{"id":"https://openalex.org/W3013419896","doi":"https://doi.org/10.1109/wacv45572.2020.9093341","title":"ULSAM: Ultra-Lightweight Subspace Attention Module for Compact Convolutional Neural Networks","display_name":"ULSAM: Ultra-Lightweight Subspace Attention Module for Compact Convolutional Neural Networks","publication_year":2020,"publication_date":"2020-03-01","ids":{"openalex":"https://openalex.org/W3013419896","doi":"https://doi.org/10.1109/wacv45572.2020.9093341","mag":"3013419896"},"language":"en","primary_location":{"id":"doi:10.1109/wacv45572.2020.9093341","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv45572.2020.9093341","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2020 IEEE Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2006.15102","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Rajat Saini","orcid":null},"institutions":[{"id":"https://openalex.org/I65181880","display_name":"Indian Institute of Technology Hyderabad","ror":"https://ror.org/01j4v3x97","country_code":"IN","type":"education","lineage":["https://openalex.org/I65181880"]}],"countries":["IN"],"is_corresponding":true,"raw_author_name":"Rajat Saini","raw_affiliation_strings":["IIT Hyderabad"],"affiliations":[{"raw_affiliation_string":"IIT Hyderabad","institution_ids":["https://openalex.org/I65181880"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Nandan Kumar Jha","orcid":null},"institutions":[{"id":"https://openalex.org/I65181880","display_name":"Indian Institute of Technology Hyderabad","ror":"https://ror.org/01j4v3x97","country_code":"IN","type":"education","lineage":["https://openalex.org/I65181880"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Nandan Kumar Jha","raw_affiliation_strings":["IIT Hyderabad"],"affiliations":[{"raw_affiliation_string":"IIT Hyderabad","institution_ids":["https://openalex.org/I65181880"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Bedanta Das","orcid":null},"institutions":[{"id":"https://openalex.org/I65181880","display_name":"Indian Institute of Technology Hyderabad","ror":"https://ror.org/01j4v3x97","country_code":"IN","type":"education","lineage":["https://openalex.org/I65181880"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Bedanta Das","raw_affiliation_strings":["IIT Hyderabad"],"affiliations":[{"raw_affiliation_string":"IIT Hyderabad","institution_ids":["https://openalex.org/I65181880"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Sparsh Mittal","orcid":null},"institutions":[{"id":"https://openalex.org/I154851008","display_name":"Indian Institute of Technology Roorkee","ror":"https://ror.org/00582g326","country_code":"IN","type":"education","lineage":["https://openalex.org/I154851008"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Sparsh Mittal","raw_affiliation_strings":["IIT Roorkee"],"affiliations":[{"raw_affiliation_string":"IIT Roorkee","institution_ids":["https://openalex.org/I154851008"]}]},{"author_position":"last","author":{"id":null,"display_name":"C. Krishna Mohan","orcid":null},"institutions":[{"id":"https://openalex.org/I65181880","display_name":"Indian Institute of Technology Hyderabad","ror":"https://ror.org/01j4v3x97","country_code":"IN","type":"education","lineage":["https://openalex.org/I65181880"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"C. Krishna Mohan","raw_affiliation_strings":["IIT Hyderabad"],"affiliations":[{"raw_affiliation_string":"IIT Hyderabad","institution_ids":["https://openalex.org/I65181880"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I65181880"],"apc_list":null,"apc_paid":null,"fwci":4.4161,"has_fulltext":false,"cited_by_count":108,"citation_normalized_percentile":{"value":0.95648705,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"1616","last_page":"1625"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998000264167786,"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":0.9998000264167786,"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/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.9948999881744385,"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/T10320","display_name":"Neural Networks and Applications","score":0.9940000176429749,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/subspace-topology","display_name":"Subspace topology","score":0.7634000182151794},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.6603999733924866},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.5983999967575073},{"id":"https://openalex.org/keywords/convolution","display_name":"Convolution (computer science)","score":0.5706999897956848},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.5468000173568726},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.49540001153945923},{"id":"https://openalex.org/keywords/field","display_name":"Field (mathematics)","score":0.4812000095844269},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.42320001125335693},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.3971000015735626}],"concepts":[{"id":"https://openalex.org/C32834561","wikidata":"https://www.wikidata.org/wiki/Q660730","display_name":"Subspace topology","level":2,"score":0.7634000182151794},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.745199978351593},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.6603999733924866},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6022999882698059},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.5983999967575073},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.5706999897956848},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.5468000173568726},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.49540001153945923},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.4812000095844269},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.42320001125335693},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3971000015735626},{"id":"https://openalex.org/C75294576","wikidata":"https://www.wikidata.org/wiki/Q5165192","display_name":"Contextual image classification","level":3,"score":0.36649999022483826},{"id":"https://openalex.org/C3826847","wikidata":"https://www.wikidata.org/wiki/Q188768","display_name":"FLOPS","level":2,"score":0.34450000524520874},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.3294000029563904},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.321399986743927},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.3151000142097473},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3133000135421753},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3043000102043152},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.29019999504089355},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.2870999872684479},{"id":"https://openalex.org/C70518039","wikidata":"https://www.wikidata.org/wiki/Q16000077","display_name":"Dimensionality reduction","level":2,"score":0.2858000099658966},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.27250000834465027},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.260699987411499},{"id":"https://openalex.org/C2778012447","wikidata":"https://www.wikidata.org/wiki/Q1034415","display_name":"Scope (computer science)","level":2,"score":0.25699999928474426},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.25540000200271606}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/wacv45572.2020.9093341","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv45572.2020.9093341","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2020 IEEE Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2006.15102","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2006.15102","pdf_url":"https://arxiv.org/pdf/2006.15102","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2006.15102","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2006.15102","pdf_url":"https://arxiv.org/pdf/2006.15102","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":46,"referenced_works":["https://openalex.org/W12634471","https://openalex.org/W1849277567","https://openalex.org/W2097117768","https://openalex.org/W2112796928","https://openalex.org/W2117539524","https://openalex.org/W2183341477","https://openalex.org/W2194775991","https://openalex.org/W2302255633","https://openalex.org/W2531409750","https://openalex.org/W2549139847","https://openalex.org/W2550553598","https://openalex.org/W2550848904","https://openalex.org/W2752782242","https://openalex.org/W2765793020","https://openalex.org/W2792641098","https://openalex.org/W2928165649","https://openalex.org/W2949391068","https://openalex.org/W2962858109","https://openalex.org/W2963091558","https://openalex.org/W2963125010","https://openalex.org/W2963163009","https://openalex.org/W2963495494","https://openalex.org/W2964350391","https://openalex.org/W2981413347","https://openalex.org/W2988396473","https://openalex.org/W4229494842","https://openalex.org/W6600609147","https://openalex.org/W6630875275","https://openalex.org/W6637373629","https://openalex.org/W6638444622","https://openalex.org/W6684191040","https://openalex.org/W6696085341","https://openalex.org/W6696569547","https://openalex.org/W6730179637","https://openalex.org/W6737664043","https://openalex.org/W6739901393","https://openalex.org/W6753038380","https://openalex.org/W6753261499","https://openalex.org/W6753412334","https://openalex.org/W6753767121","https://openalex.org/W6753955284","https://openalex.org/W6756040250","https://openalex.org/W6762396321","https://openalex.org/W6763367864","https://openalex.org/W6769955919","https://openalex.org/W6771497439"],"related_works":[],"abstract_inverted_index":{"The":[0],"capability":[1],"of":[2,31,106,159,166,199],"the":[3,8,35,114,138,146,157,164,194,213],"self-attention":[4,21],"mechanism":[5,154],"to":[6,113,155],"model":[7],"long-range":[9],"dependencies":[10],"has":[11],"catapulted":[12],"its":[13],"deployment":[14],"in":[15,121,137,192,209],"vision":[16,122],"models.":[17,123],"Unlike":[18],"convolution":[19],"operators,":[20],"offers":[22],"infinite":[23],"receptive":[24],"field":[25],"and":[26,46,92,111,128,173,180,189,196,204,215,222],"enables":[27,90],"compute-":[28],"efficient":[29],"modeling":[30],"global":[32],"dependencies.":[33],"However,":[34],"existing":[36,115],"state-of-the-art":[37],"attention":[38,72,84,108,118,153],"mechanisms":[39,119],"incur":[40],"high":[41],"compute":[42],"and/or":[43],"parameter":[44,197],"overheads,":[45],"hence":[47],"unfit":[48],"for":[49,74,86,100],"compact":[50,140,160],"convolutional":[51],"neural":[52],"networks":[53],"(CNNs).":[54],"In":[55],"this":[56],"work,":[57],"we":[58,168],"propose":[59],"a":[60,133,151,202],"simple":[61],"yet":[62],"effective":[63],"\"Ultra-Lightweight":[64],"Subspace":[65],"Attention":[66],"Mechanism\"":[67],"(ULSAM),":[68],"which":[69,96],"infers":[70],"different":[71],"maps":[73,85],"each":[75,87],"feature":[76,88,94],"map":[77],"subspace.":[78],"We":[79,186],"argue":[80],"that":[81,149],"leaning":[82],"separate":[83],"subspace":[89,107,152],"multi-scale":[91],"multi-frequency":[93],"representation,":[95],"is":[97,109,125,145],"more":[98,205],"desirable":[99],"fine-grained":[101,182,216],"image":[102,183,217],"classification.":[103],"Our":[104],"method":[105],"orthogonal":[110],"complementary":[112],"state-of-the-":[116],"arts":[117],"used":[120],"ULSAM":[124],"end-to-end":[126],"trainable":[127],"can":[129],"be":[130],"deployed":[131],"as":[132,175],"plug-and-":[134],"play":[135],"module":[136],"pre-existing":[139],"CNNs.":[141,161],"Notably,":[142],"our":[143],"work":[144],"first":[147],"attempt":[148],"uses":[150],"increase":[156],"efficiency":[158],"To":[162],"show":[163],"efficacy":[165],"ULSAM,":[167],"perform":[169],"experiments":[170],"with":[171,201],"MobileNet-V1":[172],"MobileNet-V2":[174,200],"backbone":[176],"architectures":[177],"on":[178,212],"ImageNet-1K":[179,214],"three":[181],"classification":[184,218],"datasets.":[185],"achieve":[187],"\u224813%":[188],"\u224825%":[190],"reduction":[191],"both":[193],"FLOPs":[195],"counts":[198],"0.27%":[203],"than":[206],"1%":[207],"improvement":[208],"top-1":[210],"accuracy":[211],"datasets":[219],"(respectively).":[220],"Code":[221],"trained":[223],"models":[224],"are":[225],"available":[226],"at":[227],"https://github.com/Nandan91/ULSAM.":[228]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":22},{"year":2024,"cited_by_count":39},{"year":2023,"cited_by_count":19},{"year":2022,"cited_by_count":18},{"year":2021,"cited_by_count":8}],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2020-04-03T00:00:00"}