[{"data":1,"prerenderedAt":186},["ShallowReactive",2],{"project-17029430":3},{"id":4,"name":5,"line":5,"line_source":6,"title":7,"url":8,"domain":9,"maker":10,"posted_at":11,"year":12,"points":13,"comments":14,"rank_in_year":15,"shown_in_year":16,"top":17,"tier":18,"category":19,"subcategory":20,"format":21,"keywords":22,"reception":28,"sentiment":29,"stance_mix":30,"scored_comments":34,"aspects":35,"strip":39,"post_text":41,"what":42,"said":61,"maker_in_thread":114,"shape":119,"grid":122,"attempts":147,"same_week":148,"maker_shows":33,"maker_hidden":178,"reader_version":179,"card":180},"17029430","Offline Object Detection and Tracking on a Raspberry Pi","title","Show HN: Offline Object Detection and Tracking on a Raspberry Pi","https:\u002F\u002Fmedium.com\u002Fml-everything\u002Foffline-object-detection-and-tracking-on-a-raspberry-pi-fddb3bde130","medium.com","bko",1525870989,2018,106,12,409,2155,19,"notable","ai","computer vision on embedded devices","other",[23,24,25,26,27],"object detection","raspberry pi","offline inference","object tracking","tensorflow","curious",-0.133,{"positive":31,"neutral":32,"mixed":31,"critical":33},0,4,1,5,{"docs":36,"performance":37,"execution":38},{"positive":31,"critical":31},{"positive":31,"critical":33},{"positive":31,"critical":31},[40,31,31,31,33,40,31,31,31,40,40,31],3,null,{"problem":41,"audience":41,"pricing":41,"motivation":41,"time_to_build":41,"stack":43,"openness":60,"stage":60,"team_size":60},[44,56],{"value":45,"source":46,"evidence":47},"Python","commenter",[48,52],{"id":49,"by":50,"text":51},"17030713","ythn","Why not just install all the python 3 dependencies to a docker image and then run the docker container on the pi?",{"id":53,"by":54,"text":55},"17030180","IloveHN84","Why python and not c++?",{"value":57,"source":46,"evidence":58},"Docker",[59],{"id":49,"by":50,"text":51},"unknown",{"praise":62,"objections":63,"requests":71,"questions":79,"alternatives":88},[],[64],{"value":65,"source":46,"evidence":66},"Performance may be poor on Raspberry Pi Zero due to ARMv6 architecture",[67],{"id":68,"by":69,"text":70},"17031292","MarkMMullin","You want to proceed carefully with a Pi Zero - the chip is an ARMv6, so you'll end up building more things and you loose some optimization - you might not like the performance :-(",[72],{"value":73,"source":46,"evidence":74},"Self-contained way to run trained models on a Pi without many Python dependencies",[75],{"id":76,"by":77,"text":78},"17030098","dividuum","Is there a more self-contained way of running trained models on devices like the Pi? Almost all tutorials I've come across always require me to install a lot of Python…",[80],{"value":81,"source":46,"evidence":82},"Why Python and not C++?",[83,84],{"id":53,"by":54,"text":55},{"id":85,"by":86,"text":87},"17030362","icebraining","Why C++ and not Python?",[89,93,100,107],{"value":90,"source":46,"evidence":91},"Darknet\u002FYOLO",[92],{"id":76,"by":77,"text":78},{"value":94,"source":46,"evidence":95},"Flogo",[96],{"id":97,"by":98,"text":99},"17030254","zok3102","Not exactly a single binary, but Flogo - https:\u002F\u002Fgithub.com\u002FTIBCOSoftware\u002Fflogo uses one binary + one .so for TF inferencing. See…",{"value":101,"source":46,"evidence":102},"Resin.io",[103],{"id":104,"by":105,"text":106},"17032763","brittohalloran","Did you look at Resin OS at all (resin.io)? No affiliation. Ship a container to your Pi that you can develop on your laptop and then have built for the ARM architecture on…",{"value":108,"source":46,"evidence":109},"TensorFlow.js",[110],{"id":111,"by":112,"text":113},"17032400","Maybestring","You could use tensorflow.js, not self-contained, but it only depends on the browser. And... Is it easy to get openCL working on a Pi? WebGL might be slower, but it ought to just…",{"comments":33,"answered":33,"key":115},[116],{"id":117,"text":118},"17030781","This was actually my original approach. I meant to mention that. And I got it to work but then I had to figure out how to control the hardware from the pi and I just gave up. To…",{"commenters":120,"max_depth":40,"maker_replies":33,"first_comment_mins":121,"lifespan_hours":14},11,87,[123,125,127,130,131,133,136,138,139,142,144,145],[76,31,40,124,77,31],"project\u002Fneutral\u002Fdocs",[97,31,31,126,98,31],"other\u002Fneutral\u002Fnone",[128,31,31,126,129,31],"17032172","sarthakjain",[111,31,31,126,112,31],[68,33,33,132,69,31],"project\u002Fcritical\u002Fperformance",[49,134,40,135,50,31],2,"project\u002Fneutral\u002Fexecution",[117,134,31,137,10,33],"maker\u002Fneutral\u002Fexecution",[104,134,31,126,105,31],[140,134,31,126,141,31],"17035736","joshvm",[143,134,40,135,86,31],"17030877",[53,40,40,135,54,31],[85,40,31,146,86,31],"commenter\u002Fneutral\u002Fnone",[],[149,154,159,164,169,173],{"id":150,"name":151,"points":152,"category":153},"17015043","Idyll guitar tuning post",338,"learning_science",{"id":155,"name":156,"points":157,"category":158},"17004898","Morning Cup of Coding",192,"productivity",{"id":160,"name":161,"points":162,"category":163},"16985250","LayerJS",176,"languages",{"id":165,"name":166,"points":167,"category":168},"17033574","A Native Art Gallery for the Mac",146,"creative",{"id":170,"name":171,"points":172,"category":158},"17002648","new-note",133,{"id":174,"name":175,"points":176,"category":177},"17008147","Stadia Maps",105,"web_saas",false,"v2",{"id":181,"name":5,"line":5,"ai":178,"cat":19,"year":12,"points":13,"comments":14,"rec":28,"s":182,"top":17,"kw":183,"maker":10,"domain":9,"tier":18,"attempts":31,"strip":184,"url":8,"mix":185,"praise":-1,"objection":65,"sub":20},17029430,-0.13,[23,24,25,26,27],"300013000330",[31,32,31,33],1794036565866]