{"id":"https://openalex.org/W4296412923","doi":"https://doi.org/10.48550/arxiv.2209.07947","title":"Omni-Dimensional Dynamic Convolution","display_name":"Omni-Dimensional Dynamic Convolution","publication_year":2022,"publication_date":"2022-09-16","ids":{"openalex":"https://openalex.org/W4296412923","doi":"https://doi.org/10.48550/arxiv.2209.07947"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2209.07947","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2209.07947","pdf_url":"https://arxiv.org/pdf/2209.07947","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2209.07947","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100323094","display_name":"Chao Li","orcid":"https://orcid.org/0000-0001-6110-6210","observed_orcids":["https://orcid.org/0000-0001-6110-6210","https://orcid.org/0000-0003-1068-4621"]},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Li, Chao","raw_affiliation_strings":["Intel Labs China ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Intel Labs China ,","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041431476","display_name":"Aojun Zhou","orcid":"https://orcid.org/0000-0002-4742-8624","observed_orcids":["https://orcid.org/0000-0002-4742-8624"]},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Zhou, Aojun","raw_affiliation_strings":["CUHK-SenseTime Joint Lab , The Chinese University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CUHK-SenseTime Joint Lab , The Chinese University of Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5025017745","display_name":"Anbang Yao","orcid":"https://orcid.org/0000-0002-3878-8679","observed_orcids":["https://orcid.org/0000-0002-3878-8679"]},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yao, Anbang","raw_affiliation_strings":["Intel Labs China ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Intel Labs China ,","institution_ids":["https://openalex.org/I1343180700"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":419,"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":0.9990000128746033,"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":0.9990000128746033,"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":0.0005000000237487257,"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":0.00039999998989515007,"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":0.9959999918937683},{"id":"https://openalex.org/keywords/multi-dimensional-attention","display_name":"multi-dimensional attention","score":0.746999979019165},{"id":"https://openalex.org/keywords/kernel-space","display_name":"kernel space","score":0.5419999957084656},{"id":"https://openalex.org/keywords/lightweight-convolutional-neural-network","display_name":"lightweight convolutional neural network","score":0.8050000071525574},{"id":"https://openalex.org/keywords/attention-modules","display_name":"attention modules","score":0.527999997138977},{"id":"https://openalex.org/keywords/representation-learning","display_name":"representation learning","score":0.48500001430511475}],"concepts":[{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.7800609469413757},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.7729809880256653},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.7166373133659363},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6855404376983643},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5252591967582703},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5154405832290649},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.506672203540802},{"id":"https://openalex.org/C157899210","wikidata":"https://www.wikidata.org/wiki/Q1395022","display_name":"Convolutional code","level":3,"score":0.4822309911251068},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.44052308797836304},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.42753279209136963},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4216783344745636},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.40174970030784607},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2092541754245758},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.09016638994216919},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.07981806993484497},{"id":"https://openalex.org/C118615104","wikidata":"https://www.wikidata.org/wiki/Q121416","display_name":"Discrete mathematics","level":1,"score":0.06487846374511719},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C202444582","wikidata":"https://www.wikidata.org/wiki/Q837863","display_name":"Pure mathematics","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2209.07947","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2209.07947","pdf_url":"https://arxiv.org/pdf/2209.07947","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2209.07947","doi":"https://doi.org/10.48550/arxiv.2209.07947","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2209.07947","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":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2209.07947","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2209.07947","pdf_url":"https://arxiv.org/pdf/2209.07947","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4296412923.pdf","grobid_xml":"https://content.openalex.org/works/W4296412923.grobid-xml"},"referenced_works_count":2,"referenced_works":["https://openalex.org/W2612445135","https://openalex.org/W2950141105"],"related_works":["https://openalex.org/W3034421924","https://openalex.org/W4309095463","https://openalex.org/W3128127568","https://openalex.org/W4386858688","https://openalex.org/W3217805790","https://openalex.org/W4399657307","https://openalex.org/W4417069121","https://openalex.org/W4385965877","https://openalex.org/W4409785079","https://openalex.org/W3085075731"],"abstract_inverted_index":{"Learning":[0],"a":[1,30,112,128,134,157],"single":[2,221],"static":[3],"convolutional":[4,8,35,61,72,99,142,154,255],"kernel":[5,73,77,150,222],"in":[6,24],"each":[7,98],"layer":[9],"is":[10,241],"the":[11,45,64,71,76,80,85,88,93,149,174,206,250,254],"common":[12],"training":[13],"paradigm":[14],"of":[15,33,47,75,124,148,160],"modern":[16],"Convolutional":[17],"Neural":[18],"Networks":[19],"(CNNs).":[20],"Instead,":[21],"recent":[22],"research":[23],"dynamic":[25,65,117,229],"convolution":[26,118,230],"shows":[27],"that":[28,57,180],"learning":[29,215],"linear":[31],"combination":[32],"$n$":[34],"kernels":[36,62,143],"weighted":[37],"with":[38,63,133,218,225,232],"their":[39],"input-dependent":[40],"attentions":[41,140],"can":[42,164,223],"significantly":[43],"improve":[44],"accuracy":[46,184],"light-weight":[48,193],"CNNs,":[49],"while":[50],"maintaining":[51],"efficient":[52],"inference.":[53],"However,":[54],"we":[55,106],"observe":[56],"existing":[58,228],"works":[59],"endow":[60],"property":[66],"through":[67],"one":[68,220],"dimension":[69],"(regarding":[70,84],"number)":[74],"space,":[78],"but":[79],"other":[81,245],"three":[82],"dimensions":[83,147],"spatial":[86],"size,":[87],"input":[89],"channel":[90,95],"number":[91,96],"and":[92,176,194],"output":[94,251],"for":[97,141,186,248],"kernel)":[100],"are":[101],"overlooked.":[102],"Inspired":[103],"by":[104],"this,":[105],"present":[107],"Omni-dimensional":[108],"Dynamic":[109],"Convolution":[110],"(ODConv),":[111],"more":[113],"generalized":[114],"yet":[115],"elegant":[116],"design,":[119],"to":[120,137,202,211,244],"advance":[121],"this":[122],"line":[123],"research.":[125],"ODConv":[126,163,181,217,240],"leverages":[127],"novel":[129],"multi-dimensional":[130],"attention":[131,246],"mechanism":[132],"parallel":[135],"strategy":[136],"learn":[138],"complementary":[139],"along":[144],"all":[145],"four":[146],"space":[151],"at":[152],"any":[153],"layer.":[155],"As":[156],"drop-in":[158],"replacement":[159],"regular":[161],"convolutions,":[162],"be":[165],"plugged":[166],"into":[167],"many":[168],"CNN":[169,189],"architectures.":[170],"Extensive":[171],"experiments":[172],"on":[173,205],"ImageNet":[175,207],"MS-COCO":[177],"datasets":[178],"show":[179],"brings":[182],"solid":[183],"boosts":[185],"various":[187],"prevailing":[188],"backbones":[190],"including":[191],"both":[192],"large":[195],"ones,":[196],"e.g.,":[197],"3.77%~5.71%|1.86%~3.72%":[198],"absolute":[199],"top-1":[200],"improvements":[201],"MobivleNetV2|ResNet":[203],"family":[204],"dataset.":[208],"Intriguingly,":[209],"thanks":[210],"its":[212],"improved":[213],"feature":[214],"ability,":[216],"even":[219],"compete":[224],"or":[226,253],"outperform":[227],"counterparts":[231],"multiple":[233],"kernels,":[234],"substantially":[235],"reducing":[236],"extra":[237],"parameters.":[238],"Furthermore,":[239],"also":[242],"superior":[243],"modules":[247],"modulating":[249],"features":[252],"weights.":[256]},"counts_by_year":[{"year":2026,"cited_by_count":73},{"year":2025,"cited_by_count":138},{"year":2024,"cited_by_count":142},{"year":2023,"cited_by_count":61},{"year":2022,"cited_by_count":5}],"updated_date":"2026-11-12T12:28:57.059296","created_date":"2022-09-20T00:00:00"}