{"id":"https://openalex.org/W4416748637","doi":"https://doi.org/10.1109/iros60139.2025.11246465","title":"Interactive Fine-grained Few-shot Detection of Tools <sup>*</sup>","display_name":"Interactive Fine-grained Few-shot Detection of Tools <sup>*</sup>","publication_year":2025,"publication_date":"2025-10-19","ids":{"openalex":"https://openalex.org/W4416748637","doi":"https://doi.org/10.1109/iros60139.2025.11246465"},"language":null,"primary_location":{"id":"doi:10.1109/iros60139.2025.11246465","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iros60139.2025.11246465","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","raw_type":"proceedings-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/A5072035305","display_name":"Philip Keller","orcid":null},"institutions":[{"id":"https://openalex.org/I143379178","display_name":"FZI Research Center for Information Technology","ror":"https://ror.org/04kdh6x72","country_code":"DE","type":"nonprofit","lineage":["https://openalex.org/I143379178"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Philip Keller","raw_affiliation_strings":["FZI Research Center for Information Technology,Karlsruhe,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"FZI Research Center for Information Technology,Karlsruhe,Germany","institution_ids":["https://openalex.org/I143379178"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120707165","display_name":"Leon Strecker","orcid":null},"institutions":[{"id":"https://openalex.org/I70886390","display_name":"Karlsruhe University of Applied Sciences","ror":"https://ror.org/01c0m1t63","country_code":"DE","type":"education","lineage":["https://openalex.org/I70886390"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Leon Strecker","raw_affiliation_strings":["Karlsruhe University of Applied Sciences,Karlsruhe,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Karlsruhe University of Applied Sciences,Karlsruhe,Germany","institution_ids":["https://openalex.org/I70886390"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120478694","display_name":"Felix Durchdewald","orcid":null},"institutions":[{"id":"https://openalex.org/I102335020","display_name":"Karlsruhe Institute of Technology","ror":"https://ror.org/04t3en479","country_code":"DE","type":"education","lineage":["https://openalex.org/I102335020","https://openalex.org/I1305996414"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Felix Durchdewald","raw_affiliation_strings":["Karlsruhe Institute of Technology,Karlsruhe,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Karlsruhe Institute of Technology,Karlsruhe,Germany","institution_ids":["https://openalex.org/I102335020"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092394937","display_name":"Friedrich Graaf","orcid":null},"institutions":[{"id":"https://openalex.org/I143379178","display_name":"FZI Research Center for Information Technology","ror":"https://ror.org/04kdh6x72","country_code":"DE","type":"nonprofit","lineage":["https://openalex.org/I143379178"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Friedrich Graaf","raw_affiliation_strings":["FZI Research Center for Information Technology,Karlsruhe,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"FZI Research Center for Information Technology,Karlsruhe,Germany","institution_ids":["https://openalex.org/I143379178"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056517421","display_name":"Tristan Schnell","orcid":null},"institutions":[{"id":"https://openalex.org/I143379178","display_name":"FZI Research Center for Information Technology","ror":"https://ror.org/04kdh6x72","country_code":"DE","type":"nonprofit","lineage":["https://openalex.org/I143379178"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Tristan Schnell","raw_affiliation_strings":["FZI Research Center for Information Technology,Karlsruhe,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"FZI Research Center for Information Technology,Karlsruhe,Germany","institution_ids":["https://openalex.org/I143379178"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5112552331","display_name":"R\u00fcdiger Dillmann","orcid":null},"institutions":[{"id":"https://openalex.org/I143379178","display_name":"FZI Research Center for Information Technology","ror":"https://ror.org/04kdh6x72","country_code":"DE","type":"nonprofit","lineage":["https://openalex.org/I143379178"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"R\u00fcdiger Dillmann","raw_affiliation_strings":["FZI Research Center for Information Technology,Karlsruhe,Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"FZI Research Center for Information Technology,Karlsruhe,Germany","institution_ids":["https://openalex.org/I143379178"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.33470113,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"15135","last_page":"15142"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.3422999978065491,"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":0.3422999978065491,"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/T10653","display_name":"Robot Manipulation and Learning","score":0.23360000550746918,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"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":0.08290000259876251,"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/construct","display_name":"Construct (python library)","score":0.6367999911308289},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.617900013923645},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.6133000254631042},{"id":"https://openalex.org/keywords/object-detection","display_name":"Object detection","score":0.5062999725341797},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.47850000858306885},{"id":"https://openalex.org/keywords/robot","display_name":"Robot","score":0.4465999901294708},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.3776000142097473}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7793999910354614},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.6367999911308289},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.617900013923645},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.6133000254631042},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.5062999725341797},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.47909998893737793},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.47850000858306885},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.4465999901294708},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4214000105857849},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.3776000142097473},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.37209999561309814},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3208000063896179},{"id":"https://openalex.org/C19966478","wikidata":"https://www.wikidata.org/wiki/Q4810574","display_name":"Mobile robot","level":3,"score":0.2976999878883362},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.2964000105857849},{"id":"https://openalex.org/C186967261","wikidata":"https://www.wikidata.org/wiki/Q5082128","display_name":"Mobile device","level":2,"score":0.29190000891685486},{"id":"https://openalex.org/C89505385","wikidata":"https://www.wikidata.org/wiki/Q47146","display_name":"User interface","level":2,"score":0.2621999979019165},{"id":"https://openalex.org/C203595873","wikidata":"https://www.wikidata.org/wiki/Q25389927","display_name":"Change detection","level":2,"score":0.25870001316070557},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2581999897956848},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2547999918460846},{"id":"https://openalex.org/C98025372","wikidata":"https://www.wikidata.org/wiki/Q477538","display_name":"Systems architecture","level":3,"score":0.25110000371932983}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iros60139.2025.11246465","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iros60139.2025.11246465","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320310414","display_name":"Baden-W\u00fcrttemberg Stiftung","ror":"https://ror.org/031h5fa94"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":26,"referenced_works":["https://openalex.org/W1524405667","https://openalex.org/W1861492603","https://openalex.org/W2193145675","https://openalex.org/W2570343428","https://openalex.org/W2948672349","https://openalex.org/W2963037989","https://openalex.org/W2963150697","https://openalex.org/W2983156430","https://openalex.org/W2989676862","https://openalex.org/W2996583130","https://openalex.org/W3034974675","https://openalex.org/W3092663126","https://openalex.org/W3096609285","https://openalex.org/W3097983980","https://openalex.org/W3150250022","https://openalex.org/W3175630421","https://openalex.org/W4285127796","https://openalex.org/W4312563428","https://openalex.org/W4366083832","https://openalex.org/W4380884472","https://openalex.org/W4388037272","https://openalex.org/W4390874575","https://openalex.org/W4391648915","https://openalex.org/W4395046231","https://openalex.org/W4402916212","https://openalex.org/W4404024980"],"related_works":[],"abstract_inverted_index":{"Few-shot":[0],"object":[1,79],"detection":[2,80,107],"is":[3],"especially":[4],"interesting":[5],"for":[6,54,137,160],"applications":[7],"with":[8,100,162],"mobile":[9],"robots":[10],"and":[11,34,42,44,114,158,165],"becomes":[12],"even":[13],"more":[14],"challenging":[15],"when":[16],"task-related":[17],"classes":[18,63],"are":[19,45],"very":[20],"similar.":[21],"This":[22],"work":[23],"focuses":[24],"on":[25,68,91],"such":[26,57],"a":[27,76,150],"scenario:":[28],"detecting":[29],"different":[30],"types":[31],"of":[32,93,135],"household":[33],"industrial":[35],"tools.":[36],"Such":[37],"tools":[38],"can":[39,111],"be":[40],"rare":[41],"specific":[43],"usually":[46],"not":[47,84],"covered":[48],"by":[49],"existing":[50],"large":[51],"datasets,":[52],"except":[53],"common":[55],"ones":[56],"as":[58],"screwdrivers.":[59],"Additionally,":[60],"the":[61,69,94,105,119,133],"target":[62],"might":[64],"change":[65],"frequently":[66],"depending":[67],"robot\u2019s":[70],"missions.":[71],"Therefore,":[72],"we":[73],"propose":[74],"DE-fine-ViT,":[75],"fine-grained":[77,106,138,153],"few-shot":[78],"model":[81,144],"that":[82],"does":[83],"require":[85],"fine-tuning.":[86],"We":[87,141],"build":[88],"our":[89,128,143],"architecture":[90],"top":[92],"elaborate":[95],"DE-ViT":[96],"model,":[97],"extending":[98],"it":[99],"specialized":[101],"components":[102],"to":[103,118],"improve":[104],"capabilities.":[108],"The":[109],"user":[110],"construct":[112],"class":[113,139],"part":[115],"prototypes":[116,136],"tailored":[117],"task":[120],"in":[121,145],"an":[122],"interactive":[123],"preparation":[124],"phase.":[125],"During":[126],"inference,":[127],"proposed":[129],"reevaluation":[130],"module":[131],"leverages":[132],"multi-granularity":[134],"differentiation.":[140],"evaluate":[142],"multiple":[146],"realistic":[147],"experiments,":[148],"including":[149],"specifically":[151],"created":[152],"dataset,":[154],"demonstrating":[155],"its":[156],"efficacy":[157],"suitability":[159],"scenarios":[161],"little":[163],"data":[164],"low":[166],"inter-class":[167],"variance.":[168]},"counts_by_year":[],"updated_date":"2025-11-28T22:28:52.004314","created_date":"2025-11-28T00:00:00"}