{"id":"https://openalex.org/W4413146472","doi":"https://doi.org/10.1109/cvpr52734.2025.00506","title":"TokenHSI: Unified Synthesis of Physical Human-Scene Interactions through Task Tokenization","display_name":"TokenHSI: Unified Synthesis of Physical Human-Scene Interactions through Task Tokenization","publication_year":2025,"publication_date":"2025-06-10","ids":{"openalex":"https://openalex.org/W4413146472","doi":"https://doi.org/10.1109/cvpr52734.2025.00506"},"language":"en","primary_location":{"id":"doi:10.1109/cvpr52734.2025.00506","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cvpr52734.2025.00506","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","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/A5100612025","display_name":"Liang Pan","orcid":"https://orcid.org/0000-0003-1821-4296"},"institutions":[{"id":"https://openalex.org/I113940042","display_name":"Shanghai University","ror":"https://ror.org/006teas31","country_code":"CN","type":"education","lineage":["https://openalex.org/I113940042"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Liang Pan","raw_affiliation_strings":["ShanghaiAI Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ShanghaiAI Laboratory","institution_ids":["https://openalex.org/I113940042"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Zeshi Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zeshi Yang","raw_affiliation_strings":["Independent Researcher"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Independent Researcher","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059487131","display_name":"Zhiyang Dou","orcid":"https://orcid.org/0000-0003-0186-8269"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Zhiyang Dou","raw_affiliation_strings":["The University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100655526","display_name":"Wenjia Wang","orcid":"https://orcid.org/0000-0001-9219-0494"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Wenjia Wang","raw_affiliation_strings":["The University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014074712","display_name":"Buzhen Huang","orcid":"https://orcid.org/0000-0002-0612-7489"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Buzhen Huang","raw_affiliation_strings":["Southeast University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Southeast University","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101990494","display_name":"Bo Dai","orcid":"https://orcid.org/0000-0003-0866-447X"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Bo Dai","raw_affiliation_strings":["The University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014324841","display_name":"Taku Komura","orcid":"https://orcid.org/0000-0002-2729-5860"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Taku Komura","raw_affiliation_strings":["The University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5025052672","display_name":"Jingbo Wang","orcid":"https://orcid.org/0009-0005-0740-8548"},"institutions":[{"id":"https://openalex.org/I113940042","display_name":"Shanghai University","ror":"https://ror.org/006teas31","country_code":"CN","type":"education","lineage":["https://openalex.org/I113940042"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jingbo Wang","raw_affiliation_strings":["ShanghaiAI Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ShanghaiAI Laboratory","institution_ids":["https://openalex.org/I113940042"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5100612025"],"corresponding_institution_ids":["https://openalex.org/I113940042"],"apc_list":null,"apc_paid":null,"fwci":3.1877,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.92229626,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"5379","last_page":"5391"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12290","display_name":"Human Motion and Animation","score":0.9890999794006348,"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"}},"topics":[{"id":"https://openalex.org/T12290","display_name":"Human Motion and Animation","score":0.9890999794006348,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9835000038146973,"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/T10444","display_name":"Context-Aware Activity Recognition Systems","score":0.9740999937057495,"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":0.796791672706604},{"id":"https://openalex.org/keywords/lexical-analysis","display_name":"Lexical analysis","score":0.6297398805618286},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6138375997543335},{"id":"https://openalex.org/keywords/human\u2013computer-interaction","display_name":"Human\u2013computer interaction","score":0.38415807485580444},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3685925602912903},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.07827574014663696},{"id":"https://openalex.org/keywords/systems-engineering","display_name":"Systems engineering","score":0.06224346160888672}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.796791672706604},{"id":"https://openalex.org/C176982825","wikidata":"https://www.wikidata.org/wiki/Q835922","display_name":"Lexical analysis","level":2,"score":0.6297398805618286},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6138375997543335},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.38415807485580444},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3685925602912903},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.07827574014663696},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.06224346160888672}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cvpr52734.2025.00506","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cvpr52734.2025.00506","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","score":0.4399999976158142,"display_name":"Reduced inequalities"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W4405003489","https://openalex.org/W4386014872","https://openalex.org/W1847536016","https://openalex.org/W4361193986","https://openalex.org/W3149094754","https://openalex.org/W3196817267","https://openalex.org/W1976600725"],"abstract_inverted_index":{"Synthesizing":[0],"diverse":[1],"and":[2,14,82,99,177],"physically":[3],"plausible":[4],"Human-Scene":[5],"Interactions":[6],"(HSI)":[7],"is":[8,88],"pivotal":[9],"for":[10,30],"both":[11],"computer":[12],"animation":[13],"embodied":[15],"AI.":[16],"Despite":[17],"encouraging":[18],"progress,":[19],"current":[20],"methods":[21],"mainly":[22],"focus":[23],"on":[24],"developing":[25],"separate":[26,96],"controllers,":[27],"each":[28],"specialized":[29],"a":[31,42,73,95,107,111],"specific":[32],"interaction":[33,155],"task.":[34],"This":[35],"significantly":[36,173],"hinders":[37],"the":[38,51,91,122,152],"ability":[39],"to":[40,89,139,162],"tackle":[41],"wide":[43],"variety":[44],"of":[45,53,79,136,154],"challenging":[46],"HSI":[47,181],"tasks":[48],"that":[49,169],"require":[50],"integration":[52],"multiple":[54,160],"skills,":[55,119],"e.g.":[56],"sitting":[57],"down":[58],"while":[59],"carrying":[60],"an":[61],"object":[62],"(see":[63],"Fig.":[64],"1).":[65],"To":[66],"address":[67,163],"this":[68],"issue,":[69],"we":[70,147],"present":[71],"TokenHSI,":[72],"single,":[74],"unified":[75,112],"transformer-based":[76],"policy":[77,113,127],"capable":[78],"multi-skill":[80],"unification":[81],"flexible":[83,134],"adaptation.":[84],"The":[85,166],"key":[86],"insight":[87],"model":[90],"humanoid":[92],"proprioception":[93],"as":[94],"shared":[97],"token":[98],"combine":[100],"it":[101],"with":[102],"distinct":[103],"task":[104,145],"tokens":[105],"via":[106],"masking":[108],"mechanism.":[109],"Such":[110],"enables":[114],"effective":[115],"knowledge":[116],"sharing":[117],"across":[118],"thereby":[120],"facilitating":[121],"multi-task":[123],"training.":[124],"Moreover,":[125],"our":[126,170],"architecture":[128],"supports":[129],"variable":[130],"length":[131],"inputs,":[132],"enabling":[133],"adaptation":[135],"learned":[137],"skills":[138,161],"new":[140],"scenarios.":[141],"By":[142],"training":[143],"additional":[144],"tokenizers,":[146],"can":[148,172],"not":[149],"only":[150],"modify":[151],"geometries":[153],"targets":[156],"but":[157],"also":[158],"coordinate":[159],"complex":[164],"tasks.":[165,182],"experiments":[167],"demonstrate":[168],"approach":[171],"improve":[174],"versatility,":[175],"adaptability,":[176],"extensibility":[178],"in":[179],"various":[180]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2}],"updated_date":"2026-05-05T08:41:31.759640","created_date":"2025-10-10T00:00:00"}