×

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!

Automatic Image Tagging

CowboyNeal posted more than 7 years ago | from the on-the-horizon dept.

123

bignickel writes "Researchers at Penn State have applied for a patent on software that automatically recognizes objects in photos and tags them accordingly. The 'Automatic Linguistic Indexing of Pictures Real-Time' software (catchy name) trained a database using tens of thousands of images, and new images have 15 tags suggested based on comparisons with objects or concepts in the database. Not sure how you identify a 'concept,' and they're only talking about having one correct tag in the top 15, but still cool."

cancel ×
This is a preview of your comment

No Comment Title Entered

Anonymous Coward 1 minute ago

No Comment Entered

123 comments

Not shockingly... (4, Funny)

goldmeer (65554) | more than 7 years ago | (#16698175)

The vast majority of the images on the internets including The Google include "Pornography" in it's top 15 tags suggested. The accuracy rate is surprisingly high.

So how long before... (0)

Anonymous Coward | more than 7 years ago | (#16699541)

How long before Google buys them out?

---
Yes, +5 Insightful me!

Re:Not shockingly... (-1, Redundant)

Anonymous Coward | more than 7 years ago | (#16699917)

I think it's easy to identify a 'concept':

  • if the photo has a lot of people -> probably should be tagged as crowd
  • if the photo has a pair of boobs and dick -> tag it as sex etc.
  • if the photo has a lot of people and boobs -> tag it as a group-fuck etc.. =)

Tag for /. (-1)

Anonymous Coward | more than 7 years ago | (#16698219)

rating system down.

Re:Tag for /. (0)

Anonymous Coward | more than 7 years ago | (#16698265)

i prefer the 'yes', 'no', 'amirite', 'itsatrap' tags to the slashdot stories.

clearly the way that they were meant to be used!

Re:Tag for /. (1)

Amouth (879122) | more than 7 years ago | (#16698625)

"itsatrap" is the best... but use sparingly

Re:Tag for /. (0)

Anonymous Coward | more than 7 years ago | (#16699433)

"ponies" "lasers" "sharks" and "fud" are my favourites, "fud" being the most abused tag on the site..

That Sucks (2, Funny)

JerkyBoy (455854) | more than 7 years ago | (#16698221)

Researchers at a publicly funded institution are using their research results for personal (financial gain). Pennsylvania's tax dollars at work? How is this legal?

Re:That Sucks (0)

Skater (41976) | more than 7 years ago | (#16698257)

RTFA...it's not very long. "The university has applied for a patent on the invention." Any other conclusions you'd like to jump to?

Re:That Sucks (0)

Anonymous Coward | more than 7 years ago | (#16698271)

Universities shouldn't be _able_ to get patents. It's double taxation, plain and simple.

Re:That Sucks (0)

Anonymous Coward | more than 7 years ago | (#16698607)

How is that double? They still charge tuition, and unless they *enforce* the patent you haven't paid a dime.

Re:That Sucks (1)

tomstdenis (446163) | more than 7 years ago | (#16699057)

Hint: They do enforce patents, that's why they get them.

There is a way to document prior art. It's called a journal.

These are not defensive patents. They're designed solely to make the uni money. The problem? They made those patents with your tax dollars [at least the public unis]. So you have a public institution telling you that you can't use an algorithm or idea because they took YOUR money and patented it.

Tom

Re:That Sucks (1)

grimsweep (578372) | more than 7 years ago | (#16700175)

Actually, this type of research is often primarily funded by an external source. It's not unusual to have contracts (implied or otherwise) with private sources that stipulate what can and cannot be done with the results. The most your tax dollars have done is to provide them with a means of carrying it out in a well-equipped environment. Technically, they're still taxed on it and the majority of the profit goes to the school, which indirectly earns money for the state.

Arguably, this would be like taking issue with publicly funded roads and highways being used by private businesses to make money (i.e. the freight business). We're all paying for those roads, but without them they would be unable to make a profit.

I gather this wouldn't be such a point of contention if it a patent wasn't involved. This said, I believe this study [repec.org] agrees with your statement.

Re:That Sucks (1)

Gemini_25_RB (997440) | more than 7 years ago | (#16698609)

Double taxation in what sense? Okay, maybe some public money goes into the research fund (moreso at public universities than private), but how much? I was under the impression that collegiate level research was difficult because the professors are required to obtain their own funding through corporate sponsorships, grants, and private donations. Now the case _for_ patents. I believe that it is important that universities be able to patent their developments, otherwise someone else will. It simply becomes a free-for-all among the corporate world; the first company to discover the piece of research gets the prize!

Re:That Sucks (2, Informative)

Nybarius (799156) | more than 7 years ago | (#16698299)

Contrary to what you might believe, there is nothing unethical about making money. The government even gives out grants for entrepreneurs, and lets them keep all the profits; it's good for the economy, overall. The profit motive is a much more powerful incentive to positive social change than the goodness that lies in the hearts of men,

-Nyb

Re:That Sucks (0)

Anonymous Coward | more than 7 years ago | (#16699675)

Only when the profit motive is in-line with positive social change; but often it is not, which is why (say) organized crime, rape, etc. exist. Perfectly rational acts in a world where what's good isn't necessarily rational.

Re:That Sucks (0)

Anonymous Coward | more than 7 years ago | (#16699719)

The fallacy of your argument is that there is a differential - a group of a few people get control over the product of research that was publically funded. It is arguable that the economy would be better off if the entire public got a shot at using the fruits of their tax dollars to try to create some sort of successful business venture from it rather than a few elites.

Regardless, however, we funded that research, therefore, there is no reason why a chosen few should only be allowed to use it.

Re:That Sucks (1)

Manchot (847225) | more than 7 years ago | (#16698635)

Just because the institution is partially publicly funded doesn't mean that the research is. As a matter of fact, at many public universities, the big research groups have the "opposite" of public funding. As an example, use the University of Illinois. In theory, they're a public university. In practice, they get 20% of their budget from the state. This means that the big research groups in the College of Engineering, some of whom bring in millions of dollars a year, can end up pay up to 50% in taxes to subsidize the other departments. If you didn't allow professors to get patents, that's equivalent to telling them that they can't use their research results funded by their money. The result would be that at least some of the best ones would stay in the private sector.

Re:That Sucks (1)

megaditto (982598) | more than 7 years ago | (#16698707)

Penn State is a private school (privately chartered by the Commonwealth) despite what the name implies. It also receives less than 5% of it's funding from the state.

from http://www.psu.edu/ur/about/character.html [psu.edu]
Today Penn State is one of four 'state-related' universities (along with the University of Pittsburgh, Temple University, and Lincoln University), institutions that are not state-owned and -operated but that have the character of public universities and receive substantial state appropriations.


Interestingly, Penn State offers all residents in-state tuition at a loss, without being adequately compenstated by PA.

Re:That Sucks (1)

sprocketbox (636698) | more than 7 years ago | (#16698913)

I think how well Penn State is compensated by the state can be calculated in lots of different ways. I'm sure Penn State benefits from things like state maintained roads, sewer systems and other general public infrastructure. As an educational institution there is some chance that they don't pay property taxes which means that they don't pay in to the funds that pay for that same infrastructure.

All that said, I think software patents suck, no matter who is doing the patenting.

Re:That Sucks (1)

megaditto (982598) | more than 7 years ago | (#16699119)

Agreed. I was merely addressing GPs point about Penn State being a public school.

Regarding the compensation levels, Penn State receives less than the other 'state-affiliated' private universities in PA (though my info is somewhat dated).

And I agree with your statement about the software patents, though I think .edu's generally allow the public their free not-for-profit use. Think about FreeBSD's origin at the University of California at Berkeley.

And EULAs suck, but linux has one!

Re:That Sucks (0)

Anonymous Coward | more than 7 years ago | (#16699715)

What you just said, by definition, makes it a public school.

Re:That Sucks (0)

Anonymous Coward | more than 7 years ago | (#16700027)

UK's Oxford is public too then.

Penn State and Patents (0)

Anonymous Coward | more than 7 years ago | (#16698981)

If anyone cares, Penn State has a strict policy with patents, detailed here [psu.edu] among other places. It all comes from the fact that most of the University's reseach is paid for by grants and industry cooperation. I didn't RTFA yet, but I'd bet that this is going to be immediately licensed to either the federal government or other such body, whomever funded the research. Otherwise, it could very well become public domain.

Just how broad is a concept? (0)

Anonymous Coward | more than 7 years ago | (#16698267)

I think I could categorise most things using less than 15 (admittedly very broad) tags. Animal, person, plant, machine, sports, vehicle, furniture, book, etc.

Re:Just how broad is a concept? (1)

James McGuigan (852772) | more than 7 years ago | (#16700977)

I think I could categorise most things using less than 15 (admittedly very broad) tags. Animal, person, plant, machine, sports, vehicle, furniture, book, etc.
So which of these categories does a mushroom fit into?

The other 50% is the problem (2, Informative)

Heir Of The Mess (939658) | more than 7 years ago | (#16698273)

I've seen lots of systems like this. The problem is in the 50% of the images that don't work, so basically you have to manually tag 50% of your images.

I saw an interesting one about 10 years ago. It took an X-Ray image, did an edge detection, converted all the edges to a slope vs distance 2D plot, and conerted edge curves to a radius and distance plot, then used a kind of statistical correlation algorithm to pick which part of the body the image was from. I could imagine that you could apply something similar to the luminance of an image to pick out objects, and then maybe do some color transforms and stuff to improve results. The article says they do it in 1.4 seconds per image though, which is impressive.

Re:The other 50% is the problem (2, Insightful)

cloudmaster (10662) | more than 7 years ago | (#16699385)

Since you don't know *which* 50% it'll get right, though, you end up having to look at 100% to determine if the system got it right or not. At that point, it's only saving you a few seconds of typing / picking from a drop-down list. :)

Re:The other 50% is the problem (1)

llauren (80737) | more than 7 years ago | (#16700533)

If they did it with any kind of sense, they would not tag pictures they weren't confident enough of. But of course, i didn't read the fine article. This is /. after all :)

Re:The other 50% is the problem (1)

pe1chl (90186) | more than 7 years ago | (#16700851)

It is quite common to start building a system like this (image recognition, speech recognition, automatic translation, etc etc) and publish a press release stating that "the initial results are promising".
That is because the coarse approach to the problem is relatively uncomplicated, and after building some framework and inputting some reference data it is easy to make the system do some things right. Like guessing keywords correct for 50% of the input.

What is hard is to get it correct for close to 100% of the input. That is why you usually never hear again from such projects. The initial ramp-up in "promising results" quickly flattens and no noticable further progress is made unless a completely novel approach is taken.

XRay much easier though (2, Informative)

SuperKendall (25149) | more than 7 years ago | (#16699741)

The human body is pretty much the same between people, and XRays are generally shot from similar directions person to person - so the kind of check you are describing seems like it would yield high matches for pretty much any part of the body.

In the real world we have an object you might take a picture of from any angle, using a myriad of focal lengths, with variable levels of distorition depending on the lens and camera used. Really nasty for generic object recognition. I think the best we can hope for in terms of accuracy is perhaps some kind of facial recognition autmatically recognizing and tagging people in images.

Prior art? (1)

Sirch (82595) | more than 7 years ago | (#16698285)

Not RTFA to be honest, but can I claim prior art?

http://www.relle.co.uk/papers/2003-Content_Based_I mage_Retrieval__A_Words_and_Pictures_Approach.pdf [relle.co.uk]

We didn't have enough time to train the system properly, but itstarted off well...

Re:Prior art? (0)

Anonymous Coward | more than 7 years ago | (#16699689)

here's an earlier system that builds on public domain research

http://www.beholdsearch.com/ [beholdsearch.com]

the authors may in fact patent a particular algorithmic aspect of their system, rather than their approach as a whole

Re:Prior art? (1)

SausageOfDoom (930370) | more than 7 years ago | (#16700545)

I think lots of people have done this before - I did something very similar four years ago, and it's actually pretty simple (assuming you understand degree-level vision processing), especially if they're only aiming to get 1 tag right in 15. I think that this is just another example of the US patent system is getting out of control.

LIPS (1)

ch-chuck (9622) | more than 7 years ago | (#16698323)

You've got to hand it to those cunning linguists at Penn State.

Google, others doing similar research? (1)

EMIce (30092) | more than 7 years ago | (#16698339)

I'm sure a lot of research is being done in this area, in fact there is lots of interest implement this sort of thing in DSP for robot vision. How much of what this patent covers overlaps with what the others are working on? Is this something completely out from left field or does it fit the trend of where this area research was headed anyway?

Re:Google, others doing similar research? (1)

MikeFM (12491) | more than 7 years ago | (#16698955)

I did this as a project a few years ago when I was building something similar to, the yet unheard of, Flickr. Except I didn't limit myself to 15 tags. I just went through thousands of pics and tagged them with keywords I thought of as I did them and used that information to train a system that'd then go through and tag other pics. It wasn't perfect but did work pretty well and I think it could have been awesome. I tried pitching it to Google as a mashup of Google Images and Slashdot with powerful indexing and search features but I never got any response to any of my emails. Guess you have to have a PhD to get them to care about your ideas.

Anyway - it's not to hard to train a program to pull out things such as colors, shading, patterns, and shapes from pics and then from those it can learn what combinations are good indicators of certain keywords. For example, it's pretty easy to recognize most nudity as you'll get skin tones and usually get certain shapes (breasts are usually pretty similar in shape) so that with enough training it becomes pretty easy to pick those out of random images. Likewise it's pretty easy to pick out the shape of an airplane in a photo.

You can really do a lot with neural nets if you feel like bothering. The trick is to feed them the right information so that they can do something useful.

A video an the subject (2, Informative)

damgx (132688) | more than 7 years ago | (#16698347)

Luis Van Ahn did something almost the same, his idea though is to use humans aswell.

View the video on Human Computation [google.com]

Re:A video an the subject (1)

thebes (663586) | more than 7 years ago | (#16698851)

I thought this was going to be a bunch of chinese people in a basement calculating numbers and stuff.

Nobody is impressed by this (1)

indigest (974861) | more than 7 years ago | (#16698357)

FTFA:
The analysis takes about 1.4 seconds per image and in 98 per cent of tests suggests at least one correct tag in the top 15.
I suspect you could generate a list of 15 sufficiently vague words that would cover 98% of all images. Here's a start: people, sport, animal, trees...

Re:Nobody is impressed by this (0)

Anonymous Coward | more than 7 years ago | (#16698489)

Let me round out the list for yah:

people, sport, animal, trees, porn, porno, pornography, pr0n, boobies, b00bies, tits, ass, pussy, sex, girl

I swear I saw a butterfly (1)

packetmon (977047) | more than 7 years ago | (#16698365)

at least that's what I told the psychologist. Then on the second look, it looked like splotched ink on a paper that was then folded in half... I hope this software doesn't think like me cause at the end of it all, I saw a segfaulted X server on fvwm

API (1)

dampjam (779525) | more than 7 years ago | (#16698423)

I currently work for the group doing this - a very cool new feature will be launched in the next week that I am writing (stay tuned). Yes - this project has been done many times before by many people (to lesser degrees of success than this), but the thing to keep in mind is that this is realtime. It takes less than a second for the tags to be generated. All previous systems required a much larger amount of processing time. Check out www.alipr.com to try it yourself!

Re:API (1)

OzPhIsH (560038) | more than 7 years ago | (#16699913)

Just wondering what type of classifier you're using. Is it just an extension of one of the classical classification approaches, or a weird hybrid approach? Is there an actual research/conference paper those of us in the field can read instead of this fluff article? The real time nature of classifications makes me think you're handling the incoming images as stream data. Maybe somehow you can extend it to video, which is after all, nothing more than an image stream...

Re:API (1)

dampjam (779525) | more than 7 years ago | (#16700603)

I will ask them to upload it to alipr.com tomorrow... there might be a problem because it's published and now belongs to the journal. It was just presented at the ACM Multimedia Conference at UCSB, if you get those proceedings you can find it.

Make sure to check back tomorrow to be able to search based on the tags that the computer suggests, people verify, and ones that people enter manually. I just got all the cron jobs working together.

1 out of 15 ? impressive (1)

Gothmolly (148874) | more than 7 years ago | (#16698429)

How do they get less than a 50% average that you'd get by just guessing?

(yes, assuming a normal distribution of 'concepts' in the pictures, etc)

Re:1 out of 15 ? impressive (1)

CrabbMan (724775) | more than 7 years ago | (#16698791)

How do they get less than a 50% average that you'd get by just guessing? (yes, assuming a normal distribution of 'concepts' in the pictures, etc)

I hate to be a hater. . . but firstly, I doubt any sort of Gaussian or even mixture of Gaussians will work well to describe the distribution of picture labels. And secondly, you get a 50% average by making guesses about an even Bernoulli distribution, like a coin flip.

Re:1 out of 15 ? impressive (0)

Anonymous Coward | more than 7 years ago | (#16698837)

Here's my algorithm:

def guess_tags(image):
          import random, time
          random.seed(time.time())
          words = file("/usr/share/dict/web2").readlines()
          tags = ["nature", "man-made"]
          for i in range(1, 13):
                    random_word = words[random.randint(1, len(words))]
                    tags += [random_word]
          return tags

Re:1 out of 15 ? impressive (2, Insightful)

wayward_bruce (988607) | more than 7 years ago | (#16698975)

How do they get less than a 50% average that you'd get by just guessing?
How do you get that 50% is average on guessing? Their tag pool contains 332 "concepts", which means that randomly picking 15 would give you about 1/22 chance of getting a correct tag for a picture that is tagged with one word. For a two-tag image, you get 1/11. To get up to 50% you'd have to work with images tagged with four or five words. Did I miss something here? Besides, the claim is that "in 98 per cent of tests suggests at least one correct tag in the top 15", the keywords here being "98%" and "at least". We don't know how the number of correctly identified tags is distributed, so we can't say much about that anyway. This reminds me of Pres Eckhart and John Mauchly inviting a group of female "computers" to show them their first two blocks of tubes perform a computation of 5*1000. One of these ladies later commented that they had a whole lot of equipment for such a simple computation.

w00t!!! (2, Funny)

rts008 (812749) | more than 7 years ago | (#16698433)

Now almost 7% of my pr0n will get tagged correctly!
That's cool, the rest of it will be like opening xmas presents!

*file: 123456.jpeg>open>Aghh! Goatse!*

Hmmm...This may be neat when it gets a LITTLE more accurate, but a cool start none the less.
Kudus to the gang for getting a grip on a hard problem...erm..nevermind.

Re:w00t!!! (0)

Anonymous Coward | more than 7 years ago | (#16698527)

Of course, the question in everyone's mind is: Why would you have Goatse saved in your pr0n? :)

Re:w00t!!! (2, Funny)

CCFreak2K (930973) | more than 7 years ago | (#16698597)

You named your penis "problem?"

Re:w00t!!! (1)

rts008 (812749) | more than 7 years ago | (#16698825)

A lot of the time it is. Especially it's sense of timing.

Re:w00t!!! (1)

nsillik (791687) | more than 7 years ago | (#16699347)

I ran their software on the mental image after reading your comment:
Tags: Disgusting, small, horrifying, triceratops, Grenada

Three out of five ain't bad.

Re:w00t!!! (1)

rts008 (812749) | more than 7 years ago | (#16699481)

"Tags: Disgusting, small, horrifying, triceratops, Grenada"

Okay, I get the first three, but tricerotops aand Grenada? (disclaimer: I trained Spl Forces teams to go into Grenada- and yes, it WAS as fscked up as you might have heard!- so I may have a whole different point of view/perspective about Grenada than you may have :-) )

I'm intrigued, especially about the tricerotops, and would appreciate an explaination if you would be so kind.
(no sarcasm intended or implied- I'm really curious!)

Re:w00t!!! (1)

nsillik (791687) | more than 7 years ago | (#16699799)

It was an example of how poor the software (could) be. I'm not really judging the software, but most implementations I've seen give way-off responses like that.

The obvious (0)

Anonymous Coward | more than 7 years ago | (#16698439)

This enables a word search of images. Suppose that I'm doing research on the advertising for Cat's Paw Boot Products. This technology would enable me to find such images which are incidental parts of other images. In other words, it could turn up the fact that Fred's Boot Shop in downtown Moose Jaw had such a sign in its window in 1935. The photograph might be labeled 'Downtown 1935' and there would be no mention of Cat's Paw anywhere. It might even turn up the fact that Al Capone's old limo was parked just down the street. This is actually exciting news for researchers.

As the software said ... (0)

Anonymous Coward | more than 7 years ago | (#16698467)

"Ahh, but I don't see a goat in this picture?"


Snap ... it's down.

IBM was working on this years ago... (1)

rthille (8526) | more than 7 years ago | (#16698553)


Not sure how far they got, but remember reading that IBM was working on this and had some reasonable success at object recognition in images. I'd love to be able to classify the 10k digital images I've got around. Especially if it can recognize individuals (not that it would know their names initially, but would be trainable).

Re:IBM was working on this years ago... (0)

Anonymous Coward | more than 7 years ago | (#16698747)

Especially if it can recognize individuals (not that it would know their names initially, but would be trainable).

You should be able to do that with off-the-shelf facial recognition software.

Re:IBM was working on this years ago... (1)

Concerned Onlooker (473481) | more than 7 years ago | (#16700599)

Yep, it is called QBIC [ibm.com]--query by image content. The web site points to another web site or two that use QBIC for retrieving images from a collection.

Facial recognition is one thing, but if you just want to try to categorize your current collection you might try imgSeek [imgseek.net], which is a pretty cool program. Keep in mind that no one has really yet hit upon a great general purpose algorithm for finding matches to images or query by content. There is a large subjective component in categorizing images. If an image is mostly monochromatic blue and it's a picture of a boat, does it get classified as a boat or does it get lumped in with other predominantly blue images? How do you decide whether something is more blue than it is a boat? I suppose at that point a human has to step in to say "I'm really interested in the shape of the object more than I am the color right now."

Reportedly (2, Funny)

stunt_penguin (906223) | more than 7 years ago | (#16698621)

Reportedly the researchers showed the system a picture of a Death Star, and it correctly tagged the image with 'thatsnomoon'.

The system has clearly been let crawl the web for far too long.

Google will buy it... (1)

Duggeek (1015705) | more than 7 years ago | (#16698789)

Unless Jupiter Media [jupitermedia.com] gets to it first.

Someone like myself would understand the hours of data-entry and database development that goes into indexing imagery. I research photo copyrights for a living.

The fact that there is a feasible, automated system that can do the work will certainly cut down the man-hours for that sort of work; at least by half.

Pity, though. I heard that Google and others had a telecommuting thing that paid people to recognize what's in a photo. Sorry to hear they'll be out of a job soon.

Workarounds....... (1)

Anachragnome (1008495) | more than 7 years ago | (#16698813)

.........SexSurfer logs in to begin his daily search of the web to find more images to rip in an effort to increase his database of porn images, utilizing this technology, only to find that most of the images consist of naked women with political statements printed on their asses......

Seriously now, I am sure their are people out there that have already got ideas rolling around in their heads about how they can use this technology to hijack images to their advantage. Once somebody understands how the technology works it is only a matter of time before it is used for nefarious purposes, by means of "tricking" the technology. And in the process, invalidating any possible means by which the developers can realize a return on their investments.

Personally, I'd love to use such a technology(if it actually works) to sift through the plethora of "crap" images I have to search through on the web. It can be really frustrating to do a search only to find that a vast amount of the results are TOTALLY out of context simply because of the title tag attached.

Won't this... (0)

Anonymous Coward | more than 7 years ago | (#16698853)

Won't this take away many of the pr0n-sorting jobs and slow down the pr0n economy?

I for one will not stand for this!

One tag in 15 correct (1)

theeddie55 (982783) | more than 7 years ago | (#16698891)

well that really had to happen, i've just tried it and you can't really go wrong if one of the top 15 tags is 'photo' and another of the tags is 'thing'.

image spam filtering (1)

grapeape (137008) | more than 7 years ago | (#16698935)

Now all they need to do is come up with a way to recognize spam words in image text without the overhead of OCR and they can make a fortune on that alone.

Re:image spam filtering (0)

Anonymous Coward | more than 7 years ago | (#16700861)

well true current OCR are cpu intensive per image but they are used anyway... and guess what spammers realised that and they adopted new technique... they send now gif which paints "slowly" different part of image + they add noise to avoid text detection.. like with bayesian... they get antispam product and they find out where the crack is... and the loop continues... so antispam is good if you got antispam provider which updates techniques in the same pace as spammers..

Now all we need... (0)

Anonymous Coward | more than 7 years ago | (#16698961)

Automated dupe tagging.

I'm not... (1)

Morphine007 (207082) | more than 7 years ago | (#16699045)

... usually a pedant... but you don't train a database. It was likely a neural net, but TFA is rather thin on details. Anyone got a link to their paper?

Neural networks (0)

Anonymous Coward | more than 7 years ago | (#16699061)

Couldnt something like this be easily accomplished using neural networks.. I mean u will need a good dictinary and some base patterns to teach the network and then split up the image to compare for patterns. I am not an expert of image processing or neural networks but I have dabbled in some matlab based coding for both and it seems like a doable program the only problem I see is how to split up the image and at what point, edge detection maybe

Lots of implications for Surveillance, however. (1, Interesting)

Anonymous Coward | more than 7 years ago | (#16699497)

Makes it easier to process all that data generated by all those security cams.

Is there a "Big Brother" category on Slashdot, yet?

Wrong approach? (1)

havardi (122062) | more than 7 years ago | (#16699749)

This would be cool and all, but why not focus more on letting humans do the hard work-- like if I could take a picture of a tree and then press a button and say aloud; "redwood tree", and have that tag the file.

So they say... (1)

Ninwa (583633) | more than 7 years ago | (#16699803)

That's all fine and great that they can tell us, but why the heck couldn't they make a web-interface for it so I could try it out?

Bullshit Patents (1)

OzPhIsH (560038) | more than 7 years ago | (#16699811)

This just your standard data mining classification system, simply applied to image data as input, with the tags being possible classifications. This is an obvious application to ANYONE in the field. Software patents suck.

Re:Bullshit Patents (1)

heinousjay (683506) | more than 7 years ago | (#16699905)

If it's so obvious, how did they do it?

Re:Bullshit Patents (2, Insightful)

OzPhIsH (560038) | more than 7 years ago | (#16699997)

The application is obvious, although, I'll admit, their EXACT method isn't. But at it's core, it is basic supervised learning. Feed your classifier a training a set of images that are already tagged. Extract the features of the image and use those features to predict the tags. When the predicted classifications don't match the actual tags, adjust the model, rinse and repeat. Just pick up a data mining book. Like I said, lots of people are working on image classification, and this is an obvious application, at least to those in data mining/machine learning related fields. That doesn't make it an EASY thing to do successfully. If it were easy, there wouldn't be so much research going on. In that sense this group gets my respect for doing a pretty successful job. My concern is the patent. People already look at images and classify them based on content. That's what tagging IS. When computer software is written to automatically do something that every normal person does anyway, should that be patentable? How is this different than people giving Amazon tons of shit for their patents on their product recommender system?

confucius says... (1)

recharged95 (782975) | more than 7 years ago | (#16700093)

Considering the adage that "a picture is worth a thousand words", they're going to have a lot more words to index--where the words may not follow a specific taxonomy.

And that's one of the problems: does an image define the taxonomy or taxonomy defines the image [type]?

When it's going to be on Flickr? (1)

pigreco314 (194375) | more than 7 years ago | (#16700161)

I quickly run out of fantasy when it comes to assign tags to my pictures: an automated mass tag finder will save hours of my precious time while uploading photos to Flickr.

Neural Nets (2, Insightful)

gekoscan (1001678) | more than 7 years ago | (#16700351)

How can you take a neural network and train it, then patent that?
That's like patenting training a dog to fetch a stick, it's completely rediculous.

You take software capable of generalizing a neural network algorithm by feeding it pictures and associating each picture with certain tags. It then creates a generalized algorithm model based on what you fed it initially. So that when you give new input it is capable of outputting tags most similar to what you initially trained it.

So yes this software can recognize boxes, shapes, other objects, maybe scenes etc and associate them with tags... but ask them how the algorithm works under the hood =) They have no idea... a neural network is like a black box after it has been trained. You feed it input and it gives you output based on it's initial training. The inner workings are chaotic spaghetti values set on each neuron weighting and can't be deciphered.

How can you patent software that is a black box inside?

"Yes hello patent office? I have this box that manufactures microprocessors. I feed it all the materials and it outputs a shiny new processor. I am not sure of the manufacturing process internally but the output works great. I would like to patent this manufacturing process.

"Okay your patent number is 247286-"BLACK BOX"-9

The whole point of a neural network is it generalizes what you train it and can future predict any input based on that.

It's like having the invention of the first mirror and everytime someone put something different infront of it, that person called up the art gallery because they had a new painting that they wanted in their name (because depending what was in front of it you get a different reflection).

unimpressing (1)

Eivind (15695) | more than 7 years ago | (#16700637)

If it worked, it'd be very useful. However, getting the top 1 tag correct 50% of the time (which is the only quantifiable claim in the article) is pretty straigthforward. For most peoples photo-albums that can be done by the following AI-program: "print 'people'"

There's a few subjects that are so common that it's more or less a given they'll be in a large fraction of the photos. Outputting "people, buildings, nature, animals, plants, city" would probably give atleast 1-2 "correct" tags for 90% of whats in peoples photoalbums.

I had a class on neural networks and their (weak) sort of "ai", one task was to build a program to separate male from female names. The best programs could manage 80% or so, which is sorta decent. Until you realize that checking against static lists of the top 100 male/female names, if it's not in the list guess female if it ends in 'a', otherwise guess randomly will get you aproximately 95%. Furthermore, the latter program runs an order of magnitude faster, is more easily debuggable, can be understood by anyone, and can trivially be "extended" to reach 99% or more, simply by extending the lists of known male/female names.

link? (1)

Tom (822) | more than 7 years ago | (#16700765)

So where's the download link? How can software matter if I can't get it? ;-)
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>
Sign up for Slashdot Newsletters
Create a Slashdot Account

Loading...