{"id":"https://openalex.org/W4417446097","doi":"https://doi.org/10.48550/arxiv.2512.11884","title":"Generalization vs. Specialization: Evaluating Segment Anything Model (SAM3) Zero-Shot Segmentation Against Fine-Tuned YOLO Detectors","display_name":"Generalization vs. Specialization: Evaluating Segment Anything Model (SAM3) Zero-Shot Segmentation Against Fine-Tuned YOLO Detectors","publication_year":2025,"publication_date":"2025-12-09","ids":{"openalex":"https://openalex.org/W4417446097","doi":"https://doi.org/10.48550/arxiv.2512.11884"},"language":null,"primary_location":{"id":"pmh:oai:arXiv.org:2512.11884","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.11884","pdf_url":"https://arxiv.org/pdf/2512.11884","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2512.11884","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5025722346","display_name":"Ranjan Sapkota","orcid":"https://orcid.org/0000-0002-5417-6744"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sapkota, Ranjan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077820227","display_name":"Konstantinos I. Roumeliotis","orcid":"https://orcid.org/0000-0002-8098-1616"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Roumeliotis, Konstantinos I.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013737840","display_name":"Manoj Karkee","orcid":"https://orcid.org/0000-0001-5337-4848"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Karkee, Manoj","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5039512085","display_name":"Nikolaos D. Tselikas","orcid":"https://orcid.org/0000-0001-5799-3558"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tselikas, Nikolaos D.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9240999817848206,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9240999817848206,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.016499999910593033,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.01510000042617321,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/segmentation","display_name":"Segmentation","score":0.6208000183105469},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.6172999739646912},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5855000019073486},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.4885999858379364},{"id":"https://openalex.org/keywords/object-detection","display_name":"Object detection","score":0.4771000146865845},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.4309000074863434},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.43059998750686646},{"id":"https://openalex.org/keywords/boundary","display_name":"Boundary (topology)","score":0.38690000772476196},{"id":"https://openalex.org/keywords/image-segmentation","display_name":"Image segmentation","score":0.35760000348091125}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.733299970626831},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6920999884605408},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.6208000183105469},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.6172999739646912},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5855000019073486},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5157999992370605},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.4885999858379364},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.4771000146865845},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.4309000074863434},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.43059998750686646},{"id":"https://openalex.org/C62354387","wikidata":"https://www.wikidata.org/wiki/Q875399","display_name":"Boundary (topology)","level":2,"score":0.38690000772476196},{"id":"https://openalex.org/C124504099","wikidata":"https://www.wikidata.org/wiki/Q56933","display_name":"Image segmentation","level":3,"score":0.35760000348091125},{"id":"https://openalex.org/C2776023743","wikidata":"https://www.wikidata.org/wiki/Q7234907","display_name":"Pothole (geology)","level":2,"score":0.3379000127315521},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3287000060081482},{"id":"https://openalex.org/C2993043157","wikidata":"https://www.wikidata.org/wiki/Q4949967","display_name":"Boundary line","level":2,"score":0.3276999890804291},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.31459999084472656},{"id":"https://openalex.org/C45371612","wikidata":"https://www.wikidata.org/wiki/Q3058587","display_name":"Generalist and specialist species","level":3,"score":0.31439998745918274},{"id":"https://openalex.org/C94915269","wikidata":"https://www.wikidata.org/wiki/Q1834857","display_name":"Detector","level":2,"score":0.3061999976634979},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.3012999892234802},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2957000136375427},{"id":"https://openalex.org/C3018395757","wikidata":"https://www.wikidata.org/wiki/Q1379672","display_name":"Evaluation methods","level":2,"score":0.29440000653266907},{"id":"https://openalex.org/C17231256","wikidata":"https://www.wikidata.org/wiki/Q5156540","display_name":"Completeness (order theory)","level":2,"score":0.2935999929904938},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.2775000035762787},{"id":"https://openalex.org/C136389625","wikidata":"https://www.wikidata.org/wiki/Q334384","display_name":"Supervised learning","level":3,"score":0.26840001344680786},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.2648000121116638},{"id":"https://openalex.org/C48677424","wikidata":"https://www.wikidata.org/wiki/Q6888088","display_name":"Mode (computer interface)","level":2,"score":0.25119999051094055}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2512.11884","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.11884","pdf_url":"https://arxiv.org/pdf/2512.11884","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2512.11884","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2512.11884","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2512.11884","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.11884","pdf_url":"https://arxiv.org/pdf/2512.11884","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Deep":[0],"learning":[1],"has":[2],"advanced":[3],"two":[4],"fundamentally":[5],"different":[6],"paradigms":[7],"for":[8,54,190],"instance":[9,55,192],"segmentation:":[10],"specialized":[11,181],"models":[12,20,110,183,187],"optimized":[13],"through":[14],"task-specific":[15],"fine-tuning":[16],"and":[17,43,51,87,114,171],"generalist":[18,185],"foundation":[19,186],"capable":[21],"of":[22,46,80,147,153,163,179],"zero-shot":[23,41,123],"segmentation.":[24,56],"This":[25,149,195],"work":[26],"presents":[27],"a":[28,65,176],"comprehensive":[29],"comparison":[30],"between":[31],"SAM3":[32,118,136],"(Segment":[33],"Anything":[34],"Model,":[35],"also":[36],"called":[37],"SAMv3)":[38],"operating":[39],"in":[40,121,155,160],"mode":[42],"three":[44],"variants":[45],"Ultralytics":[47],"YOLO11":[48],"(nano,":[49],"medium,":[50],"large)":[52],"fine-tuned":[53,182],"The":[57],"evaluation":[58,169],"is":[59,198],"conducted":[60],"on":[61,200],"the":[62,103,151],"MinneApple":[63],"dataset,":[64],"dense":[66,191],"benchmark":[67],"comprising":[68],"670":[69],"orchard":[70],"images":[71],"with":[72],"28,179":[73],"annotated":[74],"apple":[75],"instances,":[76],"enabling":[77],"rigorous":[78],"validation":[79],"model":[81],"behavior":[82],"under":[83],"high":[84],"object":[85],"density":[86],"occlusion.":[88],"Our":[89],"analysis":[90],"shows":[91],"IoU":[92,105,133],"choices":[93],"can":[94],"inflate":[95],"performance":[96],"gaps":[97],"by":[98],"up":[99],"to":[100,175],"30%.":[101],"At":[102],"appropriate":[104],"=":[106],"0.15":[107],"threshold,":[108],"YOLO":[109,126],"achieve":[111],"68.9%,":[112],"72.2%,":[113],"71.9%":[115],"F1,":[116],"while":[117],"reaches":[119],"59.8%":[120],"pure":[122],"mode.":[124],"However,":[125],"exhibits":[127],"steep":[128],"degradation":[129],"48-50":[130],"points":[131],"across":[132],"ranges":[134],"whereas":[135],"drops":[137],"only":[138],"4":[139],"points,":[140],"revealing":[141],"12":[142],"times":[143],"superior":[144],"boundary":[145],"stability":[146],"SAM3.":[148],"highlights":[150],"strength":[152],"SAMv3":[154],"mask":[156],"precision":[157],"versus":[158],"specialization":[159],"detection":[161],"completeness":[162],"YOLO11.":[164],"We":[165],"provide":[166],"open-source":[167],"code,":[168],"pipelines,":[170],"methodological":[172],"recommendations,":[173],"contributing":[174],"deeper":[177],"understanding":[178],"when":[180],"or":[184],"are":[188],"preferable":[189],"segmentation":[193],"tasks.":[194],"project":[196],"repository":[197],"available":[199],"GitHub":[201],"as":[202],"https://github.com/Applied-AI-Research-Lab/Segment-Anything-Model-SAM3-Zero-Shot-Segmentation-Against-Fine-Tuned-YOLO-Detectors":[203]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2025-12-17T00:00:00"}