{"id":1388,"date":"2009-12-07T18:58:36","date_gmt":"2009-12-07T09:58:36","guid":{"rendered":"https:\/\/regex.info\/blog\/2009-12-07\/1388"},"modified":"2009-12-07T18:58:36","modified_gmt":"2009-12-07T09:58:36","slug":"giving-up-on-picasa-face-recognition","status":"publish","type":"post","link":"https:\/\/regex.info\/blog\/2009-12-07\/1388","title":{"rendered":"Giving Up on Picasa<\/i> Face Recognition"},"content":{"rendered":"\n\n\n
\nNOTE<\/b>: Images with an icon next to them have been artificially shrunk to better fit your screen; click the icon to restore them, in place, to their regular size.\n<\/div>\n\n\n

I've just wasted the last two days on something more fun (and more\nfrustrating) than any video game – Google's Picasa<\/a> photo app “face recognition”\nstuff.<\/p>\n\n

Google's Picasa<\/i> desktop photo application includes the very cool\nfeature of face recognition. Point it at a photo<\/span> and it will identify\nwhere in the photo faces are to be found, and do so with great precision. I pointed<\/span> Picasa at the 20,000 photos in my archive from this year (maybe\n100,000 faces?) and there were no false positives, unless you don't count\nsculptures of\nfaces<\/a> as a face.<\/span> Very impressive.<\/p>\n\n

But like a modern Star Wars film (lots of cool effects, but no plot or\nacting or anything else that might contribute to making a good<\/i>\nmovie), the end result was disappointing because of a clumsy<\/span> UI, poor\nrecognition of who the face might be, and worst of all, a decidedly<\/span> random\napproach to remembering what faces I've said belong to whom.<\/p>\n\n

It took about a day and a half of crunching in the background for Picasa\nto chew on the 20,000 photos I threw<\/span> it at, after which it grouped all the\nfaces by person (as best it could figure), letting me ascribe names to the\ngroups, and to fix errors in the groups. The application<\/span> UI offers several\ndifferent ways to approach the task of assigning names to faces and\nvice-versa, each with their pros and cons. Unfortunately, all the cons are\nsevere. In one mode<\/span> you can easily group photos so that the faces are all\nof the same few people, but it offers no way to group the results so you\ncan indicate “these 20 faces are all such-and-such”; you actually have to\nactually type their name 20 times.<\/p>\n\n

Another mode might allow you to indicate that a face<\/span> should be ignored\n(e.g. some random stranger in the background of a shot<\/span>), but other\notherwise-more-useful modes don't offer that important feature, so if you\nsee such a face<\/span> while looking at suggestions for one person, the best you\ncan do is indicate “not this person”, and so you're guaranteed to see the\nface again (and again and again) until you come across the face in the mode\nthat actually allows you to ignore it.<\/p>\n\n

Some modes make it very easy to say “these faces are\/aren't this\nperson”, but they make it very difficult to say “these faces\nare this other<\/i> person.”<\/p>\n\n

So, it becomes a pattern of jumping around modes to use each mode for\nwhat it seems best for, and you'd see the same faces over and over as you\nsaid “no, not this person either”, until you got fed up enough with it that\nyou visited a mode<\/span> where you could<\/i> indicate who it was.<\/p>\n\n

It was oddly addicting. It sucked<\/span> me in like nothing else has since I\nmade<\/span> the mistake of giving Magic\nPen<\/a> a try<\/span> a few<\/span> years ago. I worked<\/span> on the faces all day today, despite\nhaving promised myself a dozen<\/span> different times that “I'll stop after\ntidying up this last person's faces.”<\/p>\n\n

I should point out two very important caveats to my complaints above:\n1)<\/b> face recognition is new to Picasa (and to the world), so it's\nonly natural that the UI will be a bit<\/span> rough. It's a “version\n1” application, so should be given time to mature before one can\nreally expect a smooth<\/span> UI. And 2)<\/b>, Google provides Picasa for\nfree<\/i>. Complainers should tread lightly.<\/p>\n\n

Anyway, in the end I did abandon Picasa, and won't be returning to it\nany time soon. I'd certainly<\/span> love to have all my photos “name\ntagged” (so to speak), but Picasa seems to have a very<\/span> nasty habit of\nrandomizing the data. I might<\/span> mark 20 faces as being of the same someone,\nthen return in five minutes to see that only one of those faces is there,\nbut 47 other random faces are there as well, marked\n“confirmed”. This happened time and time again. At first<\/span> I\nthought<\/span> that I must<\/span> simply be doing something wrong, but careful testing\ntoward the end convinced me that it's just random, so no matter how cool\nthe initial “where are the faces?” technology is, everything\nelse as implemented in Picasa is not at all ready for prime time.<\/p>\n\n

I built a plugin for Lightroom to suck over the name information from\nPicasa, but I'm now much less excited to release \/ maintain it (UPDATE:<\/b> but\nnevertheless, I have<\/span>\nreleased it here<\/a>). It'll be interesting to see what Google does with\nthis in future version — I don't<\/span> doubt great things from them —\nbut for now I'll just rely on my memory for photo face recognition. (It's\nalso random, but the UI is better. \ud83d\ude42 )<\/p>\n\n\n","protected":false},"excerpt":{"rendered":"

I've just wasted the last two days on something more fun (and more frustrating) than any video game – Google's Picasa photo app \"face recognition\" stuff.<\/p>

Google's Picasa desktop photo application includes the very cool feature of face recognition. Point it at a photo and it will identify where in the photo faces are to be found, and do so with great precision. I pointed Picasa at the 20,000 photos in my archive from this year (maybe 100,000 faces?) and there were no false positives, unless you don't count sculptures of faces as a face. Very impressive.<\/p>

But like a [...]","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[7,4],"tags":[],"_links":{"self":[{"href":"https:\/\/regex.info\/blog\/wp-json\/wp\/v2\/posts\/1388"}],"collection":[{"href":"https:\/\/regex.info\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/regex.info\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/regex.info\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/regex.info\/blog\/wp-json\/wp\/v2\/comments?post=1388"}],"version-history":[{"count":0,"href":"https:\/\/regex.info\/blog\/wp-json\/wp\/v2\/posts\/1388\/revisions"}],"wp:attachment":[{"href":"https:\/\/regex.info\/blog\/wp-json\/wp\/v2\/media?parent=1388"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/regex.info\/blog\/wp-json\/wp\/v2\/categories?post=1388"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/regex.info\/blog\/wp-json\/wp\/v2\/tags?post=1388"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}