{"id":"https://openalex.org/W4404102542","doi":"https://doi.org/10.1109/tvlsi.2024.3488042","title":"Falcon: A Fused-Layer Accelerator With Layer-Wise Hybrid Inference Flow for Computational Imaging CNNs","display_name":"Falcon: A Fused-Layer Accelerator With Layer-Wise Hybrid Inference Flow for Computational Imaging CNNs","publication_year":2024,"publication_date":"2024-11-06","ids":{"openalex":"https://openalex.org/W4404102542","doi":"https://doi.org/10.1109/tvlsi.2024.3488042"},"language":"en","primary_location":{"id":"doi:10.1109/tvlsi.2024.3488042","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tvlsi.2024.3488042","pdf_url":null,"source":{"id":"https://openalex.org/S37538908","display_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","issn_l":"1063-8210","issn":["1063-8210","1557-9999"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["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 Very Large Scale Integration (VLSI) Systems","raw_type":"journal-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/A5066309622","display_name":"Yongtai Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I25846049","display_name":"National Tsing Hua University","ror":"https://ror.org/00zdnkx70","country_code":"TW","type":"education","lineage":["https://openalex.org/I25846049"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Yong-Tai Chen","raw_affiliation_strings":["Department of Electrical Engineering, National Tsing Hua University, Hsinchu, Taiwan"],"raw_orcid":"https://orcid.org/0009-0003-9174-6958","affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering, National Tsing Hua University, Hsinchu, Taiwan","institution_ids":["https://openalex.org/I25846049"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060901073","display_name":"Yen-Ting Chiu","orcid":null},"institutions":[{"id":"https://openalex.org/I25846049","display_name":"National Tsing Hua University","ror":"https://ror.org/00zdnkx70","country_code":"TW","type":"education","lineage":["https://openalex.org/I25846049"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Yen-Ting Chiu","raw_affiliation_strings":["Department of Electrical Engineering, National Tsing Hua University, Hsinchu, Taiwan"],"raw_orcid":"https://orcid.org/0009-0007-5319-0804","affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering, National Tsing Hua University, Hsinchu, Taiwan","institution_ids":["https://openalex.org/I25846049"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101269016","display_name":"Hao-Jiun Tu","orcid":null},"institutions":[{"id":"https://openalex.org/I25846049","display_name":"National Tsing Hua University","ror":"https://ror.org/00zdnkx70","country_code":"TW","type":"education","lineage":["https://openalex.org/I25846049"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Hao-Jiun Tu","raw_affiliation_strings":["Department of Electrical Engineering, National Tsing Hua University, Hsinchu, Taiwan"],"raw_orcid":"https://orcid.org/0009-0000-6804-1202","affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering, National Tsing Hua University, Hsinchu, Taiwan","institution_ids":["https://openalex.org/I25846049"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5042036635","display_name":"Chao-Tsung Huang","orcid":"https://orcid.org/0000-0002-9173-520X"},"institutions":[{"id":"https://openalex.org/I25846049","display_name":"National Tsing Hua University","ror":"https://ror.org/00zdnkx70","country_code":"TW","type":"education","lineage":["https://openalex.org/I25846049"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Chao-Tsung Huang","raw_affiliation_strings":["Department of Electrical Engineering, National Tsing Hua University, Hsinchu, Taiwan"],"raw_orcid":"https://orcid.org/0000-0002-9173-520X","affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering, National Tsing Hua University, Hsinchu, Taiwan","institution_ids":["https://openalex.org/I25846049"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I25846049"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.14097833,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"33","issue":"3","first_page":"720","last_page":"732"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9739000201225281,"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":0.9739000201225281,"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/T10320","display_name":"Neural Networks and Applications","score":0.9448000192642212,"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/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9377999901771545,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.6355562806129456},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5738710165023804},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5080056190490723},{"id":"https://openalex.org/keywords/flow","display_name":"Flow (mathematics)","score":0.41975945234298706},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3441472053527832},{"id":"https://openalex.org/keywords/nanotechnology","display_name":"Nanotechnology","score":0.19960111379623413},{"id":"https://openalex.org/keywords/materials-science","display_name":"Materials science","score":0.19376686215400696},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.18461614847183228},{"id":"https://openalex.org/keywords/mechanics","display_name":"Mechanics","score":0.10461804270744324}],"concepts":[{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.6355562806129456},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5738710165023804},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5080056190490723},{"id":"https://openalex.org/C38349280","wikidata":"https://www.wikidata.org/wiki/Q1434290","display_name":"Flow (mathematics)","level":2,"score":0.41975945234298706},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3441472053527832},{"id":"https://openalex.org/C171250308","wikidata":"https://www.wikidata.org/wiki/Q11468","display_name":"Nanotechnology","level":1,"score":0.19960111379623413},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.19376686215400696},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.18461614847183228},{"id":"https://openalex.org/C57879066","wikidata":"https://www.wikidata.org/wiki/Q41217","display_name":"Mechanics","level":1,"score":0.10461804270744324}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tvlsi.2024.3488042","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tvlsi.2024.3488042","pdf_url":null,"source":{"id":"https://openalex.org/S37538908","display_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","issn_l":"1063-8210","issn":["1063-8210","1557-9999"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["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 Very Large Scale Integration (VLSI) Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320331164","display_name":"National Science and Technology Council","ror":"https://ror.org/00wnb9798"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":35,"referenced_works":["https://openalex.org/W2242218935","https://openalex.org/W2508457857","https://openalex.org/W2607202125","https://openalex.org/W2764207251","https://openalex.org/W2917604311","https://openalex.org/W2921126577","https://openalex.org/W2933438941","https://openalex.org/W2962793481","https://openalex.org/W2963372104","https://openalex.org/W2963470893","https://openalex.org/W2966350814","https://openalex.org/W2966432437","https://openalex.org/W2979590285","https://openalex.org/W3006586535","https://openalex.org/W3007788310","https://openalex.org/W3015391117","https://openalex.org/W3047237654","https://openalex.org/W3118490733","https://openalex.org/W3160243752","https://openalex.org/W3202040256","https://openalex.org/W3205546585","https://openalex.org/W4205146836","https://openalex.org/W4214870144","https://openalex.org/W4225672218","https://openalex.org/W4249932213","https://openalex.org/W4293102247","https://openalex.org/W4308089796","https://openalex.org/W4312641363","https://openalex.org/W4312646647","https://openalex.org/W4312812783","https://openalex.org/W4360605929","https://openalex.org/W4360831791","https://openalex.org/W4380874786","https://openalex.org/W6777205017","https://openalex.org/W6810313840"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Computational":[0],"imaging":[1],"(CI)":[2],"has":[3],"advanced":[4],"significantly":[5],"due":[6],"to":[7,22,84,124,171,214],"the":[8,24,34,48,53,78,183,198,216,235],"use":[9],"of":[10,30,36,139,191,218],"convolutional":[11],"neural":[12],"networks":[13],"(CNNs).":[14],"Its":[15],"edge":[16],"deployment":[17],"relies":[18],"on":[19,46],"layer":[20],"fusion":[21],"offload":[23],"monstrous":[25],"external":[26],"memory":[27],"access":[28,170],"(EMA)":[29],"feature":[31,172],"maps,":[32],"necessitating":[33],"handling":[35],"overlapped":[37],"features":[38],"either":[39],"through":[40],"reusing":[41],"or":[42],"recomputing":[43],"them.":[44],"Depending":[45],"how":[47],"boundary-handling":[49],"strategy":[50,76],"is":[51],"organized,":[52],"induced":[54],"computing":[55,109],"complexity":[56],"and":[57,87,105,115,128,145,204,226],"EMA":[58],"can":[59],"be":[60],"optimized.":[61],"However,":[62],"state-of-the-art":[63],"CI":[64],"accelerators":[65],"primarily":[66],"apply":[67],"homogeneous":[68,232],"inference":[69,103,121,233],"flows,":[70,234],"which":[71],"employ":[72],"a":[73,107,118,151,162],"single":[74],"overlap-handling":[75],"throughout":[77],"fused":[79],"layers,":[80],"limiting":[81],"their":[82,126],"ability":[83],"balance":[85],"computation":[86],"data":[88],"access.":[89],"In":[90],"this":[91],"article,":[92],"we":[93,129,149],"explore":[94],"layer-wise":[95,113,119],"optimization":[96,132],"in":[97,175,220],"fused-layer":[98],"CNNs":[99],"by":[100,201,207],"exploiting":[101],"hybrid-strategy":[102],"flows":[104],"devising":[106],"corresponding":[108],"architecture.":[110],"We":[111],"categorize":[112],"strategies":[114],"put":[116],"forward":[117],"hybrid":[120],"flow":[122],"(LHIF)":[123],"integrate":[125],"advantages,":[127],"propose":[130],"an":[131,176],"procedure":[133],"that":[134,167,182],"explicitly":[135],"analyzes":[136],"essential":[137],"figures":[138],"merit":[140],"(FoMs),":[141],"including":[142],"throughput,":[143],"EMA,":[144],"energy":[146],"efficiency.":[147],"Furthermore,":[148],"develop":[150],"high-throughput":[152],"accelerator\u2014Falcon\u2014to":[153],"efficiently":[154],"support":[155],"LHIF":[156,195,219,237],"under":[157],"massive":[158],"parallelism,":[159],"especially":[160],"with":[161,188,230,241],"time-division-multiplexing":[163],"(TDM)":[164],"buffer":[165],"interface":[166],"enables":[168],"seamless":[169],"maps":[173],"stored":[174],"interleaved":[177],"manner.":[178],"Layout":[179],"results":[180],"show":[181],"accelerator,":[184],"delivering":[185],"41":[186],"TOPS":[187],"1.5":[189],"MB":[190],"feature-map":[192],"buffers,":[193],"supports":[194],"while":[196],"increasing":[197],"die":[199],"area":[200],"only":[202,208],"1.4%":[203],"power":[205],"consumption":[206],"0.7%.":[209],"Extensive":[210],"simulations":[211],"are":[212],"conducted":[213],"demonstrate":[215],"versatility":[217],"working":[221],"scenarios":[222],"at":[223],"operational,":[224],"design,":[225],"system":[227],"levels.":[228],"Compared":[229],"using":[231],"proposed":[236],"achieves":[238],"Pareto":[239],"optimality":[240],"up":[242],"to$2.28\\times":[243],"$higher":[244],"throughput":[245],"and$3.5\\times":[246],"$lower":[247],"EMA.":[248]},"counts_by_year":[],"updated_date":"2026-07-29T09:40:50.615796","created_date":"2025-10-10T00:00:00"}