Segment Anything Model Meets Image Harmonization

Image harmonization is a crucial technique in image composition that aims to seamlessly match the background by adjusting the foreground of composite images. Current methods adopt either global-level or pixel-level feature matching. Global-level feature matching ignores the proximity prior, treating...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2023-12
Main Authors: Chen, Haoxing, Li, Yaohui, Gu, Zhangxuan, Xu, Zhuoer, Lan, Jun, Li, Huaxiong
Format: Article
Language:English
Subjects:
Matching
Pixels
Segments
Semantic segmentation
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Chen, Haoxing
Li, Yaohui
Gu, Zhangxuan
Xu, Zhuoer
Lan, Jun
Li, Huaxiong
description Image harmonization is a crucial technique in image composition that aims to seamlessly match the background by adjusting the foreground of composite images. Current methods adopt either global-level or pixel-level feature matching. Global-level feature matching ignores the proximity prior, treating foreground and background as separate entities. On the other hand, pixel-level feature matching loses contextual information. Therefore, it is necessary to use the information from semantic maps that describe different objects to guide harmonization. In this paper, we propose Semantic-guided Region-aware Instance Normalization (SRIN) that can utilize the semantic segmentation maps output by a pre-trained Segment Anything Model (SAM) to guide the visual consistency learning of foreground and background features. Abundant experiments demonstrate the superiority of our method for image harmonization over state-of-the-art methods.
format article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2904525610</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2904525610</sourcerecordid><originalsourceid>FETCH-proquest_journals_29045256103</originalsourceid><addsrcrecordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mQwCE5Nz03NK1FwzKssycjMS1fwzU9JzVHwTU0tKVbwzE1MT1XwSCzKzc_LrEosyczP42FgTUvMKU7lhdLcDMpuriHOHroFRfmFpanFJfFZ-aVFeUCpeCNLAxNTI1MzQwNj4lQBAFGzMuM</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2904525610</pqid></control><display><type>article</type><title>Segment Anything Model Meets Image Harmonization</title><source>Publicly Available Content Database (Proquest) (PQ_SDU_P3)</source><creator>Chen, Haoxing ; Li, Yaohui ; Gu, Zhangxuan ; Xu, Zhuoer ; Lan, Jun ; Li, Huaxiong</creator><creatorcontrib>Chen, Haoxing ; Li, Yaohui ; Gu, Zhangxuan ; Xu, Zhuoer ; Lan, Jun ; Li, Huaxiong</creatorcontrib><description>Image harmonization is a crucial technique in image composition that aims to seamlessly match the background by adjusting the foreground of composite images. Current methods adopt either global-level or pixel-level feature matching. Global-level feature matching ignores the proximity prior, treating foreground and background as separate entities. On the other hand, pixel-level feature matching loses contextual information. Therefore, it is necessary to use the information from semantic maps that describe different objects to guide harmonization. In this paper, we propose Semantic-guided Region-aware Instance Normalization (SRIN) that can utilize the semantic segmentation maps output by a pre-trained Segment Anything Model (SAM) to guide the visual consistency learning of foreground and background features. Abundant experiments demonstrate the superiority of our method for image harmonization over state-of-the-art methods.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Matching ; Pixels ; Segments ; Semantic segmentation</subject><ispartof>arXiv.org, 2023-12</ispartof><rights>2023. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/2904525610?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>780,784,25753,37012,44590</link.rule.ids></links><search><creatorcontrib>Chen, Haoxing</creatorcontrib><creatorcontrib>Li, Yaohui</creatorcontrib><creatorcontrib>Gu, Zhangxuan</creatorcontrib><creatorcontrib>Xu, Zhuoer</creatorcontrib><creatorcontrib>Lan, Jun</creatorcontrib><creatorcontrib>Li, Huaxiong</creatorcontrib><title>Segment Anything Model Meets Image Harmonization</title><title>arXiv.org</title><description>Image harmonization is a crucial technique in image composition that aims to seamlessly match the background by adjusting the foreground of composite images. Current methods adopt either global-level or pixel-level feature matching. Global-level feature matching ignores the proximity prior, treating foreground and background as separate entities. On the other hand, pixel-level feature matching loses contextual information. Therefore, it is necessary to use the information from semantic maps that describe different objects to guide harmonization. In this paper, we propose Semantic-guided Region-aware Instance Normalization (SRIN) that can utilize the semantic segmentation maps output by a pre-trained Segment Anything Model (SAM) to guide the visual consistency learning of foreground and background features. Abundant experiments demonstrate the superiority of our method for image harmonization over state-of-the-art methods.</description><subject>Matching</subject><subject>Pixels</subject><subject>Segments</subject><subject>Semantic segmentation</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mQwCE5Nz03NK1FwzKssycjMS1fwzU9JzVHwTU0tKVbwzE1MT1XwSCzKzc_LrEosyczP42FgTUvMKU7lhdLcDMpuriHOHroFRfmFpanFJfFZ-aVFeUCpeCNLAxNTI1MzQwNj4lQBAFGzMuM</recordid><startdate>20231220</startdate><enddate>20231220</enddate><creator>Chen, Haoxing</creator><creator>Li, Yaohui</creator><creator>Gu, Zhangxuan</creator><creator>Xu, Zhuoer</creator><creator>Lan, Jun</creator><creator>Li, Huaxiong</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20231220</creationdate><title>Segment Anything Model Meets Image Harmonization</title><author>Chen, Haoxing ; Li, Yaohui ; Gu, Zhangxuan ; Xu, Zhuoer ; Lan, Jun ; Li, Huaxiong</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_29045256103</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Matching</topic><topic>Pixels</topic><topic>Segments</topic><topic>Semantic segmentation</topic><toplevel>online_resources</toplevel><creatorcontrib>Chen, Haoxing</creatorcontrib><creatorcontrib>Li, Yaohui</creatorcontrib><creatorcontrib>Gu, Zhangxuan</creatorcontrib><creatorcontrib>Xu, Zhuoer</creatorcontrib><creatorcontrib>Lan, Jun</creatorcontrib><creatorcontrib>Li, Huaxiong</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database (Proquest) (PQ_SDU_P3)</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Chen, Haoxing</au><au>Li, Yaohui</au><au>Gu, Zhangxuan</au><au>Xu, Zhuoer</au><au>Lan, Jun</au><au>Li, Huaxiong</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Segment Anything Model Meets Image Harmonization</atitle><jtitle>arXiv.org</jtitle><date>2023-12-20</date><risdate>2023</risdate><eissn>2331-8422</eissn><abstract>Image harmonization is a crucial technique in image composition that aims to seamlessly match the background by adjusting the foreground of composite images. Current methods adopt either global-level or pixel-level feature matching. Global-level feature matching ignores the proximity prior, treating foreground and background as separate entities. On the other hand, pixel-level feature matching loses contextual information. Therefore, it is necessary to use the information from semantic maps that describe different objects to guide harmonization. In this paper, we propose Semantic-guided Region-aware Instance Normalization (SRIN) that can utilize the semantic segmentation maps output by a pre-trained Segment Anything Model (SAM) to guide the visual consistency learning of foreground and background features. Abundant experiments demonstrate the superiority of our method for image harmonization over state-of-the-art methods.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2023-12
issn 2331-8422
language eng
recordid cdi_proquest_journals_2904525610
source Publicly Available Content Database (Proquest) (PQ_SDU_P3)
subjects Matching
Pixels
Segments
Semantic segmentation
title Segment Anything Model Meets Image Harmonization
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-27T14%3A09%3A08IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Segment%20Anything%20Model%20Meets%20Image%20Harmonization&rft.jtitle=arXiv.org&rft.au=Chen,%20Haoxing&rft.date=2023-12-20&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2904525610%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-proquest_journals_29045256103%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2904525610&rft_id=info:pmid/&rfr_iscdi=true