{"id":"https://openalex.org/W2889111331","doi":"https://doi.org/10.1145/3218603.3218615","title":"A Fully Onchip Binarized Convolutional Neural Network FPGA Impelmentation with Accurate Inference","display_name":"A Fully Onchip Binarized Convolutional Neural Network FPGA Impelmentation with Accurate Inference","publication_year":2018,"publication_date":"2018-07-23","ids":{"openalex":"https://openalex.org/W2889111331","doi":"https://doi.org/10.1145/3218603.3218615","mag":"2889111331"},"language":"en","primary_location":{"id":"doi:10.1145/3218603.3218615","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3218603.3218615","pdf_url":null,"source":{"id":"https://openalex.org/S4363608722","display_name":"Proceedings of the International Symposium on Low Power Electronics and Design","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Symposium on Low Power Electronics and Design","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/A5018588825","display_name":"Yang Li","orcid":"https://orcid.org/0000-0002-4268-4004"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li Yang","raw_affiliation_strings":["Department of Electrical and Computer Engineering, Orlando, Florida"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Orlando, Florida","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036755436","display_name":"Zhezhi He","orcid":"https://orcid.org/0000-0002-6357-236X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhezhi He","raw_affiliation_strings":["Department of Electrical and Computer Engineering, Orlando, Florida"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Orlando, Florida","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5047916979","display_name":"Deliang Fan","orcid":"https://orcid.org/0000-0002-7989-6297"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Deliang Fan","raw_affiliation_strings":["Department of Electrical and Computer Engineering, Orlando, Florida"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Orlando, Florida","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":37,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"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/T11992","display_name":"CCD and CMOS Imaging Sensors","score":0.9991999864578247,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic 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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9980999827384949,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7940295934677124},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.6696618795394897},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.6345556974411011},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6000015735626221},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.5327052474021912},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.4647940993309021},{"id":"https://openalex.org/keywords/convolution","display_name":"Convolution (computer science)","score":0.4334956705570221},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.43142980337142944},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.4309610426425934},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.41665178537368774},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.365885853767395},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.36083388328552246},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.33354389667510986},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.23771244287490845}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7940295934677124},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.6696618795394897},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.6345556974411011},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6000015735626221},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.5327052474021912},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.4647940993309021},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.4334956705570221},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.43142980337142944},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.4309610426425934},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.41665178537368774},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.365885853767395},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.36083388328552246},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.33354389667510986},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.23771244287490845},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.0},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3218603.3218615","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3218603.3218615","pdf_url":null,"source":{"id":"https://openalex.org/S4363608722","display_name":"Proceedings of the International Symposium on Low Power Electronics and Design","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Symposium on Low Power Electronics and Design","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.8600000143051147,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":20,"referenced_works":["https://openalex.org/W1836465849","https://openalex.org/W1902934009","https://openalex.org/W2194775991","https://openalex.org/W2242818861","https://openalex.org/W2300242332","https://openalex.org/W2319920447","https://openalex.org/W2335728318","https://openalex.org/W2469490737","https://openalex.org/W2565125333","https://openalex.org/W2585560244","https://openalex.org/W2604700561","https://openalex.org/W2618530766","https://openalex.org/W2753301142","https://openalex.org/W2762651727","https://openalex.org/W2963551763","https://openalex.org/W2964299589","https://openalex.org/W3102169921","https://openalex.org/W3143763605","https://openalex.org/W3147600416","https://openalex.org/W4234552385"],"related_works":["https://openalex.org/W2524802307","https://openalex.org/W2466675884","https://openalex.org/W2951390974","https://openalex.org/W2997828269","https://openalex.org/W2588201809","https://openalex.org/W2904851353","https://openalex.org/W2047588290","https://openalex.org/W2949189996","https://openalex.org/W3200060857","https://openalex.org/W3006085271"],"abstract_inverted_index":{"Deep":[0],"convolutional":[1],"neural":[2,59],"network":[3,155,164,180],"has":[4,14],"taken":[5],"an":[6],"important":[7],"role":[8],"in":[9,18,42,114,194],"machine":[10],"learning":[11],"algorithm":[12,41],"which":[13,83,107,169,222],"been":[15],"widely":[16],"used":[17],"computer":[19],"vision":[20],"tasks.":[21],"However,":[22],"its":[23],"enormous":[24],"model":[25,78],"size":[26,79],"and":[27,45,66,70,80,175,219,228],"massive":[28],"computation":[29,81],"cost":[30],"have":[31,55],"became":[32],"the":[33,57,77,109,141,153,162,173,225],"main":[34],"obstacle":[35],"for":[36,88],"deployment":[37],"of":[38],"such":[39,50],"powerful":[40],"low":[43],"power":[44],"resource":[46],"limited":[47],"embedded":[48],"system,":[49],"as":[51],"FPGA.":[52],"Recent":[53],"works":[54],"shown":[56],"binarized":[58,63],"networks":[60],"(BNN),":[61],"utilizing":[62],"(i.e.":[64,105],"+1":[65],"-1)":[67],"convolution":[68,112,121],"kernel":[69],"binary":[71,111,120],"activation":[72],"function,":[73],"can":[74],"significantly":[75],"reduce":[76,172],"complexity,":[82],"paves":[84],"a":[85,98,186],"new":[86,99,187],"road":[87],"energy-efficient":[89],"FPGA":[90,145,196,210],"implementation.":[91],"In":[92],"this":[93],"work,":[94],"we":[95],"first":[96],"propose":[97],"BNN":[100,104,116],"algorithm,":[101],"called":[102],"Parallel-Convolution":[103],"PC-BNN),":[106],"replaces":[108],"original":[110],"layer":[113],"conventional":[115],"with":[117,129,147],"two":[118],"parallel":[119],"layers.":[122],"PC-BNN":[123,139,195,208],"achieves":[124,212],"~86%":[125],"on":[126],"CIFAR-10":[127],"dataset":[128],"only":[130,148],"2.3Mb":[131],"parameter":[132,165,181],"size.":[133],"We":[134],"then":[135],"deploy":[136],"our":[137,207],"proposed":[138,193],"into":[140,166],"Xilinx":[142],"PYNQ":[143],"Z1":[144],"board":[146],"4.9Mb":[149],"on-chip":[150,167],"RAM.":[151],"Since":[152],"ultra-small":[154],"parameter,":[156],"it":[157],"is":[158,192],"feasible":[159],"to":[160,178,198,232],"store":[161],"whole":[163],"RAM,":[168],"could":[170],"greatly":[171],"energy":[174,229],"delay":[176],"overhead":[177],"load":[179],"from":[182],"off-chip":[183],"memory.":[184],"Meanwhile,":[185],"data":[188],"streaming":[189],"pipeline":[190],"architecture":[191],"implementation":[197,211],"further":[199],"improve":[200],"throughput.":[201],"The":[202],"experiment":[203],"results":[204],"show":[205],"that":[206],"based":[209],"930":[213],"frames":[214],"per":[215],"second,":[216],"387.5":[217],"FPS/Watt":[218],"396x10-4":[220],"FPS/LUT,":[221],"are":[223],"among":[224],"best":[226],"throughput":[227],"efficiency":[230],"compared":[231],"most":[233],"recent":[234],"works.":[235]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":3},{"year":2022,"cited_by_count":5},{"year":2021,"cited_by_count":10},{"year":2020,"cited_by_count":6},{"year":2019,"cited_by_count":6},{"year":2018,"cited_by_count":1},{"year":2017,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}