Visual search and image recognition is one of the holy grails of consumer Internet technologies. Picitup is jumping into the deep end of this space by announcing the launch of its public beta.
Unlike Like.com (formerly Riya) which focuses on likeness, Picitup focuses on attaining matching images. This differentiation is important as it sets the company on a completely different trajectory in terms of both offering quantifiable value to users, as well as delivering a business model at the end of the day.
An image search on Picitup begins with a textual search actually queried on Google or Yahoo. Picitup will display a set of results only from one of the two—the basis of the decision is the speed and quality of the results. The user can then select which image Picitup should fetch similar images for, or filter the results by Faces, Products, Landscapes and Color. The analysis is made in real time and is based on 100+ parameters including a propriety color space the company developed.
Erick recently wrote that:
It’s hard to compete in the search engine market, but one approach taken by several startups is to sit on top of the big search engines and try to improve their results or interface. Why reinvent the wheel when you can simply add new spokes?
From a practicality point-of-view, relying on the likes of Google and Yahoo makes sense, but it should be noted that they forbid the reordering of their results, a sticking point that surely has a negative effect on the quality of results Picitup ultimately delivers.
Picitup claims it shortens the number of pages needed for an image search from 10 to 2. However, from my experimentation its engine’s match reliability was shaky. Results were pretty good for Ford Focus, but not even close for this Running Shoe. Note that all images should theoretically correspond to the top left-hand image.
Another issue that left a sour taste in my mouth was CelebrityMatchUp, an attempt to add some light-hearted fun to the beta interface. The idea here is that users upload photos of individuals and have Picitup produce results of people they resemble. This doesn’t exactly work. For instance, consider that Michael Arrington’s photo brought back results that he resembles both Barack Obama and John McCain. Huh?
Erick Schonfeld’s photo results are also somewhat curious, although the bright side is that Erick’s wife should be delighted to know she married a Kevin Costner look-alike. For a company claiming its forté is in image matching, Picitup should not have opened this door.
Alon Atsmon, co-founder & CEO, believes the company’s technology is compelling enough to drive revenue both from ads and through licensing a white label version of the engine for integration into ecommerce sites.
True, my initial impressions of Picitup are not necessarily positive ones. However, considering Atsmon is a serial entrepreneur I’ll remain optimistic and wait for Picitup to iterate a couple of more times before I cement my judgment.
Contextual image search doesn’t work very well, even on Google image search, so this makes sense to me as an extension of the existing services. I wonder how long they’ll be on their own feet.
Peter
do you follow me on http://twitter.com/peterurban
Ok, well the concept is interesting, and I see real potential for contextual image search, but the Picitup homepage needs to do a much better job of explaining how it all works and why I should use it. It’s a classic example of extreme information minimalism.
Wow i tried it and it said i looked like Brad Pitt! I can tell this company is going to be the next google.
there is already look.com, and like.com who are already in this space…and some of them have built facebook and other apps with this technology. What is new here? What chance they have with this half-ass name.
Contextual image? Does it mean we search a image from google image first then only use Picitup for further searching ?
“It’s hard to compete in the search engine market, but one approach taken by several startups is to sit on top of the big search engines and try to improve their results or interface. Why reinvent the wheel when you can simply add new spokes?”
What’s the point in adding new spokes?
MyHeritage has some cool facial recognition software and a celebrity-morph feature whic yields some pretty comical results…
http://www.myhe...ace-recognition
hmm..really it is different..
i think site is down
I guess their photo/face matching algorithm won’t be used at airports and government buildings any time soon, eh?
Harry “these are almost always disappointing” Wang
ROFL @ 10 you crack me up
Harry “I have a small” Wang
Text search is good enough as of now.
I think web is still not evolved to a point where we need such utilities.
Its like inventing sun glasses even before there were eyes, anticipating that one day we will have eyes and will need sun glasses..(I know this example is bit confusing but you know what I mean!)
I feel these search methods are bit ahead of time, and its hard to find value in them unless web is really evolved up to that point.
Sometimes,serious technology makes you laugh as well.This is what happened to me after i searched some not-so-famous guys.
Anyway,the common utility of image searching is yet to be figured out.
the site doesn’t move… still waiting.
A few comments:
In general, we are the First to allow for web wide free visual search. We are in Beta but are making fast progress.
1. A nice example of the enignes results could be found in the following:
Before: http://tinyurl.com/5wzwnq
After: http://tinyurl.com/6rgrn5
It shows how much time can be saved, and how accurate our resuls can get.
2. Please note the running shoes sample mentioned had no second match in the original results and hence the similrity match should scratch out the closest matches it could find.
3. The Celebrity match part of the site is meant to be used when there’s a second image of the same person in the DB, which was not the case for Mr. Arrington… A nice example of that should be http://tinyurl.com/64bgpa
Alon
Smoke and mirrors.
They need more smoke.
pixsta visual search and browse – PICTURE YOUR SEARCH {seesmic_video:{”url_thumbnail”:{”value”:”http://t.seesmic.com/thumbnail/tvuDA4A4xm_th1.jpg”}”title”:{”value”:”pixsta visual search and browse – PICTURE YOUR SEARCH ”}”videoUri”:{”value”:”http://www.seesmic.com/video/l7dqFTUqEm”}}}
As anyone heard of TinEye. It is done by a Toronton based company, Idee Inc, and it is better than all of these. try it at http://www.ideeinc.com. TinEye .
see http://www.yout...h?v=sN7SJtoNgoU
Is it just me or have they basically just added a “Similar Pictures” button to the Google Image interface… from my tries it looks like they still rely much more on text search and contextual meta content than the image itself.
I agree with #17, Idee’s TinEye stands head and shoulders above this site…
@11: Hey, that’s not funny!
Harry “not all Asians are small” Wang
This sounds alot like what Mugr is doing. A small startup here in Phoenix AZ has attempted the same thing with decent results.
http://www.mugr.com
Face matching is tough, I know that picture resolution has much to do with the matching accuracy, what resolution were those photos at?
I agree with Ryan (#20).
I estimate that over 99.9% of image searches are by people trying to get from keywords to an image, and not from an image to similar images. If I look at myself and everyone close that I know, it just doesn’t feel like the need is there.
Maybe it can be useful to some verticals. Or for things like automatically tagging (or even geotagging!) photos.
But of course, it’s all subject to their technology working better than it currently appears.
Anyway, good luck in their uphill battle in this crowded space.
to #18 and #20, the site you mentioned is a tiny image DB demo , the challenge of doing a web search is 1000x harder; and http://www.picitup if the first and only site to do that…!
To #12, take a look at http://www.pici...citup/press.jsp
@alon you may want to fact check your comment.
That “tiny image DB demo” has half a billion photos indexed (and growing) in it and searches for other images based on the image you provide it.
It’s a true image search engine – Picitup is just Google Images rehashed.
Here’s a good example of the power of tineye: http://ricksega...o-be-authe.html
Try doing that with Picitup…
Ryan,
I quote “a 3 million image collection”
Best of luck Anyway….
Ryan, that explains http://www.quan...www.ideeinc.com
To #2, You were right, http://www.yout...h?v=sN7SJtoNgoU
was added
Alon… I’m not involved with them in any way… but your “3 Million” quote is well out of date.
From a recent query on the site “searched over 487 million images in 0.838 seconds”
Amber Mac gives a pretty good rundown/demo of the tool on her blog: http://ambermac...e-does-for-text
Sorry everyone didn’t line up to sing your site’s praises but dude, it’s not that original… hope you don’t have your own money in this dog…
@Ryan,
Say hi to your Toronto colleagues, I really want them to succeed
, but I do have to point out the differences ( rather than trash talking…)
But Seriously, have a real try at picitp.com, it is a very useful tool for people who seek images and is gaining terrific momentum ( says alexa et al..)
killerstartups is saying:
“Picitup is certainly fun and easy to use. It’s better than what Google currently offers and visually more interesting.”
http://www.kill...Search-Picks-Up)