[{"data":1,"prerenderedAt":442},["ShallowReactive",2],{"project-49952111":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":29,"sentiment":30,"stance_mix":31,"scored_comments":36,"aspects":37,"strip":43,"post_text":45,"what":46,"said":93,"maker_in_thread":187,"shape":194,"grid":199,"attempts":404,"same_week":405,"maker_shows":32,"maker_hidden":434,"reader_version":435,"card":436},"49952111","SCM","AI search for every photo and every frame of video on macOS","title","Show HN: AI search for every photo and every frame of video on macOS","https:\u002F\u002Fgithub.com\u002Fallenv0\u002FSCM","github.com","allenleee",1791105892,2026,184,74,331,4366,8,"notable","ai","photo and video search","desktop_app",[24,25,26,27,28],"photo search","video frame search","clip embeddings","macos","ocr","curious",-0.093,{"positive":32,"neutral":33,"mixed":34,"critical":35},2,16,0,5,23,{"execution":38,"performance":39,"idea":40,"docs":42},{"positive":34,"critical":32},{"positive":34,"critical":34},{"positive":32,"critical":41},3,{"positive":34,"critical":34},[44,34,34,34,34,34,34,34,34,34,34,41,34,41,34,34,34,34,41,41,34,34,34,44,34,34,41,44,44,41,41,35,41,34,34,34,44,35,34,41],1,null,{"problem":47,"audience":45,"pricing":53,"motivation":60,"time_to_build":45,"stack":66,"openness":91,"stage":92,"team_size":92},{"value":48,"source":49,"evidence":50},"Searching photos and video frames by natural language content","inferred",[51],{"id":52},"post",{"value":54,"source":55,"evidence":56},"Free","maker",[57],{"id":58,"by":11,"text":59},"49956510","Thank you :) It's 100% free and open source!",{"value":61,"source":55,"evidence":62},"Keep v1 portable across platforms",[63],{"id":64,"by":11,"text":65},"49954873","Hey author here! :) Great point. You're totally right. Apple Vision is generally way faster and more accurate on Apple Silicon. Pure Mac-only, I'd use it. (also smaller) Main…",[67,71,75,79,83,87],{"value":68,"source":55,"evidence":69},"JavaScript",[70],{"id":64,"by":11,"text":65},{"value":72,"source":55,"evidence":73},"Transformers.js with ONNX",[74],{"id":64,"by":11,"text":65},{"value":76,"source":55,"evidence":77},"CLIP\u002FSigLIP",[78],{"id":64,"by":11,"text":65},{"value":80,"source":55,"evidence":81},"Whisper",[82],{"id":64,"by":11,"text":65},{"value":84,"source":55,"evidence":85},"Tesseract.js WASM",[86],{"id":64,"by":11,"text":65},{"value":88,"source":55,"evidence":89},"Node.js workers",[90],{"id":64,"by":11,"text":65},"open_source","unknown",{"praise":94,"objections":109,"requests":134,"questions":149,"alternatives":172},[95,103],{"value":96,"source":97,"evidence":98},"Useful for video editing","commenter",[99],{"id":100,"by":101,"text":102},"49954742","qlasisi15","This would really help in video editing. Thanks!",{"value":104,"source":55,"evidence":105},"Works well on M1\u002FM2\u002FM3 machines",[106],{"id":107,"by":11,"text":108},"49957065","YES! I can confirm it works perfectly on M1\u002FM2\u002FM3 machines (16g\u002F64g).",[110,117,127],{"value":111,"source":97,"evidence":112},"Tesseract is slower and less accurate than Apple Vision for OCR",[113],{"id":114,"by":115,"text":116},"49953995","postalcoder","Since this is for the mac you really should be using apple's vision framework for OCR. It smokes tesseract in both speed and accuracy. Edit: I'm curious which LLM was used to…",{"value":118,"source":97,"evidence":119},"Built with JS\u002FElectron, heavy on resources",[120,124],{"id":121,"by":122,"text":123},"49953989","cpursley","Why is this a JS bloatware instead of native or Rust which is easier than ever now with LLM coding tools.",{"id":125,"by":122,"text":126},"49954712","Fair enough, but seriously, my machine has taken a beating by all these JavaScript and electron thingies.",{"value":128,"source":97,"evidence":129},"Possibly vibe coded with edge case bugs",[130],{"id":131,"by":132,"text":133},"49955861","radicality","Not OP, but probably because they have no idea what they are doing or what CLIP even is. And the whole thing is likely just from one llm prompt, and then they dumped the whole…",[135,142],{"value":136,"source":97,"evidence":137},"Search for people, faces and pets",[138],{"id":139,"by":140,"text":141},"49953551","qprofyeh","This is cool. Any way to search for People \u002F faces \u002F pets? Like on iOS?",{"value":143,"source":97,"evidence":144},"Support for external network drives",[145],{"id":146,"by":147,"text":148},"49956036","measure2xcut1x","Thanks for suggesting this. I have not tried it because the photos are on an external SMB network drive.",[150,161],{"value":151,"source":97,"evidence":152},"How long does indexing take for long videos",[153,157],{"id":154,"by":155,"text":156},"49955241","doubleorseven","how long does it takes for a FHD 90 minutes asset?",{"id":158,"by":159,"text":160},"49953208","stephenitis","I like the entire premise, the one thing stopping me from trying this is not knowing the time scales that I will need to set my computer aside for the processing of large folders…",{"value":162,"source":97,"evidence":163},"Fixed interval or scene change frame sampling",[164,168],{"id":165,"by":166,"text":167},"49967952","tomveber","How do you pick which video frames to index, fixed interval or scene changes? Curious what an hour of footage costs in disk and indexing time.",{"id":169,"by":170,"text":171},"49953505","xnx","Have you tried scene detection? I would guess camera cuts are even less frequent than keyframes.",[173,180],{"value":174,"source":97,"evidence":175},"Apple Photos natural language search",[176],{"id":177,"by":178,"text":179},"49955776","tredre3","Apple Photos can already search photos with natural language using on-device AI. Is it not working well for you?",{"value":181,"source":97,"evidence":182},"Immich",[183],{"id":184,"by":185,"text":186},"49953541","lucideer","possibly off-topic, but for anyone interested in this on a more cross-platform \u002F holistic basis, Immich does this (& by \"this\" I mean an approximate AI search for photos & videos…",{"comments":35,"answered":32,"key":188},[189,190,193],{"id":64,"text":65},{"id":191,"text":192},"49961105","thanks for this! will do some research on it",{"id":107,"text":108},{"commenters":195,"max_depth":196,"maker_replies":35,"first_comment_mins":197,"lifespan_hours":198},57,6,166,68,[200,202,206,210,213,216,219,222,225,228,230,233,235,238,241,244,246,249,252,255,257,260,262,265,267,270,273,276,279,282,285,288,291,294,298,300,303,306,309,311,314,317,320,322,325,327,329,332,334,339,341,343,344,346,347,349,353,355,358,361,363,367,369,372,374,376,378,381,385,390,394,396,400,402],[114,34,44,201,115,34],"maker\u002Fcritical\u002Fexecution",[203,34,34,204,205,34],"49954259","other\u002Fneutral\u002Fnone","nullsanity",[207,34,34,208,209,34],"49958230","commenter\u002Fneutral\u002Fnone","fhub",[211,34,34,208,212,34],"49954344","daveguy",[214,34,34,208,215,34],"49954371","spiderfarmer",[217,34,34,208,218,34],"49954527","thih9",[220,34,34,208,221,34],"49954534","xp84",[223,34,34,208,224,34],"49954758","baxtr",[226,34,34,208,227,34],"49956441","Someone",[229,34,34,204,212,34],"49956549",[231,34,34,204,232,34],"49958457","TeMPOraL",[234,34,34,208,212,34],"49970610",[236,34,34,204,237,34],"49956291","losteric",[239,34,34,208,240,34],"49956986","WokeUp420",[242,34,34,208,243,34],"49963773","joshspankit",[245,34,34,208,215,34],"49954360",[247,34,34,204,248,34],"49954709","robotmay",[250,34,34,204,251,34],"49954854","vavkamil",[253,34,34,204,254,34],"49956884","sheept",[64,34,34,256,11,44],"maker\u002Fneutral\u002Fexecution",[258,34,41,256,259,34],"49957291","novok",[191,34,34,261,11,44],"maker\u002Fpositive\u002Fnone",[263,34,34,204,264,34],"49956215","darepublic",[266,34,34,204,240,34],"49956971",[268,34,41,256,269,34],"49957228","schainks",[271,34,34,204,272,34],"49958605","amelius",[274,34,34,208,275,34],"49962573","eecks",[277,44,34,204,278,34],"49953895","alt227",[280,44,34,208,281,34],"49953916","tough",[283,44,34,208,284,34],"49954196","seemaze",[286,44,34,208,287,34],"49956115","robotresearcher",[289,44,34,208,290,34],"49954385","101008",[292,44,34,208,293,34],"49954886","modzu",[295,32,41,296,297,34],"49953375","project\u002Fneutral\u002Fperformance","hn3ufz62f7",[169,32,41,299,170,34],"maker\u002Fneutral\u002Fperformance",[301,32,41,299,302,34],"49953700","pezgordo",[304,32,34,204,305,34],"49954192","Forgeties79",[307,32,34,208,308,34],"49959460","iAMkenough",[310,32,34,208,305,34],"49959801",[312,32,34,204,313,34],"49954038","mistrial9",[315,32,34,208,316,34],"49954094","Hnrobert42",[318,32,34,208,319,34],"49955368","whilenot-dev",[184,41,44,321,185,34],"project\u002Fcritical\u002Fidea",[323,41,34,204,324,34],"49955699","mannyv",[326,41,34,204,185,34],"49956472",[328,41,34,204,324,34],"49956640",[330,41,34,204,331,34],"49959197","ryandrake",[333,41,34,208,185,34],"49962337",[335,336,41,337,338,34],"49954002",4,"maker\u002Fneutral\u002Fidea","yt1998",[131,336,44,340,132,34],"project\u002Fcritical\u002Fexecution",[342,35,41,296,147,34],"49953634",[177,35,44,321,178,34],[146,35,41,345,147,34],"project\u002Fneutral\u002Fnone",[165,35,41,299,166,34],[107,35,34,348,11,44],"project\u002Fpositive\u002Fperformance",[350,196,41,351,352,34],"49988561","project\u002Fneutral\u002Fexecution","tomsonoda",[158,354,41,296,159,34],7,[356,18,41,337,357,34],"49984391","vinayamsnl",[100,359,35,360,101,34],9,"project\u002Fpositive\u002Fidea",[362,359,34,261,11,44],"49955066",[364,365,41,296,366,34],"49964094",10,"xcc3641",[121,368,34,208,122,34],11,[370,368,34,208,371,34],"49954337","collingreen",[125,368,34,373,122,34],"commenter\u002Fneutral\u002Fperformance",[375,368,34,204,371,34],"49957843",[377,368,34,204,331,34],"49959239",[139,379,41,380,140,34],12,"project\u002Fneutral\u002Fidea",[382,383,44,321,384,34],"49954994",13,"mannanj",[386,387,41,388,389,34],"49953452",14,"maker\u002Fneutral\u002Fdocs","aavisangle",[391,392,35,360,393,34],"49956292",15,"hemedanmert",[58,392,34,395,11,44],"project\u002Fpositive\u002Fpricing",[397,33,34,398,399,34],"49953806","other\u002Fneutral\u002Fidea","freecodeio",[401,33,34,398,371,34],"49954326",[154,403,41,296,155,34],17,[],[406,411,416,421,426,430],{"id":407,"name":408,"points":409,"category":410},"49994443","Bigwords.page",728,"productivity",{"id":412,"name":413,"points":414,"category":415},"49928566","stillwet.art",392,"creative",{"id":417,"name":418,"points":419,"category":420},"49930439","Glashütte Trash Clock",236,"hardware",{"id":422,"name":423,"points":424,"category":425},"49992057","Art From Pixels",174,"learning_science",{"id":427,"name":428,"points":429,"category":20},"49937916","ldraw-nova",170,{"id":431,"name":432,"points":433,"category":415},"49931031","Audionaut",155,false,"v2",{"id":437,"name":5,"line":6,"ai":434,"cat":20,"year":13,"points":14,"comments":15,"rec":29,"s":438,"top":18,"kw":439,"maker":11,"domain":10,"tier":19,"attempts":34,"strip":440,"url":9,"mix":441,"praise":96,"objection":111,"sub":21},49952111,-0.09,[24,25,26,27,28],"1000000000030300003300010031133530001503",[32,33,34,35],1794036711379]