{"id":"https://openalex.org/W3009192917","doi":"https://doi.org/10.1109/tpami.2019.2940007","title":"SibNet: Sibling Convolutional Encoder for Video Captioning","display_name":"SibNet: Sibling Convolutional Encoder for Video Captioning","publication_year":2020,"publication_date":"2020-03-06","ids":{"openalex":"https://openalex.org/W3009192917","doi":"https://doi.org/10.1109/tpami.2019.2940007","mag":"3009192917","pmid":"https://pubmed.ncbi.nlm.nih.gov/32149622"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2019.2940007","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2019.2940007","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","pubmed"],"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/A5101874791","display_name":"Sheng Liu","orcid":"https://orcid.org/0000-0003-4121-2088"},"institutions":[{"id":"https://openalex.org/I63190737","display_name":"University at Buffalo, State University of New York","ror":"https://ror.org/01y64my43","country_code":"US","type":"education","lineage":["https://openalex.org/I63190737"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Sheng Liu","raw_affiliation_strings":["Department of Computer Science and Engineering, University at Buffalo, Buffalo, NY, USA"],"raw_orcid":"https://orcid.org/0000-0003-4121-2088","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, University at Buffalo, Buffalo, NY, USA","institution_ids":["https://openalex.org/I63190737"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083293468","display_name":"Zhou Ren","orcid":"https://orcid.org/0000-0002-1402-966X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou Ren","raw_affiliation_strings":["Wormpex AI Research, Seattle, WA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Wormpex AI Research, Seattle, WA, USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5085245110","display_name":"Junsong Yuan","orcid":"https://orcid.org/0000-0002-7901-8793"},"institutions":[{"id":"https://openalex.org/I63190737","display_name":"University at Buffalo, State University of New York","ror":"https://ror.org/01y64my43","country_code":"US","type":"education","lineage":["https://openalex.org/I63190737"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Junsong Yuan","raw_affiliation_strings":["Department of Computer Science and Engineering, University at Buffalo, Buffalo, NY, USA"],"raw_orcid":"https://orcid.org/0000-0002-7324-7034","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, University at Buffalo, Buffalo, NY, USA","institution_ids":["https://openalex.org/I63190737"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5101874791"],"corresponding_institution_ids":["https://openalex.org/I63190737"],"apc_list":null,"apc_paid":null,"fwci":7.7515,"has_fulltext":false,"cited_by_count":98,"citation_normalized_percentile":{"value":0.98050269,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":100},"biblio":{"volume":"43","issue":"9","first_page":"3259","last_page":"3272"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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":1.0,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9975000023841858,"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/T11439","display_name":"Video Analysis and Summarization","score":0.9940999746322632,"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/closed-captioning","display_name":"Closed captioning","score":0.8755970001220703},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8698872923851013},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.639266848564148},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6365190744400024},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.5590500235557556},{"id":"https://openalex.org/keywords/autoencoder","display_name":"Autoencoder","score":0.5467173457145691},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.5124643445014954},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4959934651851654},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.49587640166282654},{"id":"https://openalex.org/keywords/encode","display_name":"ENCODE","score":0.46567413210868835},{"id":"https://openalex.org/keywords/visualization","display_name":"Visualization","score":0.459073543548584},{"id":"https://openalex.org/keywords/task-analysis","display_name":"Task analysis","score":0.4441727101802826},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.3527761995792389},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.3480214476585388},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.34185346961021423},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.19914215803146362}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.8755970001220703},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8698872923851013},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.639266848564148},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6365190744400024},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.5590500235557556},{"id":"https://openalex.org/C101738243","wikidata":"https://www.wikidata.org/wiki/Q786435","display_name":"Autoencoder","level":3,"score":0.5467173457145691},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.5124643445014954},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4959934651851654},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.49587640166282654},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.46567413210868835},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.459073543548584},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.4441727101802826},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.3527761995792389},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.3480214476585388},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.34185346961021423},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.19914215803146362},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tpami.2019.2940007","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2019.2940007","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},{"id":"pmid:32149622","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/32149622","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE transactions on pattern analysis and machine intelligence","raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.4699999988079071,"display_name":"Quality Education"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320332641","display_name":"University at Buffalo","ror":"https://ror.org/01y64my43"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":138,"referenced_works":["https://openalex.org/W15066456","https://openalex.org/W24089286","https://openalex.org/W877909479","https://openalex.org/W1514535095","https://openalex.org/W1522301498","https://openalex.org/W1522734439","https://openalex.org/W1533861849","https://openalex.org/W1573040851","https://openalex.org/W1586939924","https://openalex.org/W1665214252","https://openalex.org/W1753482797","https://openalex.org/W1840435438","https://openalex.org/W1869752048","https://openalex.org/W1889081078","https://openalex.org/W1895577753","https://openalex.org/W1902237438","https://openalex.org/W1924770834","https://openalex.org/W1956340063","https://openalex.org/W1968240228","https://openalex.org/W1995820507","https://openalex.org/W2017257315","https://openalex.org/W2064675550","https://openalex.org/W2095705004","https://openalex.org/W2097117768","https://openalex.org/W2101105183","https://openalex.org/W2102409316","https://openalex.org/W2130942839","https://openalex.org/W2133564696","https://openalex.org/W2134036914","https://openalex.org/W2139501017","https://openalex.org/W2144513243","https://openalex.org/W2154652894","https://openalex.org/W2157331557","https://openalex.org/W2158139315","https://openalex.org/W2164290393","https://openalex.org/W2168894214","https://openalex.org/W2183341477","https://openalex.org/W2187089797","https://openalex.org/W2194775991","https://openalex.org/W2302086703","https://openalex.org/W2337676748","https://openalex.org/W2425121537","https://openalex.org/W2507009361","https://openalex.org/W2507365558","https://openalex.org/W2519091744","https://openalex.org/W2519434724","https://openalex.org/W2523937091","https://openalex.org/W2523993696","https://openalex.org/W2527145521","https://openalex.org/W2527238922","https://openalex.org/W2527349934","https://openalex.org/W2537916383","https://openalex.org/W2549139847","https://openalex.org/W2552839021","https://openalex.org/W2554906389","https://openalex.org/W2556388456","https://openalex.org/W2557264465","https://openalex.org/W2558834163","https://openalex.org/W2559542432","https://openalex.org/W2567070169","https://openalex.org/W2597655663","https://openalex.org/W2607119937","https://openalex.org/W2607151106","https://openalex.org/W2613904329","https://openalex.org/W2618127004","https://openalex.org/W2652651782","https://openalex.org/W2739107216","https://openalex.org/W2752191396","https://openalex.org/W2765658575","https://openalex.org/W2766375149","https://openalex.org/W2766520430","https://openalex.org/W2788277448","https://openalex.org/W2798592524","https://openalex.org/W2798725893","https://openalex.org/W2803408761","https://openalex.org/W2890428938","https://openalex.org/W2913822893","https://openalex.org/W2949382160","https://openalex.org/W2949952998","https://openalex.org/W2962681491","https://openalex.org/W2962965405","https://openalex.org/W2963069010","https://openalex.org/W2963149042","https://openalex.org/W2963201326","https://openalex.org/W2963351113","https://openalex.org/W2963386218","https://openalex.org/W2963389687","https://openalex.org/W2963403868","https://openalex.org/W2963410018","https://openalex.org/W2963446712","https://openalex.org/W2963576560","https://openalex.org/W2963758027","https://openalex.org/W2963842982","https://openalex.org/W2963843052","https://openalex.org/W2963929190","https://openalex.org/W2963971014","https://openalex.org/W2964065937","https://openalex.org/W2964121744","https://openalex.org/W2964137974","https://openalex.org/W2964241990","https://openalex.org/W2964265128","https://openalex.org/W2964308564","https://openalex.org/W2964350399","https://openalex.org/W3098682680","https://openalex.org/W4285719527","https://openalex.org/W4289542401","https://openalex.org/W4385245566","https://openalex.org/W6600593648","https://openalex.org/W6600983433","https://openalex.org/W6623995992","https://openalex.org/W6630875275","https://openalex.org/W6631190155","https://openalex.org/W6631943919","https://openalex.org/W6637242042","https://openalex.org/W6637698695","https://openalex.org/W6639082767","https://openalex.org/W6639432524","https://openalex.org/W6640212811","https://openalex.org/W6674330103","https://openalex.org/W6675401909","https://openalex.org/W6679434410","https://openalex.org/W6679436768","https://openalex.org/W6681151457","https://openalex.org/W6682631176","https://openalex.org/W6683557909","https://openalex.org/W6684090549","https://openalex.org/W6684665197","https://openalex.org/W6685145238","https://openalex.org/W6685670348","https://openalex.org/W6728621371","https://openalex.org/W6731370813","https://openalex.org/W6735377749","https://openalex.org/W6737778391","https://openalex.org/W6739682802","https://openalex.org/W6739901393","https://openalex.org/W6748322150","https://openalex.org/W6754102128","https://openalex.org/W6898505805"],"related_works":["https://openalex.org/W4210416330","https://openalex.org/W2775506363","https://openalex.org/W3088136942","https://openalex.org/W4290852288","https://openalex.org/W2949362007","https://openalex.org/W4283207562","https://openalex.org/W2963177403","https://openalex.org/W2187606256","https://openalex.org/W2275988210","https://openalex.org/W4399485860"],"abstract_inverted_index":{"Visual":[0],"captioning,":[1,84,205],"the":[2,22,26,38,103,109,114,122,130,142,178,184,191,194,213],"task":[3,19],"of":[4,24,40,102,113,129,193],"describing":[5,31],"an":[6,106],"image":[7],"or":[8,13],"a":[9,17,71,76,87,162],"video":[10,104,115,131,204],"using":[11,33,70],"one":[12],"few":[14],"sentences,":[15],"is":[16],"challenging":[18],"owing":[20],"to":[21,51,54,90,165],"complexity":[23],"understanding":[25],"copious":[27],"visual":[28,68,83,99,110],"information":[29,69,101,112,128,186],"and":[30,158,175,207],"it":[32],"natural":[34],"language.":[35],"Motivated":[36],"by":[37,140],"success":[39],"applying":[41],"neural":[42],"networks":[43,118],"for":[44,82,203],"machine":[45],"translation,":[46],"previous":[47,64],"work":[48,65],"applies":[49],"sequence":[50,52],"learning":[53],"translate":[55],"videos":[56],"into":[57,161],"sentences.":[58],"In":[59],"this":[60],"work,":[61],"different":[62,221],"from":[63,147],"that":[66,212],"encodes":[67,98,126],"single":[72],"flow,":[73],"we":[74,197],"introduce":[75],"novel":[77],"Sibling":[78],"Convolutional":[79],"Encoder":[80],"(SibNet)":[81],"which":[85,136],"employs":[86],"dual-branch":[88],"architecture":[89],"collaboratively":[91],"encode":[92],"videos.":[93,148,188],"The":[94],"first":[95],"content":[96,100,174],"branch":[97,125],"with":[105,155],"autoencoder,":[107],"capturing":[108,172],"appearance":[111],"as":[116],"other":[117],"often":[119],"do.":[120],"While":[121],"second":[123],"semantic":[124,127,176],"via":[132],"visual-semantic":[133],"joint":[134],"embedding,":[135],"brings":[137],"complementary":[138],"representation":[139],"considering":[141],"semantics":[143],"when":[144],"extracting":[145],"features":[146],"Then":[149],"both":[150,173],"branches":[151],"are":[152],"effectively":[153],"combined":[154],"soft-attention":[156],"mechanism":[157],"finally":[159],"fed":[160],"RNN":[163],"decoder":[164],"generate":[166],"captions.":[167],"With":[168],"our":[169],"SibNet":[170,215],"explicitly":[171],"information,":[177],"proposed":[179,195,214],"model":[180],"can":[181],"better":[182],"represent":[183],"rich":[185],"in":[187],"To":[189],"validate":[190],"advantages":[192],"model,":[196],"conduct":[198],"experiments":[199],"on":[200],"two":[201],"benchmarks":[202],"YouTube2Text":[206],"MSR-VTT.":[208],"Our":[209],"results":[210],"demonstrate":[211],"consistently":[216],"outperforms":[217],"existing":[218],"methods":[219],"across":[220],"evaluation":[222],"metrics.":[223]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":6},{"year":2024,"cited_by_count":12},{"year":2023,"cited_by_count":29},{"year":2022,"cited_by_count":30},{"year":2021,"cited_by_count":17},{"year":2020,"cited_by_count":3}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}