{"id":"https://openalex.org/W7134888996","doi":"https://doi.org/10.1109/asp-dac66049.2026.11420496","title":"BOA-3DGS: Backward-Striding Optimized Accelerator for Reduced Memory Contention in 3D Gaussian Splatting Training","display_name":"BOA-3DGS: Backward-Striding Optimized Accelerator for Reduced Memory Contention in 3D Gaussian Splatting Training","publication_year":2026,"publication_date":"2026-01-19","ids":{"openalex":"https://openalex.org/W7134888996","doi":"https://doi.org/10.1109/asp-dac66049.2026.11420496"},"language":null,"primary_location":{"id":"doi:10.1109/asp-dac66049.2026.11420496","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asp-dac66049.2026.11420496","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 31st Asia and South Pacific Design Automation Conference (ASP-DAC)","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/A5128702599","display_name":"Hyukjun Kweon","orcid":null},"institutions":[{"id":"https://openalex.org/I848706","display_name":"Sungkyunkwan University","ror":"https://ror.org/04q78tk20","country_code":"KR","type":"education","lineage":["https://openalex.org/I848706"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Hyukjun Kweon","raw_affiliation_strings":["Sungkyunkwan University,Semiconductor Convergence Engineering,Suwon,Korea"],"affiliations":[{"raw_affiliation_string":"Sungkyunkwan University,Semiconductor Convergence Engineering,Suwon,Korea","institution_ids":["https://openalex.org/I848706"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041027404","display_name":"Junsu Kim","orcid":"https://orcid.org/0000-0001-9583-7978"},"institutions":[{"id":"https://openalex.org/I848706","display_name":"Sungkyunkwan University","ror":"https://ror.org/04q78tk20","country_code":"KR","type":"education","lineage":["https://openalex.org/I848706"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jongyeop Kim","raw_affiliation_strings":["Sungkyunkwan University,Electrical and Computer Engineering,Suwon,Korea"],"affiliations":[{"raw_affiliation_string":"Sungkyunkwan University,Electrical and Computer Engineering,Suwon,Korea","institution_ids":["https://openalex.org/I848706"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5089174485","display_name":"Jeongwoo Park","orcid":"https://orcid.org/0000-0002-9603-9588"},"institutions":[{"id":"https://openalex.org/I848706","display_name":"Sungkyunkwan University","ror":"https://ror.org/04q78tk20","country_code":"KR","type":"education","lineage":["https://openalex.org/I848706"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jeongwoo Park","raw_affiliation_strings":["Sungkyunkwan University,Semiconductor Systems Engieering,Suwon,Korea"],"affiliations":[{"raw_affiliation_string":"Sungkyunkwan University,Semiconductor Systems Engieering,Suwon,Korea","institution_ids":["https://openalex.org/I848706"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5128702599"],"corresponding_institution_ids":["https://openalex.org/I848706"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.90149222,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"341","last_page":"347"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.06400000303983688,"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"}},"topics":[{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.06400000303983688,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.047200001776218414,"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/T14158","display_name":"Optical Systems and Laser Technology","score":0.03959999978542328,"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/training","display_name":"Training (meteorology)","score":0.5400999784469604},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.2897999882698059},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.2786000072956085},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.26269999146461487}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.71670001745224},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.5400999784469604},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4162999987602234},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3711000084877014},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.2897999882698059},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.2883000075817108},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.2786000072956085},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.26269999146461487},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2442999929189682},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.24330000579357147}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/asp-dac66049.2026.11420496","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asp-dac66049.2026.11420496","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 31st Asia and South Pacific Design Automation Conference (ASP-DAC)","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":16,"referenced_works":["https://openalex.org/W2021851106","https://openalex.org/W2293372129","https://openalex.org/W2738551266","https://openalex.org/W2901982540","https://openalex.org/W3109585842","https://openalex.org/W3133253223","https://openalex.org/W4385318467","https://openalex.org/W4393248012","https://openalex.org/W4395106423","https://openalex.org/W4400581064","https://openalex.org/W4402698355","https://openalex.org/W4402727865","https://openalex.org/W4404954627","https://openalex.org/W4407196862","https://openalex.org/W4409248634","https://openalex.org/W4409248733"],"related_works":[],"abstract_inverted_index":{"D":[0],"Gaussian":[1,59,73,87,120],"Splatting":[2,74,88],"(3DGS)":[3],"algorithms":[4],"have":[5],"gained":[6],"increasing":[7],"attention":[8],"due":[9,39,114],"to":[10,13,40,81,92,107,115,142,170,177],"their":[11],"ability":[12],"enable":[14],"realistic":[15],"3D":[16,72,86],"scene":[17],"reconstruction":[18],"with":[19],"faster":[20],"runtime.":[21],"However,":[22,102],"training":[23],"these":[24,149],"models":[25],"on":[26],"graphical":[27],"processing":[28,104],"units":[29],"(GPU)":[30],"faces":[31],"unique":[32],"challenges":[33],"during":[34,174],"the":[35,44,54,67,157,185],"backpropagation":[36,145,158,175],"stage,":[37],"primarily":[38],"memory":[41,112,140],"barriers":[42],"in":[43],"execution":[45],"model,":[46],"where":[47],"every":[48],"pixel":[49],"within":[50],"a":[51,57,76,108,125,132],"tile":[52],"computes":[53],"gradient":[55,117,134,152],"for":[56,71,100],"single":[58],"per":[60],"thread.":[61],"In":[62],"this":[63],"paper,":[64],"we":[65,137],"propose":[66],"Backward-Striding":[68],"Optimized":[69],"Accelerator":[70],"(BOA-3DGS),":[75],"hardware-software":[77],"co-optimized":[78],"accelerator":[79],"designed":[80],"optimize":[82],"backward":[83],"rasterization":[84],"of":[85,111,146,160],"by":[89],"enabling":[90],"pixels":[91],"stride":[93],"through":[94],"and":[95,119,131,155,187],"select":[96],"only":[97,182],"relevant":[98],"Gaussians":[99],"computing.":[101],"such":[103,139],"styles":[105],"lead":[106],"large":[109],"number":[110],"stalls":[113,141],"conflicting":[116],"storage":[118],"parameter":[121],"fetches.":[122],"By":[123],"introducing":[124],"pixel-independent,":[126],"funnel-like":[127],"multi-Gaussian":[128],"alpha":[129],"computation":[130],"majoritybased":[133],"accumulation":[135],"method,":[136],"avoid":[138],"efficiently":[143],"accelerate":[144,156],"3DGS.":[147],"Together,":[148],"enhancements":[150],"improve":[151],"calculation":[153],"efficiency":[154],"stage":[159],"3DGS":[161],"rasterization.":[162],"Experimental":[163],"results":[164],"demonstrate":[165],"that":[166],"BOA-3DGS":[167],"achieves":[168],"up":[169],"$1.58":[171],"\\times$":[172,184,190],"speedup":[173],"compared":[176],"prior":[178],"work,":[179],"while":[180],"utilizing":[181],"$0.86":[183],"area":[186],"consuming":[188],"$0.99":[189],"power.":[191]},"counts_by_year":[],"updated_date":"2026-03-13T14:20:09.374765","created_date":"2026-03-12T00:00:00"}