{"id":"https://openalex.org/W4416385902","doi":"https://doi.org/10.1145/3777472","title":"Re-purposing SAM into Efficient Visual Projectors for MLLM-based Referring Image Segmentation","display_name":"Re-purposing SAM into Efficient Visual Projectors for MLLM-based Referring Image Segmentation","publication_year":2025,"publication_date":"2025-11-19","ids":{"openalex":"https://openalex.org/W4416385902","doi":"https://doi.org/10.1145/3777472"},"language":"en","primary_location":{"id":"doi:10.1145/3777472","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3777472","pdf_url":null,"source":{"id":"https://openalex.org/S19610489","display_name":"ACM Transactions on Multimedia Computing Communications and Applications","issn_l":"1551-6857","issn":["1551-6857","1551-6865"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Multimedia Computing, Communications, and Applications","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100816857","display_name":"Xiaobo Yang","orcid":"https://orcid.org/0009-0003-7885-302X"},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaobo Yang","raw_affiliation_strings":["College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China","Zhejiang University, China"],"raw_orcid":"https://orcid.org/0009-0003-7885-302X","affiliations":[{"raw_affiliation_string":"College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China","institution_ids":["https://openalex.org/I76130692"]},{"raw_affiliation_string":"Zhejiang University, China","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5061298101","display_name":"Xiaojin Gong","orcid":"https://orcid.org/0000-0001-9955-3569"},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaojin Gong","raw_affiliation_strings":["College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China","Zhejiang University, China"],"raw_orcid":"https://orcid.org/0000-0001-9955-3569","affiliations":[{"raw_affiliation_string":"College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China","institution_ids":["https://openalex.org/I76130692"]},{"raw_affiliation_string":"Zhejiang University, China","institution_ids":["https://openalex.org/I76130692"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I76130692"],"apc_list":null,"apc_paid":null,"fwci":0.7618,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.77266021,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":"22","issue":"2","first_page":"1","last_page":"26"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.8519999980926514,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.8519999980926514,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.025800000876188278,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.0203000009059906,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.6353999972343445},{"id":"https://openalex.org/keywords/segmentation","display_name":"Segmentation","score":0.61080002784729},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.6085000038146973},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5015000104904175},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.43630000948905945},{"id":"https://openalex.org/keywords/image-segmentation","display_name":"Image segmentation","score":0.3970000147819519},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.39489999413490295},{"id":"https://openalex.org/keywords/visualization","display_name":"Visualization","score":0.36480000615119934}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8888000249862671},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6407999992370605},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.6353999972343445},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.61080002784729},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.6085000038146973},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.5296000242233276},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5015000104904175},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.43630000948905945},{"id":"https://openalex.org/C124504099","wikidata":"https://www.wikidata.org/wiki/Q56933","display_name":"Image segmentation","level":3,"score":0.3970000147819519},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.39489999413490295},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.36480000615119934},{"id":"https://openalex.org/C86034646","wikidata":"https://www.wikidata.org/wiki/Q474311","display_name":"Semantic gap","level":4,"score":0.3562000095844269},{"id":"https://openalex.org/C160633673","wikidata":"https://www.wikidata.org/wiki/Q355198","display_name":"Pixel","level":2,"score":0.3310999870300293},{"id":"https://openalex.org/C130318100","wikidata":"https://www.wikidata.org/wiki/Q2268914","display_name":"Semantic similarity","level":2,"score":0.30239999294281006},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.3001999855041504},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.2883000075817108},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.28299999237060547},{"id":"https://openalex.org/C2775955345","wikidata":"https://www.wikidata.org/wiki/Q7449071","display_name":"Semantic mapping","level":2,"score":0.27160000801086426},{"id":"https://openalex.org/C64754055","wikidata":"https://www.wikidata.org/wiki/Q7574053","display_name":"Spatial contextual awareness","level":2,"score":0.26339998841285706}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3777472","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3777472","pdf_url":null,"source":{"id":"https://openalex.org/S19610489","display_name":"ACM Transactions on Multimedia Computing Communications and Applications","issn_l":"1551-6857","issn":["1551-6857","1551-6865"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Multimedia Computing, Communications, and Applications","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":11,"referenced_works":["https://openalex.org/W2118246710","https://openalex.org/W4312815172","https://openalex.org/W4385572962","https://openalex.org/W4388979610","https://openalex.org/W4389117229","https://openalex.org/W4390871773","https://openalex.org/W4392948412","https://openalex.org/W4394625830","https://openalex.org/W4403939369","https://openalex.org/W4405248620","https://openalex.org/W4405595839"],"related_works":[],"abstract_inverted_index":{"Recently,":[0],"Referring":[1],"Image":[2],"Segmentation":[3],"(RIS)":[4],"frameworks":[5],"that":[6,40,82,161],"pair":[7],"the":[8,15,52,109],"Multimodal":[9],"Large":[10],"Language":[11],"Model":[12,18],"(MLLM)":[13],"with":[14],"Segment":[16],"Anything":[17],"(SAM)":[19],"have":[20],"achieved":[21],"impressive":[22],"results.":[23],"However,":[24],"adapting":[25],"MLLM":[26,176],"to":[27,34,46,67,89,113,134,148],"segmentation":[28],"is":[29],"computationally":[30],"intensive,":[31],"primarily":[32],"due":[33],"visual":[35,43,55,80,103,165,184],"token":[36,65,110],"redundancy.":[37],"We":[38],"observe":[39],"traditional":[41],"patch-wise":[42],"projectors":[44,185],"struggle":[45],"strike":[47],"a":[48,77,129,144],"balance":[49],"between":[50],"reducing":[51],"number":[53],"of":[54,125,138],"tokens":[56,166],"and":[57,98,141,155,178,180],"preserving":[58],"semantic":[59,79,84,100,118,130,145],"clarity,":[60],"often":[61],"retaining":[62],"overly":[63],"long":[64],"sequences":[66],"avoid":[68],"performance":[69],"drops.":[70],"Inspired":[71],"by":[72,87,167],"text":[73],"tokenizers,":[74],"we":[75,127],"propose":[76,128],"novel":[78],"projector":[81],"leverages":[83],"superpixels":[85,101,154],"generated":[86],"SAM":[88],"identify":[90],"\u201cvisual":[91],"words\u201d":[92],"in":[93,120],"an":[94],"image.":[95],"By":[96],"compressing":[97],"projecting":[99],"as":[102],"tokens,":[104],"our":[105,162],"approach":[106],"adaptively":[107],"shortens":[108],"sequence":[111],"according":[112],"scene":[114],"complexity":[115],"while":[116],"minimizing":[117],"loss":[119,124],"compression.":[121],"To":[122],"mitigate":[123],"information,":[126],"superpixel":[131,139,146],"positional":[132],"embedding":[133],"strengthen":[135],"MLLM\u2019s":[136],"awareness":[137],"geometry":[140],"position,":[142],"alongside":[143],"aggregator":[147],"preserve":[149],"both":[150],"fine-grained":[151],"details":[152],"inside":[153],"global":[156],"context":[157],"outside.":[158],"Experiments":[159],"show":[160],"method":[163],"cuts":[164],"\\(\\sim\\)":[168],"93%":[169],"without":[170],"compromising":[171],"performance,":[172],"notably":[173],"speeding":[174],"up":[175],"training":[177],"inference,":[179],"outperforming":[181],"existing":[182],"compressive":[183],"on":[186],"RIS.":[187]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T08:27:34.040176","created_date":"2025-11-19T00:00:00"}