UPDATE: Riya is not launching a public beta today, but is sending out emails with credentials to people who’ve signed up for the alpha. In a few days Riya will be opening up registration for everyone. More on their blog here.
Rumors are buzzing today that the long awaited, almost acquired photo facial recognition service Riya should be moving out of private alpha and launching a public beta sometime today. The main site is currently down and “asks for your patience”, which certainly suggests that something interesting is going on over there.
Riya, which is funded by Blue Run Ventures, Leapfrog Ventures, and Bay Partners, is a service that automatically recognizes people in photos and groups them. Add a tag to the person and all of the photos are tagged with that name. As your friends and family join Riya too, many of their pictures will automatically be tagged with your data (and their other friends’ data) too.
Riya is a company that I’ve loved from the start (screen shots here), even when it was called Ojos, and have written about it often. Congratulations to the entire Riya team if, in fact, it is launching today as I’ve been unofficially told. I am really hoping that they adopt a partnering strategy and get this built into Flickr and other photo sharing services. Or at the very least build a hell of an importer so that I can move my photos to them. Yahoo, if Riya works (it certainly did in my beta testing), please buy this company.
Riya is live now!
Their blog says those that signed up for the alpha and beta will be getting an email tomorrow. Looks like the rumour’s true.
i don’t see anything specil in riya.. there are plenty face recognition sw around
Sogn up page says beta will start in few weeks, so not really a public launch. Bogus!
Yep, totally agree on the partnership/acquisition angle. The last thing we need is yet another photo sharing site – integration with existing sites makes more sense. Provided they can deliver on their promises, an acquisition seems inevitable.
Check out http://www.myheritage.com. Inetersting face recognition app
Man, I hope they are work it. They are so hyped right now.
I got the email this morning about the launch of the beta. Keith’s comments, though, made the most sense.
Hey Mike!
I’ll probably get in ka ka for this, but you can send your readers to the sign up page here:
http://www.riya...om/groundTruth/
The doors will be thrown open very soon…so this is a special ’sneak peak’.
Tara
Having all those stupid boxes around people’s faces showing me their names is really annoying.
yay,, got into riya at last
I got an email too, just signed up and going to install the uploader. It looks like it may be an open launch – the link given is a normal one (not like gmail invites with a special link for each invite) and I signed up using a different email account with no problems. The front page sign up link does not seem to be updated – the sign up link in the email is http://www.riya...com/groundTruth
I would recommend Google to buy Riya. Google can totally use Riya for their image search efforts. They can use the technology to catalog and tag all the images on the web.
Once you have that then they can easily turn on the contextual image based advertising.
Lastly I do think Google does not have a good online photo service story. Picasa is a good application but not as great as Flickr.
I think the service looks pretty cool. If Google snaps them up I wonder how much they will buy them for. It seems like new tech start ups are cashing in… big time.
HHmm contectual image based advertising. That’s a pretty cool idea.
Waheed
http://www.mphnews.com
I was pretty excited when I got the email this morning with the invite, but my joy was killed when I read that it’s not available for Mac. Boo!
“I would recommend Google to buy Riya. Google can totally use Riya for their image search efforts. They can use the technology to catalog and tag all the images on the web.”
Riya only recognizes faces, not objects. There are a few other companies out there that do full object recognition, which is quite different, and in my opinion, much more valuable for image search. When I search for images of an “apple” on Google/Yahoo/MSN image search, the results are usually much worse than when I search for a person’s name.
Also, maybe you guys should use it first before “recommending” it be acquired.
“Yahoo, if Riya works (it certainly did in my beta testing), please buy this company.
Riya”
Yahoo really really sucks, why them?
Wasn’t it the non-existence of this technology in a reliable form a big part of the reason Mechanical Turk was invented? I suppose we’re getting closer to the point where AI is able to do anything we can do and without all the distracting emotions. What a world! I want ice cream with it is all I care about.
http://www.riya...om/groundTruth/
gives a 404 Not Found error.
Requesting an acquisition? That really sounds strange and show how little you know of the technology itself. I’ll recommend you read up some stuff before making such statements…I guess you want to continue to loose credibility.
requires jre 1.5! – I’m not about to move from 1.4.. and spend an hour adjusting my dev. env. to all the new changes.
I wish there would be some other way to upload files.
It seems to me almost all the people who are hyping this company have not spent much time building computer vision applications. (I did. It’s hard.) When machine vision applications are made to work robustly, they are made to work well on just a couple of things, like text, faces or buildings. But what works on faces and buildings does not work on cars and scenery and birthdays and hundred of other categories that people will naturally want to “auto-tag” their photos with. Digital photos are really about everything.
The naive consumer who uses an application that is enhanced to work on faces will expect the application to work on everything, but machine vision engineering doesn’t scale that way. If “um, I know it auto-tags X and Y, but can it auto-tag A B and C?” gets answers like “Smart men are working on the problem. Smart men.” for more than a few months, you’ll have a disappointed consumer after a year.
I do not see that much overhyping at blog.riya.com, but because it is easy to imagine what is possible (servers that can automatically describe/tag any picture), people will think its like rolling out another vertical. But its not.
When you see a technology that autotags any image from images.google.com, you know you have something that is actually scales for consumer photos. The demo will be: give us a img src url, we’ll guess what it is. (No “network effects” or Mechanical Turks needed, the training data is on images.google.com itself)
That is not Riya’s demo. For obvious reasons, Riya is going to work the “Smart men are working on the problem. Smart men.” line for as long as they can. But that doesn’t mean you should see a horse with a few stripes painted on it as a zebra.
#13, #23, get rid of the extra slash at the end of the url and it will work.
gentleman sourabh niyogi,
seems lately people have got habbit of getting into bed with smart men…..
if u have build vision system, why dont u team up with riya…
i am looking for reasons for not going on google for everything….
Abhishek:
After the alpha launch, Riya reset expectations on its capabilities. Riya supposedly works best on images with large resolutions (at least 2MP) and with EXIF details (date/time the picture was taken, location, etc).
This implies, Riya will not work on most images on the web because the majority of images on the web are low resolution and do not have any EXIF details.
So, this is not a tool Google/Yahoo will be able to use to tag all images on the web.
IBM is working on a much more sophisticated application that can apply general pattern recognition over video and images.
http://www.rese...ibm.com/marvel/
Riya sounds cool but like everyone said before, do we need another image sharing application? Plus how hard is it to just write a couple of tags for an image?
I wonder why nobody is saying this, but this software _begs_ to be desktop based. most of put only selected photos online (e.g. in flickr), but it’s on our machines that we have the massive amount of family photos and faces.
They should at least give this as an option. I do believe the desktop is where it can have a bigger impact on the individual user and therefore be worth quite a lot.