{"id":"https://openalex.org/W4366207485","doi":"https://doi.org/10.48550/arxiv.2304.07254","title":"Dynamic Mobile-Former: Strengthening Dynamic Convolution with Attention and Residual Connection in Kernel Space","display_name":"Dynamic Mobile-Former: Strengthening Dynamic Convolution with Attention and Residual Connection in Kernel Space","publication_year":2023,"publication_date":"2023-04-13","ids":{"openalex":"https://openalex.org/W4366207485","doi":"https://doi.org/10.48550/arxiv.2304.07254"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2304.07254","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2304.07254","pdf_url":"https://arxiv.org/pdf/2304.07254","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,"listed_in":[],"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-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2304.07254","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Yun, Seokju","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yun, Seokju","raw_affiliation_strings":["Machine Intelligence Laboratory , University of Seoul , Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Machine Intelligence Laboratory , University of Seoul , Korea","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5078628073","display_name":"Youngmin Ro","orcid":"https://orcid.org/0000-0002-0661-9192","observed_orcids":["https://orcid.org/0000-0002-0661-9192"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ro, Youngmin","raw_affiliation_strings":["Machine Intelligence Laboratory , University of Seoul , Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Machine Intelligence Laboratory , University of Seoul , Korea","institution_ids":[]}]}],"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":true,"cited_by_count":4,"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":0.9994999766349792,"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":0.9994999766349792,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.0003000000142492354,"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/T11605","display_name":"Visual Attention and Saliency Detection","score":9.999999747378752e-05,"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/dynamic-convolution","display_name":"dynamic convolution","score":0.8999999761581421},{"id":"https://openalex.org/keywords/lightweight-attention","display_name":"lightweight attention","score":0.6610000133514404},{"id":"https://openalex.org/keywords/kernel-space","display_name":"kernel space","score":0.578000009059906},{"id":"https://openalex.org/keywords/global-and-local-feature-fusion","display_name":"global-local feature fusion","score":0.7919999957084656},{"id":"https://openalex.org/keywords/vision-transformer","display_name":"vision transformer","score":0.6700000166893005},{"id":"https://openalex.org/keywords/object-detection","display_name":"object detection","score":0.972000002861023},{"id":"https://openalex.org/keywords/instance-segmentation","display_name":"instance segmentation","score":0.996999979019165}],"concepts":[{"id":"https://openalex.org/C3826847","wikidata":"https://www.wikidata.org/wiki/Q188768","display_name":"FLOPS","level":2,"score":0.7936073541641235},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7409378886222839},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.6704205274581909},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5774052739143372},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5497572422027588},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.5300326943397522},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.4461471140384674},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.438186913728714},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.36809754371643066},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.31305068731307983},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1436992585659027},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.0},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2304.07254","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2304.07254","pdf_url":"https://arxiv.org/pdf/2304.07254","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,"listed_in":[],"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-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2304.07254","doi":"https://doi.org/10.48550/arxiv.2304.07254","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2304.07254","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,"listed_in":[],"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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2304.07254","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2304.07254","pdf_url":"https://arxiv.org/pdf/2304.07254","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,"listed_in":[],"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-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4366207485.pdf","grobid_xml":"https://content.openalex.org/works/W4366207485.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4288417960","https://openalex.org/W2921438861","https://openalex.org/W2995343971","https://openalex.org/W2982536526","https://openalex.org/W2992221004","https://openalex.org/W3034421924","https://openalex.org/W4386858688","https://openalex.org/W4380302312","https://openalex.org/W4390971171","https://openalex.org/W4385338604"],"abstract_inverted_index":{"We":[0],"introduce":[1],"Dynamic":[2,16,23,38,58,113,126],"Mobile-Former(DMF),":[3],"maximizes":[4],"the":[5,20,73,82,125,153,192],"capabilities":[6],"of":[7,22,66,76,131,156,191],"dynamic":[8,28,78,110],"convolution":[9,79,83],"by":[10,80,164],"harmonizing":[11],"it":[12,52,137],"with":[13,27,166],"efficient":[14],"operators.Our":[15],"MobileFormer":[17],"effectively":[18],"utilizes":[19],"advantages":[21],"MobileNet":[24,59],"(MobileNet":[25],"equipped":[26],"convolution)":[29],"using":[30,197],"global":[31,49,69],"information":[32],"from":[33],"light-weight":[34],"attention.A":[35],"Transformer":[36,61],"in":[37,96,102],"Mobile-Former":[39,114,127,193],"only":[40,117,167],"requires":[41],"a":[42,55,97,129,183],"few":[43],"randomly":[44],"initialized":[45],"tokens":[46],"to":[47],"calculate":[48],"features,":[50],"making":[51],"computationally":[53],"efficient.And":[54],"bridge":[56],"between":[57],"and":[60,68,89,108,134,146,202],"allows":[62,93],"for":[63,94],"bidirectional":[64],"integration":[65],"local":[67],"features.We":[70],"also":[71,121],"simplify":[72],"optimization":[74,95],"process":[75],"vanilla":[77],"splitting":[81],"kernel":[84,88,99],"into":[85],"an":[86,90],"input-agnostic":[87],"input-dependent":[91],"kernel.This":[92],"wider":[98],"space,":[100],"resulting":[101],"enhanced":[103,109],"capacity.By":[104],"integrating":[105],"lightweight":[106],"attention":[107],"convolution,":[111],"our":[112,150],"achieves":[115,138],"not":[116],"high":[118],"efficiency,":[119],"but":[120],"strong":[122],"performance.We":[123],"benchmark":[124],"on":[128,141,158,175],"series":[130],"vision":[132,177],"tasks,":[133],"showcase":[135],"that":[136,190],"impressive":[139],"performance":[140],"image":[142],"classification,":[143],"COCO":[144],"detection,":[145],"instanace":[147],"segmentation.For":[148],"example,":[149],"DMF":[151],"hits":[152],"top-1":[154],"accuracy":[155],"79.4%":[157],"ImageNet-1K,":[159],"much":[160],"higher":[161,188],"than":[162,189],"PVT-Tiny":[163],"4.3%":[165],"1/4":[168],"FLOPs.Additionally,our":[169],"proposed":[170],"DMF-S":[171],"model":[172],"performed":[173],"well":[174],"challenging":[176],"datasets":[178],"such":[179],"as":[180],"COCO,":[181],"achieving":[182],"39.0%":[184],"mAP,which":[185],"is":[186],"1%":[187],"508M":[194],"model,":[195],"despite":[196],"3":[198],"GFLOPs":[199],"less":[200],"computations.Code":[201],"models":[203],"are":[204],"available":[205],"at":[206],"https://github.com/ysj9909/DMF":[207]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":3}],"updated_date":"2026-09-30T09:27:27.532063","created_date":"2025-10-10T00:00:00"}