{"id":"https://openalex.org/W4393578137","doi":"https://doi.org/10.1145/3626202.3637582","title":"A Flexible, Fast, Low Bandwidth Block-based Acceleration Architecture for CNN Inference on FPGAs","display_name":"A Flexible, Fast, Low Bandwidth Block-based Acceleration Architecture for CNN Inference on FPGAs","publication_year":2024,"publication_date":"2024-04-01","ids":{"openalex":"https://openalex.org/W4393578137","doi":"https://doi.org/10.1145/3626202.3637582"},"language":"en","primary_location":{"id":"doi:10.1145/3626202.3637582","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3626202.3637582","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2024 ACM/SIGDA International Symposium on Field Programmable Gate Arrays","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/A5086306222","display_name":"Yan Chen","orcid":"https://orcid.org/0000-0002-2660-2419"},"institutions":[{"id":"https://openalex.org/I177738480","display_name":"Japan Advanced Institute of Science and Technology","ror":"https://ror.org/03frj4r98","country_code":"JP","type":"education","lineage":["https://openalex.org/I177738480"]}],"countries":["JP"],"is_corresponding":true,"raw_author_name":"Yan Chen","raw_affiliation_strings":["Japan Advanced Institute of Science and Technology, Nomi, Ishikawa, Japan"],"affiliations":[{"raw_affiliation_string":"Japan Advanced Institute of Science and Technology, Nomi, Ishikawa, Japan","institution_ids":["https://openalex.org/I177738480"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5010030610","display_name":"Kiyofumi Tanaka","orcid":"https://orcid.org/0000-0003-0148-9250"},"institutions":[{"id":"https://openalex.org/I177738480","display_name":"Japan Advanced Institute of Science and Technology","ror":"https://ror.org/03frj4r98","country_code":"JP","type":"education","lineage":["https://openalex.org/I177738480"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Kiyofumi Tanaka","raw_affiliation_strings":["Japan Advanced Institute of Science and Technology, Nomi, Ishikawa, Japan"],"affiliations":[{"raw_affiliation_string":"Japan Advanced Institute of Science and Technology, Nomi, Ishikawa, Japan","institution_ids":["https://openalex.org/I177738480"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5086306222"],"corresponding_institution_ids":["https://openalex.org/I177738480"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.03616664,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"183","last_page":"183"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","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/T10036","display_name":"Advanced Neural Network Applications","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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9973000288009644,"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":0.9944999814033508,"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/computer-science","display_name":"Computer science","score":0.8154932260513306},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.7601151466369629},{"id":"https://openalex.org/keywords/dataflow","display_name":"Dataflow","score":0.6567498445510864},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.6516770124435425},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.6357793211936951},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"Memory bandwidth","score":0.5791570544242859},{"id":"https://openalex.org/keywords/chip","display_name":"Chip","score":0.44280073046684265},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.4292975962162018},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.42518073320388794},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.4076972007751465},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.40258386731147766},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.3972225785255432},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.16589799523353577},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.11765047907829285}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8154932260513306},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.7601151466369629},{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.6567498445510864},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.6516770124435425},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.6357793211936951},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.5791570544242859},{"id":"https://openalex.org/C165005293","wikidata":"https://www.wikidata.org/wiki/Q1074500","display_name":"Chip","level":2,"score":0.44280073046684265},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.4292975962162018},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.42518073320388794},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.4076972007751465},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.40258386731147766},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3972225785255432},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.16589799523353577},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.11765047907829285},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0},{"id":"https://openalex.org/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3626202.3637582","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3626202.3637582","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2024 ACM/SIGDA International Symposium on Field Programmable Gate Arrays","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2293118914","https://openalex.org/W2999668243","https://openalex.org/W1998888015","https://openalex.org/W1967938402","https://openalex.org/W2386041993","https://openalex.org/W1608572506","https://openalex.org/W2077105843","https://openalex.org/W2770465587","https://openalex.org/W2086716781","https://openalex.org/W2022510519"],"abstract_inverted_index":{"In":[0,33],"recent":[1],"years,":[2],"Convolutional":[3],"Neural":[4],"Networks":[5],"(CNNs)":[6],"have":[7],"been":[8],"widely":[9],"used":[10],"in":[11],"many":[12],"fields.":[13],"Many":[14],"accelerators":[15,44],"are":[16,25,157,176],"designed":[17],"to":[18,45,73,93],"run":[19,94],"CNN":[20,42,76],"inference":[21],"on":[22,66,126,145,151],"hardware.":[23],"They":[24],"classified":[26],"into":[27],"two":[28],"types:":[29],"Overlay":[30],"and":[31,56,78,131,148,168],"Dataflow.":[32],"this":[34],"paper,":[35],"we":[36],"propose":[37],"a":[38,90,127],"new":[39,68],"architecture":[40,69,103],"of":[41,49,111,155,174],"Inference":[43],"leverage":[46],"the":[47,75,132,160,164,169],"strengths":[48],"each":[50],"type.":[51],"It":[52,81,142],"is":[53,137],"flexible,":[54],"fast,":[55],"has":[57,89],"low":[58,139],"off-chip":[59,133,170],"memory":[60,134,171],"bandwidth":[61,135,172],"requirements.":[62],"The":[63,153],"accelerator":[64],"based":[65],"our":[67],"can":[70],"be":[71],"configured":[72],"fit":[74],"structure":[77],"FPGA":[79,129,166],"capacity.":[80],"basically":[82],"runs":[83,144],"CNNs":[84],"by":[85],"blocks,":[86],"whereas":[87],"it":[88],"fallback":[91],"path":[92],"single":[95],"Convolution":[96],"or":[97],"Fully":[98],"Connected":[99],"layers":[100],"efficiently.":[101],"Our":[102],"demonstrates":[104],"significant":[105],"advantages":[106],"when":[107,121],"running":[108,122],"networks":[109],"composed":[110],"inverted":[112],"residual":[113],"blocks.":[114],"We":[115],"archived":[116,149],"1954FPS":[117],"(frames":[118],"per":[119],"second)":[120],"8-bit":[123],"quantized":[124],"MobileNetV2":[125],"mid-range":[128],"ZU7EV,":[130],"requirement":[136,173],"as":[138,140],"2.47GiB/s.":[141],"also":[143],"cost-optimized":[146],"FPGAs,":[147],"505FPS":[150],"ZU3EG.":[152],"throughputs":[154],"ours":[156,175],"higher":[158],"than":[159],"other":[161],"types,":[162],"while":[163],"minimum":[165],"capacity":[167],"reasonable.":[177]},"counts_by_year":[],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}