{"id":"https://openalex.org/W2802632470","doi":"https://doi.org/10.5075/epfl-thesis-8420","title":"Computational Aesthetics and Image Enhancements using Deep Neural Networks","display_name":"Computational Aesthetics and Image Enhancements using Deep Neural Networks","publication_year":2018,"publication_date":"2018-01-01","ids":{"openalex":"https://openalex.org/W2802632470","doi":"https://doi.org/10.5075/epfl-thesis-8420","mag":"2802632470"},"language":"en","primary_location":{"id":"pmh:oai:infoscience.epfl.ch:255088","is_oa":true,"landing_page_url":"https://infoscience.epfl.ch/handle/20.500.14299/146179","pdf_url":null,"source":{"id":"https://openalex.org/S4306400487","display_name":"Infoscience (Ecole Polytechnique F\u00e9d\u00e9rale de Lausanne)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"doctoral thesis"},"type":"dissertation","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://infoscience.epfl.ch/handle/20.500.14299/146179","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5016501019","display_name":"Bin Jin","orcid":"https://orcid.org/0000-0002-1411-2631"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Jin, Bin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5016501019"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"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/T11019","display_name":"Image Enhancement Techniques","score":0.9243999719619751,"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/T11019","display_name":"Image Enhancement Techniques","score":0.9243999719619751,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5370897650718689},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5191318392753601},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.48891010880470276},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.4602791965007782},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.43928828835487366},{"id":"https://openalex.org/keywords/aesthetics","display_name":"Aesthetics","score":0.4186486601829529},{"id":"https://openalex.org/keywords/philosophy","display_name":"Philosophy","score":0.2352316975593567}],"concepts":[{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5370897650718689},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5191318392753601},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.48891010880470276},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4602791965007782},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.43928828835487366},{"id":"https://openalex.org/C107038049","wikidata":"https://www.wikidata.org/wiki/Q35986","display_name":"Aesthetics","level":1,"score":0.4186486601829529},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.2352316975593567}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:infoscience.epfl.ch:255088","is_oa":true,"landing_page_url":"https://infoscience.epfl.ch/handle/20.500.14299/146179","pdf_url":null,"source":{"id":"https://openalex.org/S4306400487","display_name":"Infoscience (Ecole Polytechnique F\u00e9d\u00e9rale de Lausanne)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"doctoral thesis"},{"id":"doi:10.5075/epfl-thesis-8420","is_oa":true,"landing_page_url":"https://doi.org/10.5075/epfl-thesis-8420","pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Dissertation"},{"id":"mag:2802632470","is_oa":false,"landing_page_url":"https://infoscience.epfl.ch/record/255088","pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":null}],"best_oa_location":{"id":"pmh:oai:infoscience.epfl.ch:255088","is_oa":true,"landing_page_url":"https://infoscience.epfl.ch/handle/20.500.14299/146179","pdf_url":null,"source":{"id":"https://openalex.org/S4306400487","display_name":"Infoscience (Ecole Polytechnique F\u00e9d\u00e9rale de Lausanne)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"doctoral thesis"},"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/W1896080482","https://openalex.org/W3086135827","https://openalex.org/W2954216558","https://openalex.org/W3196590312","https://openalex.org/W2366958188","https://openalex.org/W2787914747","https://openalex.org/W3023330005","https://openalex.org/W3013121826","https://openalex.org/W2612918144","https://openalex.org/W2804407723","https://openalex.org/W2612034263","https://openalex.org/W2946229600","https://openalex.org/W2516576108","https://openalex.org/W2408979503","https://openalex.org/W2217895792","https://openalex.org/W2942917907","https://openalex.org/W3175574109","https://openalex.org/W3161321444","https://openalex.org/W3174240041","https://openalex.org/W2995932445"],"abstract_inverted_index":{"Imaging":[0],"devices":[1],"have":[2],"become":[3],"ubiquitous":[4],"in":[5,126,159,165,310,368],"modern":[6],"life,":[7],"and":[8,70,128,141,250,380,472],"many":[9],"of":[10,16,28,78,156,167,197,216,260,389,408,492],"us":[11],"capture":[12],"an":[13,236,247,449],"increasing":[14],"number":[15],"images":[17,433,468],"every":[18],"day.":[19],"When":[20],"we":[21,176,189,234,254,372,414,444],"choose":[22,37],"to":[23,36,86,105,178,184,204,269,285,292,316,322,335,347,384,469,475,504],"share":[24],"or":[25],"store":[26],"some":[27],"these":[29],"images,":[30],"our":[31,223],"primary":[32],"selection":[33],"criterion":[34],"is":[35,47,84,283,345],"the":[38,109,119,124,133,160,168,195,202,208,214,226,270,274,280,286,294,308,311,323,349,353,365,390,393,397,406,430,435,440,446,466,476,489,493,497,502],"most":[39,73],"visually":[40,318],"pleasing":[41,89,244],"ones.":[42],"Yet,":[43],"quantifying":[44],"visual":[45,66],"pleasantness":[46],"a":[48,75,96,138,143,148,191,230,251,257,299,481],"challenge,":[49],"as":[50,61,137],"image":[51,58,150,237,249,301,313,412,422,441,507],"aesthetics":[52,120,134,151,157,173,210,276,450,483,490],"not":[53,206],"only":[54,207],"correlate":[55],"with":[56,452],"low-level":[57],"quality,":[59,121],"such":[60,186,217],"contrast,":[62],"but":[63,212],"also":[64,213],"high-level":[65],"processes,":[67],"like":[68],"composition":[69],"context.":[71],"For":[72,305],"users,":[74],"considerable":[76],"amount":[77,259],"manual":[79],"effort":[80],"and/or":[81],"professional":[82],"knowledge":[83],"required":[85],"get":[87],"aesthetically":[88,243],"images.":[90,111,130,478,495],"Developing":[91],"automatic":[92,421],"solutions":[93],"thus":[94,200],"benefits":[95,123,363],"large":[97,149],"community.":[98],"This":[99],"thesis":[100],"proposes":[101],"several":[102],"computational":[103],"approaches":[104],"help":[106],"users":[107,125],"obtain":[108],"desired":[110],"The":[112,153,359,457],"first":[113,255,458],"technique":[114,239],"aims":[115,486],"at":[116,487],"automatically":[117,241],"measuring":[118],"which":[122,279],"selecting":[127],"ranking":[129],"We":[131,288,296,479],"form":[132],"prediction":[135],"problem":[136],"regression":[139],"task":[140],"train":[142,445],"deep":[144],"neural":[145],"network":[146,203,367],"on":[147,194,448],"dataset.":[152],"unbalanced":[154],"distribution":[155],"scores":[158,271],"training":[161,183,436],"set":[162],"can":[163],"result":[164],"bias":[166],"trained":[169],"model":[170,224],"towards":[171],"certain":[172,252],"levels.":[174],"Therefore,":[175],"propose":[177,235,480],"add":[179],"sample":[180],"weights":[181],"during":[182],"overcome":[185],"bias.":[187],"Moreover,":[188],"build":[190],"loss":[192,484],"function":[193],"histograms":[196],"user":[198],"labels,":[199],"enabling":[201],"predict":[205],"average":[209],"quality":[211,491],"difficulty":[215],"predictions.":[218],"Extensive":[219],"experiments":[220,291],"demonstrate":[221],"that":[222,240,464,485],"outperforms":[225],"previous":[227],"state-of-the-art":[228],"by":[229,273,326,396],"notable":[231],"margin.":[232],"Additionally,":[233],"cropping":[238],"outputs":[242],"crops.":[245,262],"Given":[246],"input":[248,312,477],"template,":[253],"extract":[256],"sufficient":[258],"candidate":[261],"These":[263],"crops":[264],"are":[265,314,357,401,425,460],"later":[266],"ranked":[267],"according":[268,321],"predicted":[272],"pre-trained":[275],"network,":[277],"after":[278],"best":[281],"crop":[282],"output":[284],"users.":[287,327],"conduct":[289],"psychophysical":[290],"validate":[293],"performance.":[295],"further":[297],"present":[298],"keyword-based":[300],"color":[302,332,354,360,381,398],"re-rendering":[303,333,355,361,399],"algorithm.":[304],"this":[306,337],"task,":[307],"colors":[309],"modified":[315],"be":[317,470],"more":[319,374],"appealing":[320],"keyword":[324],"specified":[325],"Our":[328],"algorithm":[329,344],"applies":[330],"local":[331],"operations":[334,356,400],"achieve":[336,373],"goal.":[338],"A":[339],"novel":[340],"weakly-supervised":[341],"semantic":[342],"segmentation":[343,366],"developed":[346],"locate":[348],"keyword-related":[350],"regions":[351],"where":[352],"applied.":[358],"process":[362],"from":[364,434],"two":[369,459],"aspects.":[370],"Firstly,":[371],"accurate":[375],"correlation":[376],"measurements":[377],"between":[378,432],"keywords":[379,409],"characteristics,":[382],"contributing":[383],"better":[385],"re-render":[386],"rendering":[387],"results":[388],"colors.":[391],"Secondly,":[392],"artifacts":[394],"caused":[395],"significantly":[402],"reduced.":[403],"To":[404,438],"avoid":[405],"need":[407],"when":[410],"enhancing":[411],"aesthetics,":[413],"explore":[415],"generative":[416,462],"adversarial":[417],"networks":[418],"(GANs)":[419],"for":[420,427],"enhancement.":[423],"GANs":[424,447,503],"known":[426],"directly":[428],"learning":[429],"transformations":[431],"data.":[437],"learn":[439],"enhancement":[442,508],"operations,":[443],"dataset":[451],"three":[453,498],"different":[454],"losses":[455,463,499],"combined.":[456],"standard":[461],"enforce":[465],"generated":[467,494],"natural":[471],"content-wise":[473],"similar":[474],"third":[482],"improving":[488],"Overall,":[496],"together":[500],"direct":[501],"apply":[505],"appropriate":[506],"operations.":[509]},"counts_by_year":[],"updated_date":"2026-07-23T08:03:31.855105","created_date":"2025-10-10T00:00:00"}