[{"data":1,"prerenderedAt":256},["ShallowReactive",2],{"project-510264":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":27,"sentiment":28,"stance_mix":29,"scored_comments":32,"aspects":33,"strip":41,"post_text":43,"what":44,"said":74,"maker_in_thread":166,"shape":173,"grid":178,"attempts":242,"same_week":243,"maker_shows":16,"maker_hidden":248,"reader_version":249,"card":250},"510264","PittPatt face recognition demo","Our new online face recognition demo","title","Show HN: Our new online face recognition demo","http:\u002F\u002Fwebdemo.pittpatt.com\u002Frecognition_demo\u002F","webdemo.pittpatt.com","lbrandy",1236706380,2009,50,25,1,8,13,"notable","ai","face recognition","web_app",[21,24,25,26],"face detection","image matching","pose estimation","mixed",0.019,{"positive":30,"neutral":31,"mixed":31,"critical":30},3,4,14,{"performance":34,"execution":36,"idea":38,"docs":39,"privacy":40},{"positive":35,"critical":16},2,{"positive":37,"critical":37},0,{"positive":37,"critical":16},{"positive":37,"critical":16},{"positive":37,"critical":37},[42,37,30,37,37,37,37,42,37,30,37,30,30,16,30,37,37,16,37,37,16,30,30,42,30],5,null,{"problem":45,"audience":43,"pricing":43,"motivation":43,"time_to_build":43,"stack":56,"openness":71,"stage":72,"team_size":73},{"value":46,"source":47,"evidence":48},"Matches faces across two photos to tell whether they show the same person","maker",[49,52],{"id":50,"by":11,"text":51},"510321","I have a few comments for you guys about our demo: 1. Image to Image is a bit misleading because we are fairly strict in our online demo in pose requirements, so many faces are…",{"id":53,"by":54,"text":55},"510491","markessien","I totally misunderstood what this is. You recognize and match two faces across two pictures! I thought you just identify that a face is there. You REALLY need to explain a bit…",[57,63],{"value":58,"source":47,"evidence":59},"C",[60],{"id":61,"by":11,"text":62},"511642","It's a hodgepodge of low-level stuff. Some C. My favorite is the SSE-optimized assembly :)",{"value":64,"source":47,"evidence":65},"x86 assembly with SSE optimizations",[66,67],{"id":61,"by":11,"text":62},{"id":68,"by":69,"text":70},"513204","10ren","OK, I'm not familiar with MatLab performance. :-) But image processing is very computationally expensive, and the demo here is unbelievably fast. It must be optimized for its…","unknown","launched","company",{"praise":75,"objections":97,"requests":123,"questions":132,"alternatives":154},[76,93],{"value":77,"source":78,"evidence":79},"Detection and pose estimation impressive and fast","commenter",[80,84,85,89],{"id":81,"by":82,"text":83},"510548","ktharavaad","Very good demo and it also happens to be related to my interests so here's my little analysis. 1, The face detection of this program is more impressive than the face recognition.…",{"id":68,"by":69,"text":70},{"id":86,"by":87,"text":88},"510296","spydez","My first R&D project at my company when I just got out of college was in image recognition, so this is pretty interesting. Here's 2 I did:…",{"id":90,"by":91,"text":92},"510385","vaksel","Seems to work, good job.",{"value":94,"source":47,"evidence":95},"Video recognition works better than still images",[96],{"id":50,"by":11,"text":51},[98,106,110,116],{"value":99,"source":78,"evidence":100},"Frontal-pose requirement rejects many photos",[101,105],{"id":102,"by":103,"text":104},"511637","chime","It worked quite well for my face but not my wife's ( http:\u002F\u002Fwebdemo.pittpatt.com\u002Frecognition_demo\u002Fview.php?id=X3YPJH35U9579 ) because her head is slightly tilted and not facing…",{"id":50,"by":11,"text":51},{"value":107,"source":78,"evidence":108},"Unclear what the product does from the page",[109],{"id":53,"by":54,"text":55},{"value":111,"source":78,"evidence":112},"No download or access path for the technology",[113],{"id":114,"by":54,"text":115},"510508","So...how exactly does one go about getting a-hold of this technology? There is no download link and it's not clear if the contact link will result in my being sent a demo.",{"value":117,"source":78,"evidence":118},"Privacy concerns about tracking people",[119],{"id":120,"by":121,"text":122},"510556","spoiledtechie","Looks like something to give the government to track your EVERY MOVE. Good job tho guys. Its pretty impressive.",[124,128],{"value":125,"source":78,"evidence":126},"Relax photo pose requirements",[127],{"id":102,"by":103,"text":104},{"value":129,"source":78,"evidence":130},"Explain the product more clearly",[131],{"id":53,"by":54,"text":55},[133,143,150],{"value":134,"source":78,"evidence":135},"What implementation language is used",[136,139],{"id":137,"by":69,"text":138},"510500","Impressive. What's the implementation language? I assume C.",{"id":140,"by":141,"text":142},"511431","rlm","Why do you assume that? It could just as well be MatLab or something similar, that makes fiddling with multimedia very easy.",{"value":144,"source":78,"evidence":145},"Can templates be precomputed for faster comparison",[146],{"id":147,"by":148,"text":149},"511174","westside1506","Does this technology work using both images simultaneously? Or could lots of images be pre-processed individually into something that allowed a cheaper comparison? Just to try to…",{"value":151,"source":78,"evidence":152},"How to obtain the technology",[153],{"id":114,"by":54,"text":115},[155,159],{"value":156,"source":78,"evidence":157},"OpenCV",[158],{"id":81,"by":82,"text":83},{"value":160,"source":78,"evidence":161},"Facebook face tagging",[162],{"id":163,"by":164,"text":165},"510707","cool-RR","How is this better than what Facebook uses?",{"comments":30,"answered":35,"key":167},[168,169,170],{"id":50,"text":51},{"id":61,"text":62},{"id":171,"text":172},"511634","Recognition is generally broken down into two steps, processing faces into \"templates\" and then comparing those templates. Generating templates includes all the preprocessing…",{"commenters":174,"max_depth":175,"maker_replies":30,"first_comment_mins":176,"lifespan_hours":177},18,6,24,39,[179,181,183,184,186,189,191,193,194,195,197,198,200,203,205,207,210,214,216,221,224,228,231,234,237],[81,37,42,180,82,37],"project\u002Fpositive\u002Fperformance",[50,16,37,182,11,16],"maker\u002Fneutral\u002Fexecution",[137,16,30,182,69,37],[140,16,37,185,141,37],"commenter\u002Fneutral\u002Fnone",[187,16,37,185,188,37],"511519","Rod",[190,16,37,185,69,37],"511974",[192,16,37,185,141,37],"513174",[68,16,42,180,69,37],[61,16,37,182,11,16],[147,16,30,196,148,37],"maker\u002Fneutral\u002Fidea",[171,16,37,182,11,16],[102,16,30,199,103,37],"project\u002Fmixed\u002Fexecution",[201,35,30,202,69,37],"510477","project\u002Fneutral\u002Fperformance",[114,30,16,204,54,37],"project\u002Fcritical\u002Fidea",[86,31,30,206,87,37],"project\u002Fmixed\u002Fperformance",[208,31,37,185,209,37],"510315","lutorm",[211,42,37,212,213,37],"510545","other\u002Fpositive\u002Fnone","dbul",[53,175,16,215,54,37],"project\u002Fcritical\u002Fdocs",[217,218,37,219,220,37],"511207",7,"other\u002Fneutral\u002Fnone","allenbrunson",[222,218,37,219,223,37],"511403","brett",[225,17,16,226,227,37],"510328","project\u002Fcritical\u002Fperformance","dkokelley",[163,229,30,230,164,37],9,"project\u002Fneutral\u002Fidea",[120,232,30,233,121,37],10,"project\u002Fmixed\u002Fprivacy",[90,235,42,236,91,37],11,"project\u002Fpositive\u002Fnone",[238,239,30,240,241,37],"510977",12,"project\u002Fmixed\u002Fidea","thepanister",[],[244],{"id":245,"name":246,"points":35,"category":247},"512080","JStartup","productivity",false,"v2",{"id":251,"name":5,"line":6,"ai":248,"cat":20,"year":13,"points":14,"comments":15,"rec":27,"s":252,"top":18,"kw":253,"maker":11,"domain":10,"tier":19,"attempts":37,"strip":254,"url":9,"mix":255,"praise":77,"objection":99,"sub":21},510264,0.02,[21,24,25,26],"5030000503033130010013353",[30,31,31,30],1793990864102]