Yesterday, Google unveiled its new Google Similar Images search feature under Google Labs. The product is nice, and works very well. But, Microsoft was doing the same thing with Live Search — over 4 months ago.
A lot of commenters pointed it out to us yesterday, and naturally, Microsoft reached out today to let us know the same thing with what might as well have been a big, loud “FIRST!” But it’s true, Microsoft rolled out the feature on December 1 of last year. So how does it stack up?
I ran the same test I did yesterday with Google Similar Images, a query for “Apple.” Perhaps that’s a bit unfair to Microsoft since that is a chief rival, but the results were solid. The silver Apple logo similar image search returned silver apples across the board.
The colorful Apple similar image search was a little more uneven than Google’s results, but that’s just being kind of picky.
The apple fruit similar image search worked well, just as Google’s did.
I took the test of Microsoft’s tool a bit farther and did a search for “Daniel Day-Lewis,” my favorite actor. I wanted to find a picture of him as Bill the Butcher from Gangs of New York in particular. I found one, hit the similar images link, and aside from one odd picture of Penelope Cruz (which I can certainly forgive) and a few of Day-Lewis in There Will Be Blood (where he has a similar look), got some good results.
So yes, Microsoft, which gets a lot of flack for borrowing features from rivals, appears to have had one of its own borrowed by Google here. Of course others in the search space, such as like.com, have been doing this for a while. “Imitation is the greatest form of flattery and all that,” quipped a Microsoft spokesperson. Indeed.
Are you telling me that MSFT beat GOOG on something?
Oh, the humanity!!!
What will be next? 3-D porn with SMELLORAMA?
In case you haven’t notice, live search image search is way better than google. Aside from the search which MS needs to improve they have been improving more and more and it seems that they are getting things right. As always, Microsoft knows how to give some good competition.
“In case you haven’t notice”
we haven’t…Live search is the Zune of search engines. It only exists in stories, not in real search life
Last i check Zune was pretty successful for it’s initial offerings. It was never meant to take over the market instantly but long term. It could still do that and there are many indications that it could take a bite out of the Ipod market.
I called this as image search future. Nice to know MS is leading us to future.
I don’t ever know that Microsoft ever ever ever better than Google.
oh before anybody starts jumping when google adds this feature… which i have not notice it on gmail… check it out it’s called QuickAdd
http://windowsl...salt_042009_120
Now that’s a cool feature and i haven’t seen this being implemented.
Tostdata-man said…
What will be next? 3-D porn with SMELLORAMA?
Hehe, there is no doubt that some web-searchers out will be using the service to search for similar porn images, but perhaps Google or Live Search can filter porn image similarity searches.
There was a system developed by researchers at Stanford University years ago for automated screening of pornography material on the internet. It was reasonably accurate in its classification at the time (well according to its designers/authors), where they used wavelet algorithms. The summary of the abstract of their paper is shown below:
Abstract:
——–
As computers and Internet become more and more available to families, access of objectionable graphics by children is increasingly a problem that many parents are concerned about. This paper describes WIPETM (Wavelet Image Pornography Elimination) a system capable of classifying an image as objectionable or benign. The algorithm uses a combination of an icon flter, a graph-photo detector, a color
histogram filter, a texture filter, and a wavelet based shape matching algorithm to
provide robust screening of on-line objectionable images. Semantically-meaningful feature vector matching is carried out so that comparisons between a given on-line image and images in a pre-marked training data set can be performed efficiently and effectively. The system is practical for real-world applications, processing queries at
the speed of less than 2 seconds each, including the time to compute the feature vector for the query, on a Pentium Pro PC. Besides its exceptional speed, it has demonstrated
96% sensitivity over a test set of 1,076 digital photographs found on objectionable news groups. It wrongly classified 9% of a set of 10,809 benign photographs obtained from various sources. The specificity in real-world applications is expected to be much higher because benign on-line graphs can be filtered out
with our graph-photo detector with 100% sensitivity and nearly 100% specificity, and surrounding text can be used to assist the classification process.
The paper is downloadable from the link shown below, but : WARNING: Some readers might find some of the porn images offensive:
System for Screening Objectionable Images using WIPE (Wavelet Image Pornography Elimination)
I wouldn’t be surprised if both Google’s similarity image retrieval & Live Search do use wavelet somewhere in the architecture of their systems, because wavelet is best is finding similarity in signals, either 2D signals (ie, image) or 1D signal (speech, audio signals, etc,…). Wavelet has also been applied in voice print identification.
The current FBI image fingerprint database retrieval system does use wavelet which was developed at Los Alamos. Also Hollywood are big fan of wavelets, since Pixar Animation Studios debut wavelet in the making of Toy Story, now there is no doubt that every graphic studio in Hollywood jumps into the wavelet bandwagon, because it brings superior graphics compared to any other algorithms.
Wait a minute, Microsoft has a search product that competes with Google Search?
I wasn’t aware of that
Yes MS was there (maybe a imitated some other) But http://www.like.com/ doesn’tt seem to be in the search business…
Hey Tostada-man,
Yes, guys like you are biggest threat to innovation and fair competition. You are so blind that everything else you do not support brings you more disbelief. Wake up and smell the sulpher or you will be obscure.
Actually, MSFT has beat GOOG on innovation many times before, just never got credit for it.
GOOD got a big buzz for incorporating GTalk into Gmail. Hotmail had Web IM integration way back in 2003.
Christopher Payne announced in August 2004, in a seach summit panel discussion, that they were working on Desktop Search, and Google rushed to get their verison out, leveraging their web search technology. When MSFT’s desktop search came out in December many people, including folks on Slashdot (a well known anti-MS site), literraly said “this takes Google to the cleaners”.
MSFT was the first to tie deskbar to Desktop Search. GOOG had deskbar in beta for years and didn’t know what to do with it. As soon as MSN Toolbar Suite 2.0 Beta (the first release of desktop search with integration with the various toolbars) came out, GOOD did the same tie in.
The whole interactive map that Google Map is so well known for was implemented by Live Local first.
This list goes on and on. Obviously, MSFT needs to reflect upon themselve the lousy PR job they have been doing. However, GOOG have been enjoying undue innovation credit way too often and for way too long.
Wait, I got as far as Web IM in Hotmail — wasn’t that just launched yesterday?
http://windowsl...ve.com/blog/cns!2F7EB29B42641D59!38982.entry
Actually, they did that back in late 2003. It was offered to limited accounts in US, UK, and Japan. Unfortunately, MSFT did not seem to realize how innovative and important this was going to be and never deployed it widely. In fact, they kept cutting down on the number of users getting the offer because it was expensive to run.
It wasn’t until GOOG got all the praises that they started to regret never patent it, and finally put in some effort to make it widely available.
Of course, UI wise, what you got yesterday was very different from what was offered back then, since Hotmail’s UI has changed dramastically in the last few years. I actually think it had better experience then, but that may be just me.
More clarification. What they are offering now is a complete reimplementation and not built off what they had back in late 2003/early 2004. However, the idea is that WebIM integrated directly within Hotmail was offered much earlier than GTalk within GMail (or anyone else, if I’m not mistaken).
Image search with similar image… that’s pretty intense work going on in search space.
very soon in future it will be easy to locate perfect “look-alikes” in different countries without any genetic association.
Cool stuff…;) May be one can find 10 girlfriends who look alike in every sense and then choose whoever thinks the “smartest”…:) lolz..
marvin
http://yousuggest.us
Nice. Finally Google gets called out from stealing the thunder from other guys who have been doing the same thing. Google gets massive free PR for whatever they release. Everyone else struggles to get mentioned at all.
This happens with startups too. Many startups innovate and no even knows about and then google goes and copies them and everybody says ‘Google is genious!’ to the frustration of developers everywhere.
It isn’t Google’s fault no one knew this feature existed on Live Search. Sure, a few people knew, but the fact is, very few people use Live Search compared to Google search.
Hey Siegler thanks for posting an article about this… i have said before i really hope you bring balance to your articles. So thanks again for presenting this, it’s nice to see this and give credit where it is due.
msft stock going to $5 http://iamned.com/blog/ the net books revolution & web 2.0 + google desktop and SOA bad for microsoft
besides the obvious indication that you must be smoking something really strong, you just seem to be imagining things… guess that stuff get’s you up there!
It’s good to see this ‘correction’ to yesterdays ‘all hail google’ post.
OK, so we know that Google has no monopoly on inventing/launching first advanced search features.
What we learn here though is that they are able to replicate any(?) advanced feature relatively quickly so as to minimize the risk of losing users to other search engines. It’s hardly a suprise considering their scale and resources.
The big question is: considering Google’s position today is it possible for other search company (e.g. MS Live, Yahoo!, or some totally new start-up) to come up with a new approach or improvement to search which would be much more attractive that the current Google’s service AND which Google would not be able to replicate that easily?
finally some credit
Didn’t google have this feature a long time ago, but only available via putting some magic in the url?
whoopsie daisy!
Assuming Google copied MSFT: Google saw this on Live search on Dec 1, ‘08. Decided to copy, and had a product up and running 5 months later. That’d be freaking amazing IF it was true. Much more likely that BOTH MSFT and Google saw to potential when the startup that became http://www.like.com started running. Yeah they likely both copied this, but these solutions are not easy, so both have made significant investments — not sure “copy” is the right word.
Nice to see, however, because there is a lot of utility yet to be discovered here, and competition will get us there faster.
Exactly what I thought when people were commenting they copied MS.
God you guys are really Mac fans all the way. I mean, every time Microsoft borrows a little feature, you guys go crazy mad, critisizing Microsoft like it’s the most evil corporation in the world, but when it’s the other way around you shrug your shoulders and go “eh” and pretend in it never happened…what is it with you guys? Microsoft seems like it’s being a realllllly good sport by saying immitation is the highest form of flattery so be more enthusiastic!!! Jst so you know, I’m the biggest Google fan and own a Mac. :p
God you guys are really Mac fans all the way. I mean, every time Microsoft borrows a little feature, you guys go crazy mad, critisizing Microsoft like it’s the most evil corporation in the world, but when it’s the other way around you shrug your shoulders and go “eh” and pretend in it never happened…what is it with you guys? Microsoft seems like it’s being a realllllly good sport by saying immitation is the highest form of flattery so be more enthusiastic!!! Jst so you know, I’m the biggest Google fan and own a Mac. :p
Actually, similar image search is an old topic. Engineers in digital imaging industry have done most of the basic work in the 80-90’s. GOOG and MSFT guys are just testing if it’s worth building something out of it.
too bad anything microsoft can do google can do better.
-Jack
Hey live wier
Who said Google did it better? Have you tested it thoroughly? Why you guys are so fast in judging and just saying stuff like “google can do better”? Test both of them side-by-side… then use your god-given intelligence in your head to speak out. Google does many things better, but that does not imply they can do everything better. For example their products search sucked big time – but people like you will ignore those. Anyway, does not matter, as blind people like you will lose at the end as other non-biased people will adopt whatever is better and use them.
I don’t even know the url for microsoft’s search engine.
http://www.live.com
NickyD said…
GOOG and MSFT guys are just testing if it’s worth building something out of it.
Totally agree.
I drink your milkshake? REALLY?
It’s not like anyone stole anyone else’s idea here. Image comparison algorithms have been discussed in the literature for eons, and adding that feature to an image search is an obvious idea (and not as easy to implement well as it may seem).
To Microsoft’s credit, they were the first *big* company to make the feature available. To Google’s credit, they were definitely better at marketing it. So who matters, in the end?
ugh huh..LOL
It is not Google’s fault no one knew this feature existed on Live Search. Sure, a few people knew, but the fact is, very few people use Live Search compared to Google search because Google is user friendly
It is not Google’s fault no one knew this feature existed on Live Search. Sure, a few people knew, but the fact is, very few people use Live Search compared to Google search because Google is user friendly
Sorry, forgot to add great post! Can’t wait to see your next post!
“because Google is user friendly”
Eh? What are you talking about? People use Google cos its better. What does “user-friendly” mean?
Its all about keeping some information private, lesson learned by Live!
Found this one on Reddit:
http://tinyurl.com/cvwhep
The copyscape of Images is here.
This will allow owners to track their copyrighted image usage.
Hi,
Actually, when was the last time that Google did anything truly original, more often then not it’s an ignorant media on the google cool-aid that hypes the company and it’s creations no matter historical fact – just re-publishing those press releases and happy to give attention to company spokespeople.
Microsoft, as well as Amazon, as well as Yahoo, even Aol, et al, have repeatedly out-innovated well in advance, and have not got coverage for their work, yet google launches mobile products, or latitude or street-view or sea-view/space in Earth and hours and hours of air-time and column-inches are given over to it.
Kind regards,
Shakir Razak
I think somehow along the way, as technology developed, Google established itself as a trusted brandname, and so inevitably often creates more hype than any other similar product that may have been released before Google did it. We call web search “Googling” or we say, “Let’s Google it”, and not “Let’s Live Search it”. Could be a vicious cycle: when Google first started out, there was a large media coverage probably because it was then, a new product, provided free-of-charge and user-friendly. And probably, a refreshing change and challenge to the Microsoft monopoly which would have been welcomed by Microsoft detractors. Naturally, the media continued to provide coverage on Google innovation because that is what the masses want to know.
As much as I think other developers have provided similar products and services that may be better than what Google is offering, I think the advantage that Google has is just that it has become a one-stop solution for all the user’s web search needs. As a user, I’d rather go to a place that can provide all that I need, even if it’s individual products and services may not be as good as what another singular developer is offering.
TinEye offered this way before Microsoft and Google if I’m not wrong.
“Microsoft… appears to have had one of its own borrowed by Google here”
Wow, don’t go too far out of your way to correct yourself there.
Given the flack that Microsoft gets, I think the call out was legitimate. But despite the comparison, your tone at the end makes it sound like “Google’s is still better, and if you insist on pointing out Microsoft, well, guess what – Microsoft was one of *many* we didn’t cover.” That doesn’t justify the exclusion as much as it shows apparent bias.
Have a look at the Xyggy beta image search at:
http://www.xyggy.com/image.php
Drag ONE or MORE images from the left-hand set of images into the search oval. The results will appear on the right.
Add additional images to the search by dragging and dropping more images. Remove images from the query by clicking on the image. Click the search oval to clear the search and start a new one.
Isn’t this what image search should be all about? Imagine being on any web page and dragging and dropping images from that page into the search oval to find similar images.
Dinesh