A Hierarchical Feature Extraction Network for Fast Scene Segmentation

Semantic segmentation is one of the most active research topics in computer vision with the goal to assign dense semantic labels for all pixels in a given image. In this paper, we introduce HFEN (Hierarchical Feature Extraction Network), a lightweight network to reach a balance between inference spe...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Liu Miao, Yi Zhang
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	semantic segmentation scene understanding hierarchical feature extraction Chemical technology TP1-1185
Acceso en línea:	https://doaj.org/article/09c21491cc544f31a234f355c24dfedf
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:09c21491cc544f31a234f355c24dfedf
record_format	dspace
spelling	oai:doaj.org-article:09c21491cc544f31a234f355c24dfedf2021-11-25T18:58:52ZA Hierarchical Feature Extraction Network for Fast Scene Segmentation10.3390/s212277301424-8220https://doaj.org/article/09c21491cc544f31a234f355c24dfedf2021-11-01T00:00:00Zhttps://www.mdpi.com/1424-8220/21/22/7730https://doaj.org/toc/1424-8220Semantic segmentation is one of the most active research topics in computer vision with the goal to assign dense semantic labels for all pixels in a given image. In this paper, we introduce HFEN (Hierarchical Feature Extraction Network), a lightweight network to reach a balance between inference speed and segmentation accuracy. Our architecture is based on an encoder-decoder framework. The input images are down-sampled through an efficient encoder to extract multi-layer features. Then the extracted features are fused via a decoder, where the global contextual information and spatial information are aggregated for final segmentations with real-time performance. Extensive experiments have been conducted on two standard benchmarks, Cityscapes and Camvid, where our network achieved superior performance on NVIDIA 2080Ti.Liu MiaoYi ZhangMDPI AGarticlesemantic segmentationscene understandinghierarchical feature extractionChemical technologyTP1-1185ENSensors, Vol 21, Iss 7730, p 7730 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	semantic segmentation scene understanding hierarchical feature extraction Chemical technology TP1-1185
spellingShingle	semantic segmentation scene understanding hierarchical feature extraction Chemical technology TP1-1185 Liu Miao Yi Zhang A Hierarchical Feature Extraction Network for Fast Scene Segmentation
description	Semantic segmentation is one of the most active research topics in computer vision with the goal to assign dense semantic labels for all pixels in a given image. In this paper, we introduce HFEN (Hierarchical Feature Extraction Network), a lightweight network to reach a balance between inference speed and segmentation accuracy. Our architecture is based on an encoder-decoder framework. The input images are down-sampled through an efficient encoder to extract multi-layer features. Then the extracted features are fused via a decoder, where the global contextual information and spatial information are aggregated for final segmentations with real-time performance. Extensive experiments have been conducted on two standard benchmarks, Cityscapes and Camvid, where our network achieved superior performance on NVIDIA 2080Ti.
format	article
author	Liu Miao Yi Zhang
author_facet	Liu Miao Yi Zhang
author_sort	Liu Miao
title	A Hierarchical Feature Extraction Network for Fast Scene Segmentation
title_short	A Hierarchical Feature Extraction Network for Fast Scene Segmentation
title_full	A Hierarchical Feature Extraction Network for Fast Scene Segmentation
title_fullStr	A Hierarchical Feature Extraction Network for Fast Scene Segmentation
title_full_unstemmed	A Hierarchical Feature Extraction Network for Fast Scene Segmentation
title_sort	hierarchical feature extraction network for fast scene segmentation
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/09c21491cc544f31a234f355c24dfedf
work_keys_str_mv	AT liumiao ahierarchicalfeatureextractionnetworkforfastscenesegmentation AT yizhang ahierarchicalfeatureextractionnetworkforfastscenesegmentation AT liumiao hierarchicalfeatureextractionnetworkforfastscenesegmentation AT yizhang hierarchicalfeatureextractionnetworkforfastscenesegmentation
_version_	1718410445306462208

A Hierarchical Feature Extraction Network for Fast Scene Segmentation

Ejemplares similares