{"id":"https://openalex.org/W4298364821","doi":"https://doi.org/10.1145/3532626","title":"A Survey on Temporal Sentence Grounding in Videos","display_name":"A Survey on Temporal Sentence Grounding in Videos","publication_year":2022,"publication_date":"2022-05-20","ids":{"openalex":"https://openalex.org/W4298364821","doi":"https://doi.org/10.1145/3532626"},"language":"en","primary_location":{"id":"doi:10.1145/3532626","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3532626","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3532626","source":{"id":"https://openalex.org/S19610489","display_name":"ACM Transactions on Multimedia Computing Communications and Applications","issn_l":"1551-6857","issn":["1551-6857","1551-6865"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Multimedia Computing, Communications, and Applications","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3532626","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5009016908","display_name":"Xiaohan Lan","orcid":"https://orcid.org/0000-0001-5382-6699"},"institutions":[{"id":"https://openalex.org/I4210114105","display_name":"Tsinghua\u2013Berkeley Shenzhen Institute","ror":"https://ror.org/02hhwwz98","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210114105","https://openalex.org/I95457486","https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaohan Lan","raw_affiliation_strings":["Tsinghua Shenzhen International Graduate School, Shenzhen, China"],"raw_orcid":"https://orcid.org/0000-0001-5382-6699","affiliations":[{"raw_affiliation_string":"Tsinghua Shenzhen International Graduate School, Shenzhen, China","institution_ids":["https://openalex.org/I4210114105"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011133696","display_name":"Yitian Yuan","orcid":"https://orcid.org/0000-0001-8701-7689"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yitian Yuan","raw_affiliation_strings":["Meituan, Chaoyang District, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-8701-7689","affiliations":[{"raw_affiliation_string":"Meituan, Chaoyang District, Beijing, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022927606","display_name":"Xin Wang","orcid":"https://orcid.org/0000-0002-0351-2939"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xin Wang","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-0351-2939","affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100376399","display_name":"Zhi Wang","orcid":"https://orcid.org/0000-0002-5462-6178"},"institutions":[{"id":"https://openalex.org/I4210114105","display_name":"Tsinghua\u2013Berkeley Shenzhen Institute","ror":"https://ror.org/02hhwwz98","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210114105","https://openalex.org/I95457486","https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhi Wang","raw_affiliation_strings":["Tsinghua Shenzhen International Graduate School, Shenzhen, China"],"raw_orcid":"https://orcid.org/0000-0002-5462-6178","affiliations":[{"raw_affiliation_string":"Tsinghua Shenzhen International Graduate School, Shenzhen, China","institution_ids":["https://openalex.org/I4210114105"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100339293","display_name":"Wenwu Zhu","orcid":"https://orcid.org/0000-0003-2236-9290"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenwu Zhu","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-2236-9290","affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":3.3213,"has_fulltext":true,"cited_by_count":43,"citation_normalized_percentile":{"value":0.93933777,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":100},"biblio":{"volume":"19","issue":"2","first_page":"1","last_page":"33"},"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":0.9997000098228455,"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":0.9883999824523926,"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/computer-science","display_name":"Computer science","score":0.8977636694908142},{"id":"https://openalex.org/keywords/sentence","display_name":"Sentence","score":0.6723892688751221},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.540492594242096},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5380306839942932},{"id":"https://openalex.org/keywords/benchmarking","display_name":"Benchmarking","score":0.4980764389038086},{"id":"https://openalex.org/keywords/taxonomy","display_name":"Taxonomy (biology)","score":0.46824753284454346},{"id":"https://openalex.org/keywords/modalities","display_name":"Modalities","score":0.44019147753715515},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.37899696826934814},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3563425540924072},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.3492780327796936},{"id":"https://openalex.org/keywords/data-science","display_name":"Data science","score":0.33046257495880127}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8977636694908142},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.6723892688751221},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.540492594242096},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5380306839942932},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.4980764389038086},{"id":"https://openalex.org/C58642233","wikidata":"https://www.wikidata.org/wiki/Q8269924","display_name":"Taxonomy (biology)","level":2,"score":0.46824753284454346},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.44019147753715515},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.37899696826934814},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3563425540924072},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3492780327796936},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.33046257495880127},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.0},{"id":"https://openalex.org/C59822182","wikidata":"https://www.wikidata.org/wiki/Q441","display_name":"Botany","level":1,"score":0.0},{"id":"https://openalex.org/C144024400","wikidata":"https://www.wikidata.org/wiki/Q21201","display_name":"Sociology","level":0,"score":0.0},{"id":"https://openalex.org/C162853370","wikidata":"https://www.wikidata.org/wiki/Q39809","display_name":"Marketing","level":1,"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/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","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},{"id":"https://openalex.org/C36289849","wikidata":"https://www.wikidata.org/wiki/Q34749","display_name":"Social science","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3532626","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3532626","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3532626","source":{"id":"https://openalex.org/S19610489","display_name":"ACM Transactions on Multimedia Computing Communications and Applications","issn_l":"1551-6857","issn":["1551-6857","1551-6865"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Multimedia Computing, Communications, and Applications","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1145/3532626","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3532626","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3532626","source":{"id":"https://openalex.org/S19610489","display_name":"ACM Transactions on Multimedia Computing Communications and Applications","issn_l":"1551-6857","issn":["1551-6857","1551-6865"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Multimedia Computing, Communications, and Applications","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.6600000262260437,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[{"id":"https://openalex.org/G119957897","display_name":null,"funder_award_id":"62102222","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G3720940119","display_name":null,"funder_award_id":"2020AAA0106300","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"},{"id":"https://openalex.org/G6655011991","display_name":null,"funder_award_id":"No. 2020AAA0106300","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320322392","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549"},{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4298364821.pdf","grobid_xml":"https://content.openalex.org/works/W4298364821.grobid-xml"},"referenced_works_count":49,"referenced_works":["https://openalex.org/W1516807289","https://openalex.org/W1889081078","https://openalex.org/W2102605133","https://openalex.org/W2111078031","https://openalex.org/W2133598069","https://openalex.org/W2186222003","https://openalex.org/W2531563875","https://openalex.org/W2596164567","https://openalex.org/W2596567068","https://openalex.org/W2597601064","https://openalex.org/W2602856279","https://openalex.org/W2604292070","https://openalex.org/W2607303097","https://openalex.org/W2750779823","https://openalex.org/W2768010949","https://openalex.org/W2791295466","https://openalex.org/W2890502146","https://openalex.org/W2891456603","https://openalex.org/W2915043960","https://openalex.org/W2945961616","https://openalex.org/W2948958195","https://openalex.org/W2951748488","https://openalex.org/W2963521717","https://openalex.org/W2963541336","https://openalex.org/W2966715458","https://openalex.org/W2981750465","https://openalex.org/W2991681548","https://openalex.org/W3034743747","https://openalex.org/W3040468852","https://openalex.org/W3093471110","https://openalex.org/W3095206982","https://openalex.org/W3098671152","https://openalex.org/W3099181346","https://openalex.org/W3101429639","https://openalex.org/W3103542727","https://openalex.org/W3107593835","https://openalex.org/W3143453078","https://openalex.org/W3166712493","https://openalex.org/W3207798279","https://openalex.org/W3210101297","https://openalex.org/W3216763528","https://openalex.org/W4205480697","https://openalex.org/W4214540501","https://openalex.org/W4232688373","https://openalex.org/W4236965008","https://openalex.org/W4243183553","https://openalex.org/W4301409532","https://openalex.org/W4365799947","https://openalex.org/W6792490411"],"related_works":["https://openalex.org/W4238897586","https://openalex.org/W435179959","https://openalex.org/W2619091065","https://openalex.org/W2059640416","https://openalex.org/W1490753184","https://openalex.org/W2284465472","https://openalex.org/W2291782699","https://openalex.org/W1993948687","https://openalex.org/W2000169967","https://openalex.org/W4399363378"],"abstract_inverted_index":{"Temporal":[0],"sentence":[1,22,153],"grounding":[2],"in":[3,28,121,200,209,221],"videos":[4],"(TSGV),":[5],"which":[6,97,214],"aims":[7],"at":[8],"localizing":[9],"one":[10],"target":[11],"segment":[12],"from":[13,38,60],"an":[14],"untrimmed":[15],"video":[16],"with":[17,237],"respect":[18],"to":[19,118,190],"a":[20,92,107],"given":[21],"query,":[23],"has":[24],"drawn":[25],"increasing":[26],"attentions":[27],"the":[29,33,39,100,111,140,147,184,210],"research":[30,134,193,220],"community":[31],"over":[32],"past":[34],"few":[35],"years.":[36],"Different":[37],"task":[40],"of":[41,102,110,129,142],"temporal":[42,152],"action":[43,62],"localization,":[44],"TSGV":[45,65,161,201],"is":[46,66,146],"more":[47,67,218],"flexible":[48],"since":[49,69],"it":[50,70],"can":[51],"locate":[52],"complicated":[53],"activities":[54],"via":[55],"natural":[56],"languages,":[57],"without":[58],"restrictions":[59],"predefined":[61],"categories.":[63],"Meanwhile,":[64],"challenging":[68],"requires":[71],"both":[72],"textual":[73],"and":[74,85,116,123,133,177,187,239],"visual":[75],"understanding":[76],"for":[77,95,136],"semantic":[78],"alignment":[79],"between":[80],"two":[81],"modalities":[82],"(i.e.,":[83,114],"text":[84],"video).":[86],"In":[87],"this":[88,145],"survey,":[89],"we":[90,157,182,196,224],"give":[91],"comprehensive":[93],"overview":[94],"TSGV,":[96,122],"(i)":[98],"summarizes":[99],"taxonomy":[101],"existing":[103,160],"methods,":[104,171,173,176],"(ii)":[105],"provides":[106],"detailed":[108],"description":[109],"evaluation":[112,188,212],"protocols":[113],"datasets":[115,186],"metrics)":[117],"be":[119],"used":[120],"(iii)":[124],"in-depth":[125],"discusses":[126],"potential":[127,205],"problems":[128,206],"current":[130,192,211],"benchmarking":[131],"designs":[132],"directions":[135],"further":[137],"investigations.":[138],"To":[139],"best":[141],"our":[143,227],"knowledge,":[144],"first":[148,158],"systematic":[149],"survey":[150],"on":[151,229,243],"grounding.":[154],"More":[155],"specifically,":[156],"discuss":[159,197],"approaches":[162],"by":[163],"grouping":[164],"them":[165],"into":[166],"four":[167,234],"categories,":[168],"i.e.,":[169],"two-stage":[170],"single-stage":[172],"reinforcement":[174],"learning-based":[175],"weakly":[178],"supervised":[179],"methods.":[180],"Then":[181],"present":[183],"benchmark":[185],"metrics":[189],"assess":[191],"progress.":[194],"Finally,":[195],"some":[198],"limitations":[199],"through":[202],"pointing":[203],"out":[204],"improperly":[207],"resolved":[208],"protocols,":[213],"may":[215],"push":[216],"forwards":[217],"cutting-edge":[219],"TSGV.":[222,244],"Besides,":[223],"also":[225],"share":[226],"insights":[228],"several":[230],"promising":[231],"directions,":[232],"including":[233],"typical":[235],"tasks":[236],"new":[238],"practical":[240],"settings":[241],"based":[242]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":15},{"year":2024,"cited_by_count":12},{"year":2023,"cited_by_count":11},{"year":2022,"cited_by_count":3}],"updated_date":"2026-07-16T13:24:37.021932","created_date":"2025-10-10T00:00:00"}