learn the complementary high-level context for enhancing the modeling capability of the overall deraining network. The MPAM, designed as a plug-and-play module to be utilized in the arbitrary network, is composed of multi-scale convolution blocks to learn a wide variety of contexts in multiple receptive fields, and then carries out progressive context aggregation between adjacent scales with residual connections, which is expected to concurrently disentangle the multi-scale structures of scene contents and multiple rain layers in the rainy images, and models more representative contexts for reconstructing the clean image. To reduce the learnable parameters in the MPAM, we further explore a context hallucinate block for replacing the multi-scale convolution block, and propose a lightweight MPAM. Moreover, for being specially adaptive to deal with the input rainy images with a lot of unwanted components (rain layers), we delve into the artifact-attenuating pooling and activation functions via taking into consideration of the surrounding spatial context instead of pixel-wise operation and propose the spatial context-aware pooling (SCAP) and activation (SCAA) for incorporating with our deraining network to boost performance. Extensive experiments on the benchmark datasets demonstrate that our proposed method performs favorably against state-of-the-art deraining approaches.Kohei YamamichiXian-Hua HanIEEEarticleDeep residual blockmulti-scale progressive aggregationcontext hallucinate blockartifact-attenuating pooling and activationimage derainingElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 146948-146958 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Deep residual block multi-scale progressive aggregation context hallucinate block artifact-attenuating pooling and activation image deraining Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Deep residual block multi-scale progressive aggregation context hallucinate block artifact-attenuating pooling and activation image deraining Electrical engineering. Electronics. Nuclear engineering TK1-9971 Kohei Yamamichi Xian-Hua Han Lightweight Multi-Scale Context Aggregation Deraining Network With Artifact-Attenuating Pooling and Activation Functions
description	Single image deraining is a fundamental pre-processing step in many computer vision applications for improving the visual effect and system performance of the downstream high-level tasks in adverse weather conditions. This study proposes a novel multi-scale context aggregation network, to effectively solve the single image deraining problem. Specifically, we exploit a lightweight residual structure subnet as the baseline architecture to extract fine and detailed texture context at the original scale and further incorporate a multi-scale progressive aggregation module (MPAM) to learn the complementary high-level context for enhancing the modeling capability of the overall deraining network. The MPAM, designed as a plug-and-play module to be utilized in the arbitrary network, is composed of multi-scale convolution blocks to learn a wide variety of contexts in multiple receptive fields, and then carries out progressive context aggregation between adjacent scales with residual connections, which is expected to concurrently disentangle the multi-scale structures of scene contents and multiple rain layers in the rainy images, and models more representative contexts for reconstructing the clean image. To reduce the learnable parameters in the MPAM, we further explore a context hallucinate block for replacing the multi-scale convolution block, and propose a lightweight MPAM. Moreover, for being specially adaptive to deal with the input rainy images with a lot of unwanted components (rain layers), we delve into the artifact-attenuating pooling and activation functions via taking into consideration of the surrounding spatial context instead of pixel-wise operation and propose the spatial context-aware pooling (SCAP) and activation (SCAA) for incorporating with our deraining network to boost performance. Extensive experiments on the benchmark datasets demonstrate that our proposed method performs favorably against state-of-the-art deraining approaches.
format	article
author	Kohei Yamamichi Xian-Hua Han
author_facet	Kohei Yamamichi Xian-Hua Han
author_sort	Kohei Yamamichi
title	Lightweight Multi-Scale Context Aggregation Deraining Network With Artifact-Attenuating Pooling and Activation Functions
title_short	Lightweight Multi-Scale Context Aggregation Deraining Network With Artifact-Attenuating Pooling and Activation Functions
title_full	Lightweight Multi-Scale Context Aggregation Deraining Network With Artifact-Attenuating Pooling and Activation Functions
title_fullStr	Lightweight Multi-Scale Context Aggregation Deraining Network With Artifact-Attenuating Pooling and Activation Functions
title_full_unstemmed	Lightweight Multi-Scale Context Aggregation Deraining Network With Artifact-Attenuating Pooling and Activation Functions
title_sort	lightweight multi-scale context aggregation deraining network with artifact-attenuating pooling and activation functions
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/e9d13871567345b88f253a9e301491ef
work_keys_str_mv	AT koheiyamamichi lightweightmultiscalecontextaggregationderainingnetworkwithartifactattenuatingpoolingandactivationfunctions AT xianhuahan lightweightmultiscalecontextaggregationderainingnetworkwithartifactattenuatingpoolingandactivationfunctions
_version_	1718441435053686784

Lightweight Multi-Scale Context Aggregation Deraining Network With Artifact-Attenuating Pooling and Activation Functions

Ejemplares similares