Beta

Slashdot: News for Nerds

×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Searching by Image Instead of Keywords

samzenpus posted more than 9 years ago | from the find-me-something-square-and-green dept.

Graphics 184

Content based image retrieval (CBIR), the technique to search for images not by keywords, but by comparing features of the images themselves has been the focus of much research ever since the web emerged. Consider for instance adding CBIR to Google Images, where you would be able to search for images similar to a query image instead of using keywords. A research project at Penn State University has recently been applied to the biggest aviation photo database in the world with close to 800,000 images. You can search for images similar to a photo already in their database (click "View similar photos") or submit your own query image. Some queries generate better results than others but CBIR is certainly here to stay and will be standard in many image applications of the future.

cancel ×

184 comments

Think of the greatness to society! (5, Funny)

qewl (671495) | more than 9 years ago | (#12437393)

I can't wait to put a nipple into it!

Re:Think of the greatness to society! (0)

Anonymous Coward | more than 9 years ago | (#12437582)

Why is it that the two people with that bunny in their sigs are referring to nipples?

See parent and http://slashdot.org/comments.pl?sid=148396&cid=124 37445 [slashdot.org]

Re:Think of the greatness to society! (2, Insightful)

rpcxdr (796317) | more than 9 years ago | (#12437641)

Ehm - not no be too much of a geek, but here are some airplanes that (don't) look like the Slashdot logo...

http://www.airliners.net/similarity/index.php?imag e_url=http://images.slashdot.org/title.gif [airliners.net]

BTW, if you want to post other searches, this URL format seems to work.

Skin Cancer Detection (3, Interesting)

Anonymous Coward | more than 9 years ago | (#12437665)

Talking about greatness to society and a little bit of skin. At university one of my projects was a system that used CBIR to try and diagnose skin cancer. The doctor would take an image of the suspect area it then would be compared against a database of cancers. It would then return a suggested likelyhood of being cancer. It also allowed the doctor to build a history of images allowing easy comparision over time.

I always felt good about working on projects like this, gives a warm fuzzy feeling.

Search for this image of child murder in Iraq (-1, Troll)

Anonymous Coward | more than 9 years ago | (#12437740)

How's this [yahoo.com] for something to search for?

How about images [google.com] of Beslan.

Say what you want, you fucking leftist, anti-American, stupid shitheads. The US of A does not deliberately target children.

Look at that picture long and hard. And think long and hard about who you want to win in Iraq.

I sure hope you're not rooting for the side that doesn't let women vote much less wear anything short of a tent in public, and buries gays alive under dump trucks full of stones.

And let's not even begin to mention murdering children.

Those images make me sick. Almost as sick as the sheltered, useful idiots who support such actions.

PS - if you agree with me, I dare you to post those images at school or at work. Time to fight against those that support child murderers.

Re:Search for this image of child murder in Iraq (0)

Anonymous Coward | more than 9 years ago | (#12438226)

As "sickening" as you may find it, children are actually one of the easiest items to produce, and as such one of the easiest items to replace. For instance, the 3 billion USD design of a marginally improved turbine is worth more than the lives of 100 children of the age of 5, considering even a high of 300 000 in productive capacity from unique attributes with perhaps 20 000 worth of productive time spent in emotional raising annually. Compared to 3 000 000 000 + 100 000 000, the 300 000 000 000 entering a burning building for instance it is better to save the plan than the children by practical measures. It is for the same sort of practical measures, over which all war is actually waged and not the smear/glorification campaigns dreamed up afterwards or falsified by hedging comments before or afterwards.

False positives (2, Funny)

suso (153703) | more than 9 years ago | (#12438077)

Actually, it will be hillarious what will happen when grandma puts in a picture of her grandson taking a drink from the hose in the backyard.

Its almost like telling someone to go to whitehouse.com

Re:False positives (1)

Knnniggit (800801) | more than 9 years ago | (#12438160)

Or how about this: my little brother with a garden hose between his legs, struggling to fit a water baloon over the nozzle while water sprays out. I gotta find that picture...

america the beautiful (-1, Offtopic)

Anonymous Coward | more than 9 years ago | (#12437395)

free the michael sims seven

jesus loves all the little critters and varmints of slashdot

wtf? (0, Redundant)

merpal (873013) | more than 9 years ago | (#12437402)

I don't understand how this is relevant to slashdot...

If this technology has been in development for a long time, why hasn't google taken the ball and run with it?

Re:wtf? (4, Interesting)

Rei (128717) | more than 9 years ago | (#12437442)

Because it still has problems - you'll note that the pictures seem to be compared simply based on color similarity. That's the same thing imgSeek [python-hosting.com] does (a great program for sorting and searching your photos) on photo searches. It works wonderfully if you're searching a very limited picture subset (say, airplanes), but if you search a wide variety of pictures, the results can be quite amusing.

Re:wtf? (1)

Xcott Craver (615642) | more than 9 years ago | (#12437661)

You also have the opposite problem with color histograms: two very similar images, even two images taken of the same static scene, seconds apart, can have substantially different color histograms.

I used to do research in CBIR, and in my image library I had 5 photos of a christmas tree. By any efficient metric, one of them was always way far away from the others.

Xcott

Re:wtf? (1, Funny)

merpal (873013) | more than 9 years ago | (#12437754)

don't mod me as redundant you niggers

Re:wtf? (2, Interesting)

chefmonkey (140671) | more than 9 years ago | (#12438189)

Google actually did take this technology and try it. The first version of their image search had a "find similar" link next to every image. These tended to work okay at first (they weren't great, but you usually got enough photos back that you could visually scan them and find something of interest that was related to the original image). After a few months, for some reason, the "find similar" links started returning increasingly nonsensical results. After it degenerated to the point of near uselessness, they took the "find similar" link away from the image search results. I expected it to turn up again once they got the kinks worked out, but apparently they just decided to stop working on it.

Arm jokes... (2, Funny)

sharkey (16670) | more than 9 years ago | (#12437408)

and set for goatse!

Re:Arm jokes... (0)

Trejkaz (615352) | more than 9 years ago | (#12437790)

Actually on a serious note, has anyone actually tried drawing a guy bent over and stretching his anus, and made sure it found the right image? I'm genuinely curious.

Location? (4, Funny)

poopdeville (841677) | more than 9 years ago | (#12437416)

What an awful beach [airliners.net] .

Re:Location? (0)

Anonymous Coward | more than 9 years ago | (#12437464)

Indeed. No boobies.

Old photos (2, Interesting)

green pizza (159161) | more than 9 years ago | (#12437491)

Those must be old photos. There is no way that beach would be open to the public in the post 9/11 world.

Re:Old photos (1)

YrWrstNtmr (564987) | more than 9 years ago | (#12437532)

Here's [airliners.net] one taken there about 6 weeks ago.

Re:Old photos (2, Informative)

znaps (470170) | more than 9 years ago | (#12437610)

It is, I was on it in December.

It's an amazingly scary experience to sit there at night when a large plane is landing.

Re:Old photos (0)

Anonymous Coward | more than 9 years ago | (#12437923)

photo taken: Netherlands Antilles, February 18, 2002

Re:Old photos (2, Informative)

stuffman64 (208233) | more than 9 years ago | (#12437990)

No, it still happens.

In fact, there is a bar [sunsetbeachbar.com] located right in the flight path of the runway. I just met a guy who came back from there, and said it's quite interesting to have planes landing so close to you.

Re:Old photos (2, Informative)

eander315 (448340) | more than 9 years ago | (#12438165)

They still fly that approach every day. Here's a picture [airliners.net] taken this year from the same beach on St Maarten (SXM). That airport is famous for it's low approach. That's a nude beach, by the way, and there are many photos to prove it if you dig around airliners.net. From Wikipedia [objectsspace.com] :

"The island is served by many major airlines that bring in large jets, including Boeing 747s, carrying tourists from across the world on a daily basis. This fuels the island's largest revenue source, tourism. The airport is famous for its short landing strip - only 2130 meters, which is barely enough for heavy jets. Because of this, the planes approach the island flying extremely low, right over the beach. Countless photos of large jets flying at 10-20 meters over relaxing tourists at the beach have been dismissed as photoshopped many times, but are nevertheless real."

Re:Old photos (0)

Anonymous Coward | more than 9 years ago | (#12438321)

I bet somebody could take a fishing rod and catch themselves a hell of an undercarriage.

Re:Location? (1)

PenguinBoyDave (806137) | more than 9 years ago | (#12437592)

Pucker Factor #9...and rising!

Re:Location? (1)

jmb-d (322230) | more than 9 years ago | (#12437621)

Oh My ${deity}, the noise must be horrific! My ears are hurting just from *looking* at that picture.

FULL THROTTLE! (1)

darkitecture (627408) | more than 9 years ago | (#12437968)


Oh My ${deity}, the noise must be horrific! My ears are hurting just from *looking* at that picture.

Actually, engines aren't at full throttle during landing, so it's actually not that loud at all :P

Re:Location? (1)

Knnniggit (800801) | more than 9 years ago | (#12438136)

What makes it even funnier is that, aside from a few people, no one is even acknowledging it. Nothing out of the ordinary. They've gotta have pretty low standards.

Re:Location? (0)

Anonymous Coward | more than 9 years ago | (#12438228)

Funny how it doesnt cast a shadow - like everything else!!

Re:Location? (1)

agraupe (769778) | more than 9 years ago | (#12438278)

Let me clear some things up:

1) These photos are not digitally edited. This beach and airport is real.

2) It is not an awful beach; it is the holy grail of aircraft spotting, and a place that most aviation enthusiasts (myself included) would like to visit someday

3) If you say 9/11 in relation to this subject, such as, "OMG! it's unsafe there will be terrorisms LOLZ", I will be forced to kill you.

Wow (5, Interesting)

themoodykid (261964) | more than 9 years ago | (#12437418)

I was just thinking about this the other day. I think content-based image search is one of the Next Big Things. Cameras are so ubquitous now (for better or worse), but having to rely on metadata to give meaning to images requires lots of effort up front.

It will be interesting if we ever get to a stage where we can just search for a random object (or person) in a database of photos. Then you could take pictures of everything with an always-on camera and if you need to find where you put your car keys, just do a search.

Re:Wow (1)

mboverload (657893) | more than 9 years ago | (#12437520)

But shit, this REALLY works! It's amazing.

Re:Wow (4, Informative)

theguyfromsaturn (802938) | more than 9 years ago | (#12437893)

If you are only interested in searching for images on your own computer, have a look at imgSeek. http://imgseek.python-hosting.com/ [python-hosting.com]

It's been around for some time now. You can not only use an existing image to search, but also do a rough sketch. Check the screenshots: [sourceforge.net]

Nice complement to what has been presented in this article.

Re:Wow (0)

Anonymous Coward | more than 9 years ago | (#12438092)

The problem isn't as simple as you think. Humans visualize things as 3d objects. If you were to look for a key in a picture, you know what it looks like in multiple dimensions. You then look at the picture, translating objects from 2D to see if it matches said key.

To solve this, multiple viewpoint cameras (that will get a better approximation of the object boundaries) and cameras that take pictures of objects from multiple dimensions will need to be used.

Until then, you can come up with a number of tricks that will find similar pictures, often by color or intensity.

Uses in stomping out child pornography (0)

Anonymous Coward | more than 9 years ago | (#12437422)

How long until this is employed by the FBI to attempt to discover illegal images on Usenet etc?

More importantly, how many agents will jump at the begruding task of sorting through the thousands of false-positives?

This soo not works ! (0)

Anonymous Coward | more than 9 years ago | (#12437426)

I just tried to search with a photo of my keys... It came out without a valid response !

Next up : Mission remote control.

Brilliant! (0)

kid_wonder (21480) | more than 9 years ago | (#12437427)

This that I have an image to search with...which is why I am searching for images to begin with - to find an image.

Re:Brilliant! (0)

Anonymous Coward | more than 9 years ago | (#12437441)

Ted: "Whoah."

So easily Subtroverted (0)

Anonymous Coward | more than 9 years ago | (#12437428)

Green money says if this technology were rolled out in force, pr0n sites would find a way to hijack it.

Re:So easily Subtroverted (0)

Anonymous Coward | more than 9 years ago | (#12438052)

Hijack it? hell this would be the primary use and fuel the development of the technology.
all good tech starts with pr0n, I pitty the fool who dont know that.

It works! (0)

Anonymous Coward | more than 9 years ago | (#12437429)

I took a photo of me giving my webcam The Finger and uploaded it. The first thing that came up was a picture of George W. Bush! I guess it's a match because we are both good at giving the entire world The Finger.

Dupe-tastic (0)

Anonymous Coward | more than 9 years ago | (#12437434)

omg a dupe from last year

They would need major manpower to maintain this db (3, Funny)

guardiangod (880192) | more than 9 years ago | (#12437437)

This is just asking for trouble. As most of you would probably imagine, some self-proclaim "comdeian" would post either porn pictures, or pictures that resembles porn body position.

They would need a team of outsource Indian workers to go through each picture one by one!

I am not Indian but...can I apply for the image filtering job?
I said this first, I should get the job ;) .

Top Search (3, Funny)

daishin (753851) | more than 9 years ago | (#12437445)

Something with two circles and dots in the middle of each circle.

Re:Top Search (0)

Anonymous Coward | more than 9 years ago | (#12437603)

You'll get a propeller.

Re:Top Search (1)

kryogen1x (838672) | more than 9 years ago | (#12437609)

like this: (.)(.)?

Re:Top Search (1)

falzer (224563) | more than 9 years ago | (#12437972)

Axe wound.

Re:Top Search (1)

qewl (671495) | more than 9 years ago | (#12438240)

Whoa, mad props to the guy who's not only the first to copy my sig (as far as I know), but ALSO unintentionally dupe my comment! *added to friends list!*

security issues (0, Offtopic)

green pizza (159161) | more than 9 years ago | (#12437448)

Forget slashdotting Airliners.net, how long before the TSA shuts down that website? The trainspotting hobby has already died off following terrorism fears, I can't help but think that other enthusiast sites like Airliners.net will be next.

Re:security issues (1)

avalys (221114) | more than 9 years ago | (#12437483)

Maybe trainspotting has died down because all you get on Google now are results for that wretched movie.

Re:security issues (1)

oldwolf13 (321189) | more than 9 years ago | (#12437554)

Now I've never seen trainspotting, but I find that this happenns with a lot of movies/topics. Trying to filter through reviews, dvd sales, video games, etc. when trying to find info on the subject. (ie: I noticed this first when "The Mothman Prophecies" first came out and I wanted to know more about these supposed mothmen.

Re:security issues (1)

michaelhood (667393) | more than 9 years ago | (#12437809)

We always tell people not to mod down improper observations, so I'm going to try to practice what I preach so to speak.

On what grounds could the TSA squelch photographers and their right to share their creative works (which is their livelihood)?

how long will it take (1)

scenestar (828656) | more than 9 years ago | (#12437449)

before some troll adds goatse or some spammer adding random advertisements.

for the love of god think of the children

The ever popular 'Breast' option... (1)

farmkid (15226) | more than 9 years ago | (#12437454)

Hmmm...

+"34b"

-"puffy" ... Profit! (Oops, or something, grabbing for a Kleenex)

I tried something like this before. (0)

Anonymous Coward | more than 9 years ago | (#12437486)

But I found I just couldn't type images as easily as I could words so I went back to using google.

Similar images (2, Funny)

iMaple (769378) | more than 9 years ago | (#12437487)

Some Applications of Our Research
1. Airliners.net
A site with almost 1,000,000 aviation images.


Wow !!! I tested their Sample search [airliners.net] and all the results were aeroplane photos !!! Ok, ok the site only has airplanes but still ..:)

On a more serious note the alogorithms seem to look for similatity in the colors and lighting rather than the subjects (for example it shows the interior of a cabin in photos similar to a whole plane in the sky. To really see its effectiveness we need to test in in the real world (google images) . The 'artisticly revealing' photo you always liked ... now you should be able get similar pr0n (^H^H^H^H I mean art) with these algorithms

Re:Similar images (0)

Anonymous Coward | more than 9 years ago | (#12437537)

Yes , I noticed that too the images are just scanned for color similarity , no big deal. There is already software for this sort of things (imgSeek or something similar)

The Human Brain (1)

smokeslikeapoet (598750) | more than 9 years ago | (#12437494)

Isn't this kind of how the human brain works to identify objects your eye hasn't seen before?

IANABP, I am not a bio-physicist but it seems very much like artificial intelligence to me.

And for 'lynx' users... (5, Funny)

mikael (484) | more than 9 years ago | (#12437504)

... the search engine will support ASCII art image searches.

Drawing a picture (0)

Anonymous Coward | more than 9 years ago | (#12437556)

Would it be possible to simply draw a picture and have a search be based upon that?
If you have any skills at drawing, it would be quite a useful tool.

Some relevant research papers (5, Informative)

FleaPlus (6935) | more than 9 years ago | (#12437575)

There's a bunch of interesting papers out there on content-based image analysis and retrieval. Below is a sampling from my bibtex file. Does anyone else have others they'd like to share?

* Finding Naked People [hmc.edu] (Fleck et al, 1996)

* Video google: A text retrieval approach to object matching in videos [ieee.org] (Sivic & Zisserman, 2003): web page demo here [ox.ac.uk]

* Names and Faces in the News [columbia.edu] (Berg et al, 2004)

* FACERET: An Interactive Face Retrieval System Based on Self-Organizing Maps [springerlink.com] (Ruiz-del-Solar et al, 2002)

* Costume: A New Feature for Automatic Video Content Indexing [www.irit.fr] (Jaffre 2005)

One more: automatic film character retrieval (3, Informative)

FleaPlus (6935) | more than 9 years ago | (#12437636)

I forgot one more, where specific faces were automatically retrieved from feature-length movies and Fawlty Towers:

Automatic Face Recognition for Film Character Retrieval in Feature-Length Films [cam.ac.uk] (Arandjelovic & Zisserman, 2005)

The objective of this work is to recognize all the frontal faces of a character in the closed world of a movie or situation comedy, given a small number of query faces. This is challenging because faces in a feature-length film are relatively uncontrolled with a wide variability of scale, pose, illumination, and expressions, and also may be partially occluded. We develop a recognition method based on a cascade of processing steps that normalize for the effects of the changing imaging environment. In particular there are three areas of novelty: (i) we suppress the background surrounding the face, enabling the maximum area of the face to be retained for recognition rather than a subset; (ii) we include a pose refinement step to optimize the registration between the test image and face exemplar; and (iii) we use robust distance to a sub-space to allow for partial occlusion and expression change. The method is applied and evaluated on several feature length films. It is demonstrated that high recall rates (over 92%) can be achieved whilst maintaining good precision (over 93%).

Re:Some relevant research papers (1)

XSforMe (446716) | more than 9 years ago | (#12437832)

QBIC is part of IBM's DB2 content manager. It has been available for at least 5 years now, and is now part of a DB2 extender. You can check it out here:

http://wwwqbic.almaden.ibm.com/ [ibm.com]

that first one is so pointless (1)

QuantumG (50515) | more than 9 years ago | (#12438171)

why bother making an algorithm that can recognise which images are porn and which are not when you can just set up a web site where people will do it for free? It reminds me of those "enter the characters in this image" tests that places like Yahoo do to ensure you can't sign up for a million email accounts a day. They're so easy to get around cause all you have to do is present the image to a man who wants porn and he'll happily provide his character recognition skills without charge.

I for one (0, Offtopic)

kbjnash (761564) | more than 9 years ago | (#12437594)

tried a search for hot midget monkey sex... no hits? WTF?

ohh (-1, Offtopic)

Anonymous Coward | more than 9 years ago | (#12437607)

SWEET NIZZLE

and then we have reverse "Googling" for images.. (5, Interesting)

dotpavan (829804) | more than 9 years ago | (#12437669)

Here is a google game which is reverse of google's image search:

One has to guess the search word which generated a given set of 20 images in google's image search [robinson.name]

When things are moving forward, we have soomthing to talk about "those good ole days" but frankly the game is interesting initially but later gets boring due to the frequent repetitions..

$100 bill image (0)

Anonymous Coward | more than 9 years ago | (#12437675)

They might have to filter out currency searches- one lousy scan of a dollar bill might give results of high quality color images. There have been enough problems with amatuer counterfeiting already without making it any easier. Maybe I shouldn't even be posting this idea on /. Oh well...

Is it just colour? (4, Interesting)

Bifurcati (699683) | more than 9 years ago | (#12437676)

I just did a quick search based on this [designer.am] image of a Qantas logo (that's the main Australian airline, in case you're wondering...) It's red, with a white kangaroo in the middle. My theoretical aim was to find photos of Qantas planes.

What I got was an awful lot of red planes - some of which were actually Qantas planes, but I think more by coincidence (i.e., they're red) than design. Many images had nothing to do with Qantas, or even a red plane - they simply had a lot of red in the image.

This is impressive in some ways, but in others it seems like it's simply looking for similar patches of colour. I haven't done enough testing to see what happens if,say, I gave it a half red half green image.

Interesting, but not ready for public consumption just yet. A bit like A.L.I.C.E. the artifial intelligence system actually - neat, but not practical. Yet!

How about: (.)(.) (0)

Anonymous Coward | more than 9 years ago | (#12437679)

It had to be said

Damn filters. "(.)(.)" doesn't work

Great! (4, Funny)

SetupWeasel (54062) | more than 9 years ago | (#12437700)

Now I can find all the other naked pictures of Bea Arthur on the web!

eVision? (1)

Polarweasel (33867) | more than 9 years ago | (#12437701)

I guess eVision [evisionglobal.com] were just too early to market with their visual search engine. Here's a demo or three [evisionglobal.com] of eVe in action.

It sure was cool, just too far ahead of its time...

Several image viewers do this already (1)

Ezza (413609) | more than 9 years ago | (#12437703)

Programs like GQview (unix/linux) offer functions to search for similar images, mainly used to find duplicates.

It's not quite "put in an image and find me all the similar ones" but the underlying technology is the same, usually creating some kind of "signature" of each image and then comparing the signatures to find others visually similar.

Great for de-duplicating your por^M^M^MPhoto Collection.

IP Enforcement Nightmare (1, Interesting)

buckhead_buddy (186384) | more than 9 years ago | (#12437718)

The Mona Lisa (famous and out of copyright) is often plagarized in whole or in part as part of commercial or satiric artistic works. These types of visual database engines have frequently been explained to me as being able to input the Mona Lisa and get a list of images that used the entirety of the image or just a part (such as the highly-praised subtle smile).

The big problem to me is specifying input. I know the "look" of the Mona Lisa's smile, but even with the best pen input methods I'd never be able to mimic DaVinci's subtle emotion of the smile; my hands just aren't capable of doing so. Using photos of the painting could simplify this, but this almost assumes that I'm only looking for the parody's and commercial exploiters of the image rather than the image itself (after all, I have the image to start with). And it raises the further issue that many photographic reproductions of the Mona Lisa that I can get my hands on are still under copyright and I'd be doing something legally questionable with an image long in the public domain.

Add to this the "infinite number of monkeys" issue where legally litigious companies will use technologies like this to scan the internet for litigation targets. Imagine Disney using a cell of Rafiki from the Lion King to find legally similar images that were created after the Lion King was released even if they were only superficially similar. Now do this for all movies back to Snow White or Steamboat WIlly and you could get to be a real visual mob boss with ownership (or at least threat of litigation) over huge libraries of works that weren't even created to intentionally violate Disney "Intellectual Property".

My need for this technology is small considering the input problems I'd have with my artistic abilities, while the litigation nightmare from large databases of "similar" visual data would seem to be more bothersome than helpful. I rather hope these visual search and categorizing methods don't catch on.

_Mind at Light Speed_ by David Nolte (1)

Kyrka (20144) | more than 9 years ago | (#12437734)

I've not finished it, but I started a book called "Mind at Light Speed" by David Nolte a while back. He describes three stages of machines of light, and I can't do the book justice here.

However, he put forth the concept of replacing the bit as the common unit of data with actual images - best described as holographic images of light manipulated by light. A picture really _would_ be worth a thousand words in such a system!

You can do this on your desktop (0)

Anonymous Coward | more than 9 years ago | (#12437743)

Download the demo for Taos if you've got a good 3d card. http://www.taos3d.com/ [taos3d.com]

He was my professor, it sucked. (1)

dalamarian (741404) | more than 9 years ago | (#12437774)

Yeah, in my early undergrad days I took a 200 level course for a gen ed requirement of discrete mathmatics. Wang was the professor, and too this day I haven't had a course that was as difficult or completely freaking insane as the one he gave. Glad he is doing more research and less teaching.

This reminds me of Gibson's Pattern Recognition (2, Interesting)

WhiteDragon (4556) | more than 9 years ago | (#12437775)

Pattern Rcognition [williamgibsonbooks.com] is a novel by William Gibson, basically set in the present day or very near future. Image based search plays a central role in the plot. It's a very good read.

How does it do that? (2, Funny)

exp(pi*sqrt(163)) (613870) | more than 9 years ago | (#12437792)

I was looking at a picture of a plane on that web site and there was a link that said "Click for similar images". And what do you know? It brought up more pictures of planes. This is amazing stuff. How did it understand that I was looking at a picture of a plane?

more sophisticated than colour matching (1)

jtangen (861406) | more than 9 years ago | (#12437793)

Whatever algorithm they're using, it seems to be sensitve to the horizon line, colour, shading, orientation of the aircraft, etc. It seems to be operating at the level of a pigeon (who have been shown to discriminate photos depicting trees, water, and particular people - as well as art by Picasso and Monet. See http://www.pigeon.psy.tufts.edu/avc/huber [tufts.edu] for other examples. It will be some time before algorithms can match on the basis of model numbers and such. It took humans quite a while to evolve a cortex to enable such fine discriminations.

I drew a picture... (0)

Anonymous Coward | more than 9 years ago | (#12437804)

of a black and white, really crude shape of a plane from above...
It actually found pictures that were very similar!!!!

I'm imagining and interface in google image search that lets me quickly paint a picture then search on that... of course I'm using a wacom tablet, so it's much easier, but with the tablet pc's coming...

open source implementation (0)

Anonymous Coward | more than 9 years ago | (#12437957)

Check out imgSeek: http://imgseek.sourceforge.net/ [sourceforge.net]

I wonder what a search for GW Bush would produce (0)

Anonymous Coward | more than 9 years ago | (#12437960)

I wonder what an image search for GW Bush would produce, considering that he's the greatest leader the world has seen since Gidget.

There is a GNU project related to this GIFT (4, Interesting)

capedgirardeau (531367) | more than 9 years ago | (#12438007)

From gnu.org:

The GIFT (the GNU Image-Finding Tool) is a Content Based Image Retrieval System (CBIRS). It enables you to do Query By Example on images, giving you the opportunity to improve query results by relevance feedback. For processing your queries the program relies entirely on the content of the images, freeing you from the need to annotate all images before querying the collection.

GIFT [gnu.org] It worked pretty well for me in the demos they linked too. I have been waiting for this type of application to gain momentum.

It works! (1)

t0ny747 (849486) | more than 9 years ago | (#12438027)

Wow it works I sent a picture of a cockpit and I got cockpits back! It took a while but it worked :)

Sorry, someone else is currently using the system. (1)

Ninwa (583633) | more than 9 years ago | (#12438030)

Got this error when testing out the system, they certainly need to fix this, especially if they want it to become popular.

Would it work for animated .gifs? (2, Interesting)

Tzarius (688342) | more than 9 years ago | (#12438038)

'Coz I'm looking for more information on this image. [pacific.net.au]

It says "multi lock on" and a date, but all Google reports is other forum posts looking for the creator of the image. Apparently, there's a high-res version of it too.

Eyeball Addressable Image Browser. (0)

Anonymous Coward | more than 9 years ago | (#12438043)

http://www.cs.umd.edu/hcil/photomesa/ [umd.edu]

"PhotoMesa is a zoomable image browser. It allows the user to view multiple directories of images in a zoomable environment, and uses a set of simple navigation mechanisms to move through the space of images. It also supports grouping of images by metadata available from the file system. It requires only a set of images on disk, and does not require the user to add any metadata, or manipulate the images at all before browsing, thus making it easy to get started with existing images."

Apps Already do this. (1)

AgNO3 (878843) | more than 9 years ago | (#12438148)

iView Media pro already does this. You just tell it to find dupes and set the tolerence to loose. http://www.iview-multimedia.com/ [iview-multimedia.com]

It's db of airplane pictures, they are all similar (1)

skeptictank (841287) | more than 9 years ago | (#12438185)

I clicked on the search for similar pictures link for a 727 and one of the results was a helicopter.

I guess anyone can get a research grant these days.

Pretty controllable test (1)

omahajim (723760) | more than 9 years ago | (#12438214)

A.net is very stringent on the photos they accept. You can submit hundreds of photos, and get rejected for such things as 'badmotive' (a runway sign blocking a single tire), very mildly soft focus, and lots of other pretty anal things (IMHO). So while the image count they are dealing with is high, the obvious resulting similarity among images will result in a high number of matches.

Now, do this for something like Google Images or PBase or collections spanning infinite numbers of subjects and image sizes, then I'll get excited.

No, I've never had a rejection from A.net, I've never submitted there. Two minutes in their forum will tell you how anal their 'screeners' are, for whatever reason. It's just freakin' pictures of airplanes, for chrissakes.

Re:Pretty controllable test (2, Informative)

Anonymous Coward | more than 9 years ago | (#12438248)

It seems that a favorite use of the image similarity search over there at airliners.net is for the spotters to run pix on airline and flightsim sites through the search, to see who on anet has been infringed upon copyright-wise.

Look up Bombardier in the forums on airliners.net, they have frequently asked a photog for permission to use their photos (for pay), then later say they elected not to use them (and therefore no payment to photog). But then they use the photos anyways without payment or acknowledgement to the photographer.

So these spotters trawl the web looking for aircraft photos to 'vet' to see if they are stolen from an a.net photographer and band together to stamp out the piracy (sound familiar??????)

You can tell pretty quickly how they match (1)

skeptictank (841287) | more than 9 years ago | (#12438237)

They are doing it based upon the shades of color in the image. So if you query for a image of an aircraft in flight with a lot of white clouds behind it, you get more of the same, but you also get aircraft parked on snow-covered ground.

Music (1)

sn0wflake (592745) | more than 9 years ago | (#12438251)

So when can we search for music? I'm trying to find this song that goes like "dah daahh, dumpiti, dum dum"? Any answers?

Spotlight (1)

fr3nch13 (675181) | more than 9 years ago | (#12438258)

Imagine if this was incorporated into Spotlight, not having to rename your entire photo library.

Is this a joke? (2, Interesting)

Daikiki (227620) | more than 9 years ago | (#12438320)

I've tried two different images of airplanes; one of a bright red flying car on bright green grass and one of SpaceShip One against a deep blue sky. Both times, the results looked surprisingly like my query images in color composition only. Red planes on grass and white planes against a blue sky. Inauspicious start.

Next experiment: I took a picture of a highly distinctive plane, a harrier, climbing at a steep angle and viewed in profile. I got, in return, a list of passenger jets, and even a helicopter. Hardly surprisingly, all of the result pictures had the same bluish white sky as my original image. That was literally the only similarity.

According to the introduction on the search page the heuristics used compares colors, contrast and shapes in the images themselves. I saw no correlation whatsoever between shapes, and any correlation in contrast seems to be to be the result of the search engine simply looking for images that contain the same colors in a similar ratio to the original. In short, nothing to see here, move along.

On the other hand, one of the projects listed under the Penn State University link looks fairly fascinating. The Riemann a-LIP project [psu.edu] (automatic linguistic indexing of pictures) doesn't allow user input of images, unfortunately, but it does show some fairly fascinating attempts at verbally qualifying image data. For example, it describes a blue and orange mandelbrot as pattern agate shimer abstract scene, and a sunset over a lake as Berlin Devon Namibia landscape lake scene. Okay, it may still need some work, but it sure beats the hell out of the "find the same color airplane engine".
Load More Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Create a Slashdot Account

Loading...