[{"data":1,"prerenderedAt":490},["ShallowReactive",2],{"project-15383336":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":35,"aspects":36,"strip":44,"post_text":46,"what":47,"said":88,"maker_in_thread":187,"shape":200,"grid":205,"attempts":450,"same_week":455,"maker_shows":45,"maker_hidden":482,"reader_version":483,"card":484},"15383336","fast-style-transfer-deeplearnjs","Paint yourself in the style of classic art, in your browser","title","Show HN: Paint yourself in the style of classic art, in your browser","https:\u002F\u002Fgithub.com\u002Freiinakano\u002Ffast-style-transfer-deeplearnjs\u002F","github.com","reiinakano",1506948072,2017,266,99,89,2242,4,"legend","creative","style transfer","web_app",[21,24,25,26,27],"neural style","deeplearn.js","webgl","fast neural style","roasted",-0.407,{"positive":31,"neutral":32,"mixed":33,"critical":34},6,11,3,27,47,{"design":37,"idea":40,"execution":41,"performance":43},{"positive":38,"critical":39},1,0,{"positive":18,"critical":33},{"positive":39,"critical":42},24,{"positive":39,"critical":39},[45,38,45,33,39,39,39,45,33,33,38,38,38,33,38,33,33,38,38,38,38,38,33,45,39,33,33,38,38,38,38,38,39,38,33,38,38,38,38,38],5,null,{"problem":48,"audience":46,"pricing":46,"motivation":58,"time_to_build":64,"stack":68,"openness":86,"stage":87,"team_size":86},{"value":49,"source":50,"evidence":51},"Applies classic art styles to user photos without server-side processing","maker",[52,55],{"id":53,"by":11,"text":54},"15384495","Fast Neural Style Transfer works like this: Each style image is actually trained on a neural network outside of the browser (using Python on Tensorflow) for around 6+ hours with…",{"id":56,"by":11,"text":57},"15383442","Demo website: https:\u002F\u002Freiinakano.github.io\u002Ffast-style-transfer-deeplearnjs Deeplearn.JS library doesn't work on mobile yet, sorry to folks on the go!",{"value":59,"source":50,"evidence":60},"Quick proof-of-concept demo of deeplearn.js",[61],{"id":62,"by":11,"text":63},"15384521","It is currently VERY slow since there are a lot of non-optimized (non-GPU) calls being made. Once those are implemented in deeplearn.JS, things should run much faster. In fact,…",{"value":65,"source":50,"evidence":66},"About a week",[67],{"id":62,"by":11,"text":63},[69,74,78],{"value":25,"source":50,"evidence":70},[71],{"id":72,"by":11,"text":73},"15385317","Didn't really use any of their code. I reimplemented their model in a new library called deeplearn.JS. If the issue is indeed with the library, there's little I can do. I will…",{"value":75,"source":50,"evidence":76},"TensorFlow (model training)",[77],{"id":53,"by":11,"text":54},{"value":79,"source":80,"evidence":81},"JavaScript","inferred",[82],{"id":83,"by":84,"text":85},"15383650","sabeltann","Uncaught ReferenceError: dialogPolyfill is not defined at HTMLDocument.\u003Canonymous> ((index):22672) at (index):24","unknown","prototype",{"praise":89,"objections":102,"requests":147,"questions":159,"alternatives":171},[90],{"value":91,"source":92,"evidence":93},"Impressive results and cool effect","commenter",[94,98],{"id":95,"by":96,"text":97},"15384222","TekMol","Cool. I'm getting some great results. Would be awesome if you could train your own model in the browser too. By the way: How does a NN learn from a single image? I understand how…",{"id":99,"by":100,"text":101},"15383873","davedx","Weirdly, something about the generated images (colour range?) breaks Twitter's profile photo update. I actually took screenshots of the original taken with my webcam and the deep…",[103,114,125,136],{"value":104,"source":92,"evidence":105},"Very slow in browser and hangs on large images",[106,110],{"id":107,"by":108,"text":109},"15384484","earlybike","tl;dr: Impressive Prisma clone with six filters but running all calculations locally in the browser. Compared to Prisma it feels slightly slower or as slow, one filter almost…",{"id":111,"by":112,"text":113},"15384211","rsoto","Hey, it looks very cool. However, it's hanging my browser. Haven't looked at the source, but maybe this should be implemented with web workers to increase perceived performance.…",{"value":115,"source":92,"evidence":116},"Outputs look blurry or like nightmares on some setups",[117,121],{"id":118,"by":119,"text":120},"15384863","hnlmorg","I don't know if it's my machine or whether my expectations are too high, but every generated image looks like the same blurry version of the original but with a different colour…",{"id":122,"by":123,"text":124},"15384864","Belphemur","Am I the only one that get \"the stuff of nightmares\" kind of results ? Ubuntu 17.04 FF 57 & Chrome 59 with face and Udnie:…",{"value":126,"source":92,"evidence":127},"Does not work on mobile",[128,132],{"id":129,"by":130,"text":131},"15384193","code-is-code","Looks like it does not work on mobile chrome",{"id":133,"by":134,"text":135},"15392604","lerie82","I don't know if it's loading or what but it's just blank on my phone's chrome browser.",{"value":137,"source":92,"evidence":138},"Seen as imitation rather than art",[139,143],{"id":140,"by":141,"text":142},"15386926","michrassena","When I see the style transfers, I'm reminded of the early days of Photoshop, and whatever else was a commonplace photo editing application of the mid-90s (like Fractal Painter).…",{"id":144,"by":145,"text":146},"15391741","TheOtherHobbes","It still looks like Poor Photoshop Art Imitation++ though. Outside of the developer community, no one cares how this works. They only care about whether or not they connect with…",[148,155],{"value":149,"source":92,"evidence":150},"Support mobile devices",[151],{"id":152,"by":153,"text":154},"15384489","make3","you should likely put a warning message for mobile users to let them know the website doesn't work on mobile",{"value":156,"source":92,"evidence":157},"Train your own model in the browser",[158],{"id":95,"by":96,"text":97},[160,164],{"value":161,"source":92,"evidence":162},"How does a network learn style from a single image",[163],{"id":95,"by":96,"text":97},{"value":165,"source":92,"evidence":166},"What size images can it handle",[167],{"id":168,"by":169,"text":170},"15384023","everdev","What size images can it generate? I've used other libraries that can't go past 800x600.",[172,176,183],{"value":173,"source":92,"evidence":174},"Prisma",[175],{"id":107,"by":108,"text":109},{"value":177,"source":92,"evidence":178},"Deep Art Effects",[179],{"id":180,"by":181,"text":182},"15392658","wowtip","Is it using the same tech as android\u002Fios app Deep Art Effects? https:\u002F\u002Fwww.deeparteffects.com\u002F Results look very similar",{"value":184,"source":92,"evidence":185},"Photoshop filters",[186],{"id":140,"by":141,"text":142},{"comments":188,"answered":189,"key":190},34,13,[191,194,197],{"id":192,"text":193},"15384969","You're lengstrom from Github! I must let you know your models are exactly what I used! I don't have a beefy GPU so I was forced to use what you already had, haha. For everyone…",{"id":195,"text":196},"15392499","Fast Neural Style transfer is a variation of a \"regular\" Neural Style algorithm. With regular Neural Style, you can do style transfer between any two arbitrary images. The…",{"id":198,"text":199},"15389687","Copying from nature with tiny little variations I guess?",{"commenters":201,"max_depth":202,"maker_replies":188,"first_comment_mins":203,"lifespan_hours":204},49,8,22,26,[206,210,212,216,218,221,224,228,232,234,236,239,242,245,246,248,251,254,256,259,262,263,265,269,271,274,276,279,281,282,284,286,289,291,293,295,298,300,304,307,309,311,313,314,315,317,319,323,326,328,330,333,335,338,340,342,344,346,349,350,351,354,357,361,363,365,367,370,371,374,377,380,382,384,386,389,392,395,397,398,400,404,406,408,412,414,416,420,422,426,428,430,432,434,436,440,443,445,448],[207,39,45,208,209,39],"15384949","project\u002Fpositive\u002Fdesign","logane",[192,39,39,211,11,38],"commenter\u002Fneutral\u002Fnone",[213,39,39,214,215,39],"15387572","other\u002Fneutral\u002Fnone","selljamhere",[140,38,38,217,141,39],"project\u002Fcritical\u002Fidea",[219,38,39,211,220,39],"15387872","jordanwallwork",[222,38,39,211,223,39],"15387928","math0ne",[225,38,45,226,227,39],"15388110","project\u002Fpositive\u002Fidea","themcaffee",[229,38,45,230,231,39],"15391733","project\u002Fpositive\u002Fnone","h2rra",[233,38,38,217,141,39],"15392436",[195,38,39,235,11,38],"project\u002Fneutral\u002Fidea",[237,38,39,211,238,39],"15388766","robbrown451",[240,38,39,211,241,39],"15388912","SimbaOnSteroids",[243,38,39,211,244,39],"15389627","musage",[198,38,39,214,11,38],[247,38,39,211,244,39],"15390932",[249,38,39,214,250,39],"15388946","vacri",[252,38,39,214,253,39],"15389186","tbabb",[255,38,45,226,238,39],"15390767",[257,38,39,214,258,39],"15389912","ak93",[260,38,45,226,261,39],"15391556","JacobiX",[144,38,38,217,145,39],[264,38,39,226,11,38],"15391943",[266,38,33,267,268,39],"15392096","project\u002Fmixed\u002Fexecution","notahacker",[270,38,33,235,261,39],"15392277",[118,272,38,273,119,39],2,"project\u002Fcritical\u002Fexecution",[275,272,38,273,119,39],"15384987",[277,272,39,278,11,38],"15385178","project\u002Fneutral\u002Fexecution",[280,272,38,273,119,39],"15385268",[72,272,39,278,11,38],[283,272,39,278,11,38],"15392541",[56,33,39,285,11,38],"project\u002Fneutral\u002Fnone",[287,33,38,273,288,39],"15383933","Beltiras",[290,33,39,278,11,38],"15383952",[292,33,33,278,288,39],"15383979",[294,33,39,278,11,38],"15385219",[296,33,38,273,297,39],"15384287","iaml",[299,33,39,278,11,38],"15384649",[301,18,33,302,303,39],"15385836","project\u002Fneutral\u002Fperformance","gwbas1c",[305,18,33,302,306,39],"15386111","wrigby",[308,18,39,211,11,38],"15386168",[310,18,39,214,306,39],"15386256",[107,45,33,312,108,39],"project\u002Fmixed\u002Fperformance",[62,45,39,302,11,38],[99,31,38,273,100,39],[316,31,39,278,11,38],"15383946",[318,31,39,278,11,38],"15384259",[320,321,38,273,322,39],"15384275",7,"chias",[324,321,38,273,325,39],"15384361","supermdguy",[327,321,33,278,322,39],"15384436",[329,321,33,278,325,39],"15384467",[331,321,38,273,332,39],"15384393","etrevino",[334,321,39,278,11,38],"15384428",[336,321,38,273,337,39],"15384643","BugsJustFindMe",[339,321,39,278,11,38],"15384676",[341,321,33,278,337,39],"15384723",[343,321,39,278,11,38],"15384796",[345,321,39,214,337,39],"15385398",[347,321,39,348,11,38],"15385531","commenter\u002Fpositive\u002Fnone",[95,202,45,226,96,39],[53,202,39,235,11,38],[352,202,39,353,96,39],"15384693","commenter\u002Fneutral\u002Fidea",[355,202,39,356,11,38],"15384856","maker\u002Fneutral\u002Fdocs",[358,202,39,359,360,39],"15386175","other\u002Fneutral\u002Fidea","SamBam",[180,362,33,235,181,39],9,[364,362,39,235,11,38],"15392702",[111,366,33,312,112,39],10,[368,366,39,369,11,38],"15384627","maker\u002Fneutral\u002Fperformance",[83,32,38,273,84,39],[372,32,39,373,11,38],"15383680","maker\u002Fneutral\u002Fexecution",[375,32,39,376,84,39],"15383693","commenter\u002Fneutral\u002Fexecution",[378,32,38,273,379,39],"15383839","davecardwell",[381,32,39,373,11,38],"15383918",[383,32,39,373,11,38],"15384022",[122,385,38,273,123,39],12,[387,385,38,273,388,39],"15385290","dabber",[390,385,33,278,391,39],"15387623","whataretensors",[393,385,38,273,394,39],"15391931","oever",[396,385,39,373,11,38],"15392548",[133,189,38,273,134,39],[399,189,39,373,11,38],"15392615",[401,402,39,214,403,39],"15391767",14,"mercurysmessage",[168,405,33,302,169,39],15,[407,405,39,369,11,38],"15384045",[409,410,38,273,411,39],"15384187",16,"fiatjaf",[413,410,39,373,11,38],"15384284",[415,410,33,278,411,39],"15386337",[417,418,38,273,419,39],"15384077",17,"reacweb",[421,418,39,373,11,38],"15384108",[423,424,38,273,425,39],"15384782",18,"samstave",[427,424,39,356,11,38],"15384788",[129,429,38,273,130,39],19,[431,429,39,373,11,38],"15384278",[433,429,38,273,130,39],"15386909",[152,435,38,273,153,39],20,[437,438,38,273,439,39],"15386735",21,"laythea",[441,438,39,211,442,39],"15386755","TheRealPomax",[444,438,39,211,439,39],"15386966",[446,203,38,273,447,39],"15384318","sergiotapia",[449,203,39,373,11,38],"15384631",[451,454],{"id":452,"t":453,"points":33,"comments":38},"15372761",1506778036,{"id":4,"t":12,"points":14,"comments":15},[456,460,465,469,473,477],{"id":457,"name":458,"points":459,"category":20},"15393446","Thousand Ether Homepage",344,{"id":461,"name":462,"points":463,"category":464},"15357394","RMStudio",154,"dev_tools",{"id":466,"name":467,"points":468,"category":464},"15367645","ARKit Playground",151,{"id":470,"name":471,"points":472,"category":464},"15386981","InstaBuild",124,{"id":474,"name":475,"points":476,"category":464},"15356599","Sitebulb",120,{"id":478,"name":479,"points":480,"category":481},"15392777","Nematoduino",101,"hardware",false,"v2",{"id":485,"name":5,"line":6,"ai":482,"cat":20,"year":13,"points":14,"comments":15,"rec":28,"s":486,"top":18,"kw":487,"maker":11,"domain":10,"tier":19,"attempts":272,"strip":488,"url":9,"mix":489,"praise":91,"objection":104,"sub":21},15383336,-0.41,[21,24,25,26,27],"5153000533111313311111350331111101311111",[31,32,33,34],1793990964655]