{"id":"https://openalex.org/W4286967710","doi":"https://doi.org/10.48550/arxiv.2109.11071","title":"Learning to Downsample for Segmentation of Ultra-High Resolution Images","display_name":"Learning to Downsample for Segmentation of Ultra-High Resolution Images","publication_year":2021,"publication_date":"2021-09-22","ids":{"openalex":"https://openalex.org/W4286967710","doi":"https://doi.org/10.48550/arxiv.2109.11071"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2109.11071","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2109.11071","pdf_url":"https://arxiv.org/pdf/2109.11071","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2109.11071","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Jin, Chen","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jin, Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064157732","display_name":"Ryutaro Tanno","orcid":"https://orcid.org/0000-0002-8107-6730","observed_orcids":["https://orcid.org/0000-0002-8107-6730"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tanno, Ryutaro","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091336286","display_name":"Thomy Mertzanidou","orcid":"https://orcid.org/0000-0001-9706-7399","observed_orcids":["https://orcid.org/0000-0001-9706-7399"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mertzanidou, Thomy","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003782562","display_name":"Eleftheria Panagiotaki","orcid":"https://orcid.org/0000-0002-2697-3447","observed_orcids":["https://orcid.org/0000-0002-2697-3447"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Panagiotaki, Eleftheria","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5033449704","display_name":"Daniel C. Alexander","orcid":"https://orcid.org/0000-0003-2439-350X","observed_orcids":["https://orcid.org/0000-0003-2439-350X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Alexander, Daniel C.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"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":26,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9955999851226807,"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":0.9955999851226807,"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/T10052","display_name":"Medical Image Segmentation Techniques","score":0.0017000000225380063,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.0007999999797903001,"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/image-segmentation","display_name":"image segmentation","score":0.843999981880188},{"id":"https://openalex.org/keywords/very-high-resolution-imagery","display_name":"very high resolution imagery","score":0.9610000252723694},{"id":"https://openalex.org/keywords/end-to-end-learning","display_name":"end-to-end learning","score":0.6549999713897705},{"id":"https://openalex.org/keywords/memory-efficient-segmentation","display_name":"memory-efficient segmentation","score":0.5260000228881836},{"id":"https://openalex.org/keywords/street-level-imagery","display_name":"street view images","score":0.7089999914169312},{"id":"https://openalex.org/keywords/aerial-images","display_name":"aerial images","score":0.9139999747276306},{"id":"https://openalex.org/keywords/medical-images","display_name":"medical images","score":0.7710000276565552},{"id":"https://openalex.org/keywords/accuracy-efficiency-trade-off","display_name":"accuracy-efficiency trade-off","score":0.8989999890327454}],"concepts":[{"id":"https://openalex.org/C110384440","wikidata":"https://www.wikidata.org/wiki/Q1143270","display_name":"Upsampling","level":3,"score":0.9846469759941101},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.7454358339309692},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7145041227340698},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7144978642463684},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.5695353746414185},{"id":"https://openalex.org/C160633673","wikidata":"https://www.wikidata.org/wiki/Q355198","display_name":"Pixel","level":2,"score":0.5493665337562561},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.5020203590393066},{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.49380049109458923},{"id":"https://openalex.org/C65885262","wikidata":"https://www.wikidata.org/wiki/Q7429708","display_name":"Scale-space segmentation","level":4,"score":0.4909117817878723},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.4515461325645447},{"id":"https://openalex.org/C124504099","wikidata":"https://www.wikidata.org/wiki/Q56933","display_name":"Image segmentation","level":3,"score":0.4419589936733246},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4240732789039612},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.36485159397125244},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.29525572061538696},{"id":"https://openalex.org/C106131492","wikidata":"https://www.wikidata.org/wiki/Q3072260","display_name":"Filter (signal processing)","level":2,"score":0.0645783543586731},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C43617362","wikidata":"https://www.wikidata.org/wiki/Q170050","display_name":"Chromatography","level":1,"score":0.0},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:arXiv.org:2109.11071","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2109.11071","pdf_url":"https://arxiv.org/pdf/2109.11071","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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"},{"id":"pmh:oai:eprints.ucl.ac.uk.OAI2:10167503","doi":null,"is_oa":false,"landing_page_url":"https://discovery.ucl.ac.uk/id/eprint/10167503/","pdf_url":null,"source":{"id":"https://openalex.org/S4306400024","display_name":"UCL Discovery (University College London)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I45129253","host_organization_name":"University College London","host_organization_lineage":["https://openalex.org/I45129253"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"In: Proceedings of the 10th International Conference on Learning Representations: ICLR 2022. ICLR 2022 (2022)","raw_type":"Proceedings paper"},{"id":"doi:10.48550/arxiv.2109.11071","doi":"https://doi.org/10.48550/arxiv.2109.11071","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2109.11071","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article-journal"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2109.11071","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2109.11071","pdf_url":"https://arxiv.org/pdf/2109.11071","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G2810537366","display_name":null,"funder_award_id":"EP/M020533/1","funder_id":"https://openalex.org/F4320334627","funder_display_name":"Engineering and Physical Sciences Research Council"},{"id":"https://openalex.org/G3953918217","display_name":null,"funder_award_id":"EP/R006032/1","funder_id":"https://openalex.org/F4320334627","funder_display_name":"Engineering and Physical Sciences Research Council"}],"funders":[{"id":"https://openalex.org/F4320334627","display_name":"Engineering and Physical Sciences Research Council","ror":"https://ror.org/0439y7842"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4394591959","https://openalex.org/W2981669902","https://openalex.org/W7213546408","https://openalex.org/W4406947678","https://openalex.org/W4410356951","https://openalex.org/W4386075929","https://openalex.org/W7138048617","https://openalex.org/W7134896262","https://openalex.org/W4313640889","https://openalex.org/W4322718150"],"abstract_inverted_index":{"Many":[0],"computer":[1],"vision":[2],"systems":[3],"require":[4],"low-cost":[5],"segmentation":[6,53,57,79,152],"algorithms":[7],"based":[8],"on":[9,156],"deep":[10],"learning,":[11],"either":[12],"because":[13,55],"of":[14,18,89,96,158,171],"the":[15,29,52,56,77,87,101,118,127,133,138,151],"enormous":[16],"size":[17],"input":[19,30,102],"images":[20,31,103,165],"or":[21],"limited":[22],"computational":[23],"budget.":[24],"Common":[25],"solutions":[26,112],"uniformly":[27],"downsample":[28],"to":[32,130,142,175],"meet":[33],"memory":[34],"constraints,":[35],"assuming":[36],"all":[37],"pixels":[38],"are":[39],"equally":[40],"informative.":[41],"In":[42],"this":[43,48,63],"work,":[44],"we":[45,120],"demonstrate":[46,166],"that":[47,125],"assumption":[49],"can":[50,72],"harm":[51],"performance":[54],"difficulty":[58],"varies":[59],"spatially.":[60],"We":[61,85,136],"combat":[62],"problem":[64,88],"by":[65],"introducing":[66],"a":[67,122],"learnable":[68],"downsampling":[69,92,139,178,183],"module,":[70],"which":[71],"be":[73],"optimised":[74],"together":[75],"with":[76],"given":[78,104],"model":[80],"in":[81,169],"an":[82],"end-to-end":[83],"fashion.":[84],"formulate":[86],"training":[90],"such":[91],"module":[93,140],"as":[94],"optimisation":[95],"sampling":[97,128],"density":[98],"distributions":[99],"over":[100],"their":[105],"low-resolution":[106],"views.":[107],"To":[108],"defend":[109],"against":[110],"degenerate":[111],"(e.g.":[113],"over-sampling":[114],"trivial":[115],"regions":[116],"like":[117],"backgrounds),":[119],"propose":[121],"regularisation":[123],"term":[124],"encourages":[126],"locations":[129],"concentrate":[131],"around":[132],"object":[134],"boundaries.":[135],"find":[137],"learns":[141],"sample":[143],"more":[144],"densely":[145],"at":[146],"difficult":[147],"locations,":[148],"thereby":[149],"improving":[150],"performance.":[153],"Our":[154],"experiments":[155],"benchmarks":[157],"high-resolution":[159],"street":[160],"view,":[161],"aerial":[162],"and":[163,179],"medical":[164],"substantial":[167],"improvements":[168],"terms":[170],"efficiency-and-accuracy":[172],"trade-off":[173],"compared":[174],"both":[176],"uniform":[177],"two":[180],"recent":[181],"advanced":[182],"techniques.":[184]},"counts_by_year":[{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":9},{"year":2023,"cited_by_count":8},{"year":2022,"cited_by_count":4}],"updated_date":"2026-10-16T09:59:22.235173","created_date":"2022-07-25T00:00:00"}