[{"data":1,"prerenderedAt":225},["ShallowReactive",2],{"project-14409255":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":10,"maker":11,"posted_at":12,"year":13,"points":14,"comments":15,"rank_in_year":16,"shown_in_year":17,"top":18,"tier":19,"category":20,"subcategory":21,"format":22,"keywords":23,"reception":28,"sentiment":29,"stance_mix":30,"scored_comments":34,"aspects":35,"strip":39,"post_text":41,"what":42,"said":53,"maker_in_thread":134,"shape":143,"grid":148,"attempts":187,"same_week":188,"maker_shows":217,"maker_hidden":218,"reader_version":219,"card":220},"14409255","fontjoy","Choosing font combinations with deep learning","title","Show HN: Choosing font combinations with deep learning","https:\u002F\u002Fgithub.com\u002FJack000\u002Ffontjoy","github.com","Jack000",1495626377,2017,121,18,350,2242,16,"notable","creative","font pairing tool","web_app",[24,25,26,27],"font pairing","font combinations","deep learning","font design","debated",0.05,{"positive":31,"neutral":32,"mixed":33,"critical":31},3,2,0,8,{"idea":36,"execution":38},{"positive":37,"critical":32},1,{"positive":32,"critical":37},[33,33,33,33,33,37,31,37,31,33,40,33,40,33,33,33,37,40],5,null,{"problem":41,"audience":41,"pricing":41,"motivation":41,"time_to_build":41,"stack":43,"openness":51,"stage":52,"team_size":52},[44],{"value":45,"source":46,"evidence":47},"VGG16 for font feature extraction","maker",[48],{"id":49,"by":11,"text":50},"14412487","it uses VGG16 to extract font features, then compares those features to figure out the best contrast between fonts. feel free to use the vectors in the Github.","open_source","unknown",{"praise":54,"objections":71,"requests":94,"questions":106,"alternatives":114},[55],{"value":56,"source":57,"evidence":58},"Font pairings look nice and are decent most of the time","commenter",[59,63,67],{"id":60,"by":61,"text":62},"14416077","rubatuga","Although a title with deep learning in it makes me roll my eyes nowadays, I really should be more open to the ideas behind them. The font pairings at the end of the article look…",{"id":64,"by":65,"text":66},"14417641","accountyaccount","This does an alright job most of the time (7\u002F10 options are decent). That's impressive.",{"id":68,"by":69,"text":70},"14411605","justswim","This is pretty awesome. I've been working on my side project Mixfont (https:\u002F\u002Fwww.mixfont.com) which is a similar sort of font generator. Your pairings are a lot nicer than those…",[72,79,90],{"value":73,"source":57,"evidence":74},"Results are often poor",[75],{"id":76,"by":77,"text":78},"14417863","usaphp","Honestly most of the results are really poor. I would recommend using https:\u002F\u002Fwww.typewolf.com instead.",{"value":80,"source":57,"evidence":81},"ML curation favors similar results and lacks novelty",[82,86],{"id":83,"by":84,"text":85},"14417219","Declanomous","One of my biggest complaints about things 'curated' via ML is that you are only introduced to things that are relatively similar. It's like trying to find new music by looking at…",{"id":87,"by":88,"text":89},"14416400","intoverflow2","Interesting, but not all design is about harmony. I'd be more interested in examples where ML could create combinations that shouldn't work together but work. Think like some of…",{"value":91,"source":57,"evidence":92},"Not all design is about harmony",[93],{"id":87,"by":88,"text":89},[95,102],{"value":96,"source":57,"evidence":97},"Generate fonts for character sets not in the original font, such as Hindi or Cyrillic",[98],{"id":99,"by":100,"text":101},"14415308","LeonidBugaev","I wonder, is it possible to generate fonts based on this algorithm, for character sets which are not included into the original font. For example, I have English font, and want…",{"value":103,"source":57,"evidence":104},"Find unconventional combinations that shouldn't work but do",[105],{"id":87,"by":88,"text":89},[107],{"value":108,"source":57,"evidence":109},"How to identify a font from an image",[110],{"id":111,"by":112,"text":113},"14415012","ams6110","Tangentially, I've been trying to identify a font. I have a small example as a jpeg image. I tried https:\u002F\u002Fwww.myfonts.com\u002FWhatTheFont\u002F without finding an exact match. Are there…",[115,119,123,127],{"value":116,"source":57,"evidence":117},"typewolf",[118],{"id":76,"by":77,"text":78},{"value":120,"source":57,"evidence":121},"Mixfont",[122],{"id":68,"by":69,"text":70},{"value":124,"source":57,"evidence":125},"WhatTheFont",[126],{"id":111,"by":112,"text":113},{"value":128,"source":57,"evidence":129},"identifont",[130],{"id":131,"by":132,"text":133},"14415142","teraflop","You might want to try http:\u002F\u002Fwww.identifont.com\u002F.",{"comments":31,"answered":32,"key":135},[136,139,142],{"id":137,"text":138},"14415029","I suspect the tough part of whatthefont isn't so much the visual search but having a large database of labelled fonts. There are other services that does this but personally I've…",{"id":140,"text":141},"14415322","there's another project that does this: https:\u002F\u002Ferikbern.com\u002F2016\u002F01\u002F21\u002Fanalyzing-50k-fonts-using-deep-neural-networks.html the results are a bit blurry but as the author notes…",{"id":49,"text":50},{"commenters":144,"max_depth":145,"maker_replies":31,"first_comment_mins":146,"lifespan_hours":147},15,4,327,30,[149,151,152,153,156,159,161,165,166,167,169,171,173,175,178,181,183,185],[111,33,33,150,112,33],"other\u002Fneutral\u002Fnone",[137,33,33,150,11,37],[131,33,33,150,132,33],[154,33,33,150,155,33],"14416367","jimhefferon",[157,33,33,150,158,33],"14419027","theoh",[87,37,37,160,88,33],"project\u002Fcritical\u002Fidea",[162,37,31,163,164,33],"14417145","project\u002Fneutral\u002Fidea","danso",[83,37,37,160,84,33],[99,32,31,163,100,33],[140,32,33,168,11,37],"project\u002Fneutral\u002Fexecution",[68,31,40,170,69,33],"project\u002Fpositive\u002Fexecution",[49,31,33,172,11,37],"maker\u002Fneutral\u002Fexecution",[60,145,40,174,61,33],"project\u002Fpositive\u002Fidea",[176,145,33,150,177,33],"14416317","throwanem",[179,145,33,150,180,33],"14417672","gjm11",[182,145,33,150,177,33],"14417978",[76,40,37,184,77,33],"project\u002Fcritical\u002Fexecution",[64,186,40,170,65,33],6,[],[189,194,199,204,209,213],{"id":190,"name":191,"points":192,"category":193},"14409760","Notational",257,"productivity",{"id":195,"name":196,"points":197,"category":198},"14411328","Funky Karts",251,"games",{"id":200,"name":201,"points":202,"category":203},"14402738","Triplebyte quiz",227,"learning_science",{"id":205,"name":206,"points":207,"category":208},"14409553","browseemall.com",129,"languages",{"id":210,"name":211,"points":212,"category":208},"14395087","wireworld-nim",115,{"id":214,"name":215,"points":216,"category":20},"14376821","mjmdavis.com",109,11,false,"v2",{"id":221,"name":5,"line":6,"ai":218,"cat":20,"year":13,"points":14,"comments":15,"rec":28,"s":29,"top":18,"kw":222,"maker":11,"domain":10,"tier":19,"attempts":33,"strip":223,"url":9,"mix":224,"praise":56,"objection":73,"sub":21},14409255,[24,25,26,27],"000001313050500015",[31,32,33,31],1794036555507]