{"id":"https://openalex.org/W3034584023","doi":"https://doi.org/10.1145/3397271.3401122","title":"3D Self-Attention for Unsupervised Video Quantization","display_name":"3D Self-Attention for Unsupervised Video Quantization","publication_year":2020,"publication_date":"2020-07-25","ids":{"openalex":"https://openalex.org/W3034584023","doi":"https://doi.org/10.1145/3397271.3401122","mag":"3034584023"},"language":"en","primary_location":{"id":"doi:10.1145/3397271.3401122","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3397271.3401122","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval","raw_type":"proceedings-article"},"type":"conference-paper","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/A5036987388","display_name":"Jingkuan Song","orcid":"https://orcid.org/0000-0002-2549-8322"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jingkuan Song","raw_affiliation_strings":["University of Electronic Science and Technology of China, ChengDu, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, ChengDu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022853838","display_name":"Ruimin Lang","orcid":null},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ruimin Lang","raw_affiliation_strings":["University of Electronic Science and Technology of China, ChengDu, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, ChengDu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079828357","display_name":"Xiaosu Zhu","orcid":"https://orcid.org/0000-0001-7728-2518"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaosu Zhu","raw_affiliation_strings":["University of Electronic Science and Technology of China, ChengDu, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, ChengDu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5009074046","display_name":"Xing Xu","orcid":"https://orcid.org/0000-0001-5685-3123"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xing Xu","raw_affiliation_strings":["University of Electronic Science and Technology of China, ChengDu, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, ChengDu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066645546","display_name":"Lianli Gao","orcid":"https://orcid.org/0000-0002-2522-6394"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lianli Gao","raw_affiliation_strings":["University of Electronic Science and Technology of China, ChengDu, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, ChengDu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5052993469","display_name":"Heng Tao Shen","orcid":"https://orcid.org/0000-0002-2999-2088"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Heng Tao Shen","raw_affiliation_strings":["University of Electronic Science and Technology of China, ChengDu, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, ChengDu, China","institution_ids":["https://openalex.org/I150229711"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I150229711"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1061","last_page":"1070"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9998999834060669,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9998999834060669,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9995999932289124,"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":0.8467077016830444},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.7655935287475586},{"id":"https://openalex.org/keywords/pixel","display_name":"Pixel","score":0.6808396577835083},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6764272451400757},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.5862783789634705},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5043185949325562},{"id":"https://openalex.org/keywords/codebook","display_name":"Codebook","score":0.4807731509208679},{"id":"https://openalex.org/keywords/residual","display_name":"Residual","score":0.46401938796043396},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.446990430355072},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.40088769793510437},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.16694828867912292}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8467077016830444},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.7655935287475586},{"id":"https://openalex.org/C160633673","wikidata":"https://www.wikidata.org/wiki/Q355198","display_name":"Pixel","level":2,"score":0.6808396577835083},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6764272451400757},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.5862783789634705},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5043185949325562},{"id":"https://openalex.org/C127759330","wikidata":"https://www.wikidata.org/wiki/Q637416","display_name":"Codebook","level":2,"score":0.4807731509208679},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.46401938796043396},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.446990430355072},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.40088769793510437},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.16694828867912292},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"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":1,"locations":[{"id":"doi:10.1145/3397271.3401122","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3397271.3401122","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":44,"referenced_works":["https://openalex.org/W28988658","https://openalex.org/W138284169","https://openalex.org/W1522734439","https://openalex.org/W1777628566","https://openalex.org/W1927052826","https://openalex.org/W1944615693","https://openalex.org/W1956333070","https://openalex.org/W1974647172","https://openalex.org/W1977182282","https://openalex.org/W1983364832","https://openalex.org/W2016053056","https://openalex.org/W2044072771","https://openalex.org/W2067766814","https://openalex.org/W2096621047","https://openalex.org/W2111006384","https://openalex.org/W2124509324","https://openalex.org/W2156303437","https://openalex.org/W2194775991","https://openalex.org/W2335437633","https://openalex.org/W2507009361","https://openalex.org/W2520799877","https://openalex.org/W2527195341","https://openalex.org/W2553594924","https://openalex.org/W2593190442","https://openalex.org/W2620629206","https://openalex.org/W2767687205","https://openalex.org/W2786585376","https://openalex.org/W2889119508","https://openalex.org/W2904452845","https://openalex.org/W2916245046","https://openalex.org/W2951019013","https://openalex.org/W2963403868","https://openalex.org/W2963524571","https://openalex.org/W2966153044","https://openalex.org/W2966196493","https://openalex.org/W2979656080","https://openalex.org/W2981317786","https://openalex.org/W2981582341","https://openalex.org/W2982512683","https://openalex.org/W2988283545","https://openalex.org/W3098959905","https://openalex.org/W3099677434","https://openalex.org/W4244017338","https://openalex.org/W6600186770"],"related_works":["https://openalex.org/W2293149949","https://openalex.org/W2026099691","https://openalex.org/W4284672201","https://openalex.org/W2377486419","https://openalex.org/W2017956276","https://openalex.org/W2943202426","https://openalex.org/W2736714427","https://openalex.org/W2950156284","https://openalex.org/W2163679795","https://openalex.org/W2137816434"],"abstract_inverted_index":{"Unsupervised":[0],"video":[1,15,37,68,72],"quantization":[2,34,134,139,174],"is":[3,106,183],"to":[4,9,32,61,66,78,95,141],"compress":[5],"the":[6,51,97,123,143,159,166,171,177],"original":[7],"videos":[8],"compact":[10],"binary":[11],"codes":[12],"so":[13],"that":[14,154,164],"retrieval":[16,38,44],"can":[17,120,175],"be":[18],"conducted":[19],"in":[20],"an":[21,63],"efficient":[22,83],"way.":[23],"In":[24,50],"this":[25],"paper,":[26],"we":[27,54,89,130],"make":[28],"a":[29,91,114],"first":[30],"attempt":[31],"combine":[33],"method":[35,156],"with":[36,46],"called":[39],"3D-UVQ,":[40],"which":[41,135],"obtains":[42],"high":[43],"accuracy":[45],"low":[47],"storage":[48],"cost.":[49],"proposed":[52],"framework,":[53],"address":[55],"two":[56],"main":[57],"problems:":[58],"1)":[59],"how":[60,77],"design":[62],"effective":[64],"pipeline":[65],"perceive":[67],"contextual":[69,101],"information":[70],"for":[71,82],"features":[73,81,144],"extraction;":[74],"and":[75,99,170],"2)":[76],"quantize":[79],"these":[80,87],"retrieval.":[84,180],"To":[85],"tackle":[86],"problems,":[88],"propose":[90,131],"3D":[92,167],"self-attention":[93,168],"module":[94,169],"exploit":[96],"spatial":[98],"temporal":[100],"information,":[102],"where":[103],"each":[104,118],"pixel":[105,119],"influenced":[107],"by":[108],"its":[109],"surrounding":[110],"pixels.":[111,128],"By":[112],"taking":[113],"further":[115],"recurrent":[116],"operation,":[117],"finally":[121],"capture":[122],"global":[124],"context":[125],"from":[126],"all":[127],"Then,":[129],"gradient-based":[132,172],"residual":[133,173],"consists":[136],"of":[137,179],"several":[138],"blocks":[140],"approximate":[142],"gradually.":[145],"Extensive":[146],"experimental":[147],"results":[148],"on":[149],"three":[150],"benchmark":[151],"datasets":[152],"demonstrate":[153],"our":[155],"significantly":[157],"outperforms":[158],"state-of-the-arts.":[160],"Ablation":[161],"study":[162],"shows":[163],"both":[165],"improve":[176],"performance":[178],"Our":[181],"model":[182],"publicly":[184],"available":[185],"at":[186],"https://github.com/brownwolf/3D-UVQ.":[187]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}