h a l f b a k e r yNo, not that kind of baked.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Do you got an old, dusty picture of your high school sweetheart? Just upload it to GooglePix and get the latest renditions to view (although they may be 20 pounds heavier).
Do you have an old copy of the newspaper? Upload it to GooglePix and see today's news online.
Do you like that chick from
the chat room you were on last nite? With GooglePix, upload her pic and it might possibly come back with:
Search Results: 1 - 10 of about 2,831,000,000: Click thumbnail images for full screen photos!
[link]
|
| |
MSR Beijing last year showed an application called MiAlbum. One of the functions demonstrated was the sorting of a very large photo database by marking an image and asking the application to "show me more photos like this one" ("like" the one you marked). There is a paper about it in the ACM archives but you have to be an ACM member to get access to it (or I would have posted a link). |
|
| |
Hmmm. What I saw demonstrated would have been very difficult to do with text classifications but, having no proof, I defer to PeterSealy. |
|
| |
CIRES (your second link) is great and is virtually identical to what I saw being demonstrated with a couple of exceptions: in addition to picking just one target image, you could pick a range of images ("find me pics like THESE") and you could select images that you DIDN'T want ("find me images like these, but NOT like these"). It was an amazingly fast way to navigate through a DB of 6,000+ pictures. It was even cooler that it was running on a touch-screen electronic picture frame and once your selection was returned you could add the results to a slide show on the picture frame. |
|
| |
if flickr or wikipedia could do this, you'd be able to take a photo on your mobile and get a result showing telling you what you've taken a photo of. probably. |
|
| |
http://tinyurl.com/2lvh9z |
|
| |
search photosynth microsoft |
|
| |