A Late-Stage Bitemporal Feature Fusion Network for Semantic Change Detection

Semantic change detection is an important task in geoscience and earth observation. By producing a semantic change map for each temporal phase, both the land use land cover categories and change information can be interpreted. Recently some multi-task learning based semantic change detection methods...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhou, Chenyao, Zhang, Haotian, Guo, Han, Zou, Zhengxia, Shi, Zhenwei
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Zhou, Chenyao
Zhang, Haotian
Guo, Han
Zou, Zhengxia
Shi, Zhenwei
description Semantic change detection is an important task in geoscience and earth observation. By producing a semantic change map for each temporal phase, both the land use land cover categories and change information can be interpreted. Recently some multi-task learning based semantic change detection methods have been proposed to decompose the task into semantic segmentation and binary change detection subtasks. However, previous works comprise triple branches in an entangled manner, which may not be optimal and hard to adopt foundation models. Besides, lacking explicit refinement of bitemporal features during fusion may cause low accuracy. In this letter, we propose a novel late-stage bitemporal feature fusion network to address the issue. Specifically, we propose local global attentional aggregation module to strengthen feature fusion, and propose local global context enhancement module to highlight pivotal semantics. Comprehensive experiments are conducted on two public datasets, including SECOND and Landsat-SCD. Quantitative and qualitative results show that our proposed model achieves new state-of-the-art performance on both datasets.
doi_str_mv 10.48550/arxiv.2406.10678
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2406_10678</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2406_10678</sourcerecordid><originalsourceid>FETCH-LOGICAL-a678-c73dbd684e4d6ec2f15d60e68ad5854a3337374da4060636783742752388a0903</originalsourceid><addsrcrecordid>eNotj81OwzAQhH3pAbU8ACf8AkmdrP84toEAUgSH9h4t8aa1aJLKdfl5e0zhNBppZnY_xm4KkUurlFhi-PIfeSmFzguhjb1izYo3GCnbRNwRX_tIw3EKeOA1YTwH4vX55KeRv1D8nMI776fANzTgGH3Hqz2OqXVPkbqYUgs26_Fwout_nbNt_bCtnrLm9fG5WjUZpptZZ8C9OW0lSaepK_tCOS1IW3TKKokAYMBIh-lNoSFVkiuNKsFaFHcC5uz2b_aC0x6DHzB8t79Y7QULfgBAA0Y7</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>A Late-Stage Bitemporal Feature Fusion Network for Semantic Change Detection</title><source>arXiv.org</source><creator>Zhou, Chenyao ; Zhang, Haotian ; Guo, Han ; Zou, Zhengxia ; Shi, Zhenwei</creator><creatorcontrib>Zhou, Chenyao ; Zhang, Haotian ; Guo, Han ; Zou, Zhengxia ; Shi, Zhenwei</creatorcontrib><description>Semantic change detection is an important task in geoscience and earth observation. By producing a semantic change map for each temporal phase, both the land use land cover categories and change information can be interpreted. Recently some multi-task learning based semantic change detection methods have been proposed to decompose the task into semantic segmentation and binary change detection subtasks. However, previous works comprise triple branches in an entangled manner, which may not be optimal and hard to adopt foundation models. Besides, lacking explicit refinement of bitemporal features during fusion may cause low accuracy. In this letter, we propose a novel late-stage bitemporal feature fusion network to address the issue. Specifically, we propose local global attentional aggregation module to strengthen feature fusion, and propose local global context enhancement module to highlight pivotal semantics. Comprehensive experiments are conducted on two public datasets, including SECOND and Landsat-SCD. Quantitative and qualitative results show that our proposed model achieves new state-of-the-art performance on both datasets.</description><identifier>DOI: 10.48550/arxiv.2406.10678</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2024-06</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2406.10678$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2406.10678$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Zhou, Chenyao</creatorcontrib><creatorcontrib>Zhang, Haotian</creatorcontrib><creatorcontrib>Guo, Han</creatorcontrib><creatorcontrib>Zou, Zhengxia</creatorcontrib><creatorcontrib>Shi, Zhenwei</creatorcontrib><title>A Late-Stage Bitemporal Feature Fusion Network for Semantic Change Detection</title><description>Semantic change detection is an important task in geoscience and earth observation. By producing a semantic change map for each temporal phase, both the land use land cover categories and change information can be interpreted. Recently some multi-task learning based semantic change detection methods have been proposed to decompose the task into semantic segmentation and binary change detection subtasks. However, previous works comprise triple branches in an entangled manner, which may not be optimal and hard to adopt foundation models. Besides, lacking explicit refinement of bitemporal features during fusion may cause low accuracy. In this letter, we propose a novel late-stage bitemporal feature fusion network to address the issue. Specifically, we propose local global attentional aggregation module to strengthen feature fusion, and propose local global context enhancement module to highlight pivotal semantics. Comprehensive experiments are conducted on two public datasets, including SECOND and Landsat-SCD. Quantitative and qualitative results show that our proposed model achieves new state-of-the-art performance on both datasets.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj81OwzAQhH3pAbU8ACf8AkmdrP84toEAUgSH9h4t8aa1aJLKdfl5e0zhNBppZnY_xm4KkUurlFhi-PIfeSmFzguhjb1izYo3GCnbRNwRX_tIw3EKeOA1YTwH4vX55KeRv1D8nMI776fANzTgGH3Hqz2OqXVPkbqYUgs26_Fwout_nbNt_bCtnrLm9fG5WjUZpptZZ8C9OW0lSaepK_tCOS1IW3TKKokAYMBIh-lNoSFVkiuNKsFaFHcC5uz2b_aC0x6DHzB8t79Y7QULfgBAA0Y7</recordid><startdate>20240615</startdate><enddate>20240615</enddate><creator>Zhou, Chenyao</creator><creator>Zhang, Haotian</creator><creator>Guo, Han</creator><creator>Zou, Zhengxia</creator><creator>Shi, Zhenwei</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20240615</creationdate><title>A Late-Stage Bitemporal Feature Fusion Network for Semantic Change Detection</title><author>Zhou, Chenyao ; Zhang, Haotian ; Guo, Han ; Zou, Zhengxia ; Shi, Zhenwei</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a678-c73dbd684e4d6ec2f15d60e68ad5854a3337374da4060636783742752388a0903</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>Zhou, Chenyao</creatorcontrib><creatorcontrib>Zhang, Haotian</creatorcontrib><creatorcontrib>Guo, Han</creatorcontrib><creatorcontrib>Zou, Zhengxia</creatorcontrib><creatorcontrib>Shi, Zhenwei</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Zhou, Chenyao</au><au>Zhang, Haotian</au><au>Guo, Han</au><au>Zou, Zhengxia</au><au>Shi, Zhenwei</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A Late-Stage Bitemporal Feature Fusion Network for Semantic Change Detection</atitle><date>2024-06-15</date><risdate>2024</risdate><abstract>Semantic change detection is an important task in geoscience and earth observation. By producing a semantic change map for each temporal phase, both the land use land cover categories and change information can be interpreted. Recently some multi-task learning based semantic change detection methods have been proposed to decompose the task into semantic segmentation and binary change detection subtasks. However, previous works comprise triple branches in an entangled manner, which may not be optimal and hard to adopt foundation models. Besides, lacking explicit refinement of bitemporal features during fusion may cause low accuracy. In this letter, we propose a novel late-stage bitemporal feature fusion network to address the issue. Specifically, we propose local global attentional aggregation module to strengthen feature fusion, and propose local global context enhancement module to highlight pivotal semantics. Comprehensive experiments are conducted on two public datasets, including SECOND and Landsat-SCD. Quantitative and qualitative results show that our proposed model achieves new state-of-the-art performance on both datasets.</abstract><doi>10.48550/arxiv.2406.10678</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2406.10678
ispartof
issn
language eng
recordid cdi_arxiv_primary_2406_10678
source arXiv.org
subjects Computer Science - Computer Vision and Pattern Recognition
title A Late-Stage Bitemporal Feature Fusion Network for Semantic Change Detection
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-19T10%3A52%3A13IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20Late-Stage%20Bitemporal%20Feature%20Fusion%20Network%20for%20Semantic%20Change%20Detection&rft.au=Zhou,%20Chenyao&rft.date=2024-06-15&rft_id=info:doi/10.48550/arxiv.2406.10678&rft_dat=%3Carxiv_GOX%3E2406_10678%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true