Beta
×

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!

Seeing the Forest For the Trees

kdawson posted more than 4 years ago | from the i-can-see-clearly-now dept.

Software 64

swframe writes "A new object recognition system developed at MIT and UCLA looks for rudimentary visual features shared by multiple examples of the same object. Then it looks for combinations of those features shared by multiple examples, and combinations of those combinations, and so on, until it has assembled a model of the object that resembles a line drawing. Popular Science has a summary of the research. I've been working on something similar and I think this accomplishment looks very promising."

cancel ×

64 comments

Sorry! There are no comments related to the filter you selected.

rtfg (-1, Redundant)

Anonymous Coward | more than 4 years ago | (#32166832)

dfggrg FIRST POST! similarize this!

The plot thickens (-1, Offtopic)

For a Free Internet (1594621) | more than 4 years ago | (#32166852)

I have the sinking feeling that this was created by the Italians to facilitate their global dandruff conspiracy against me.

This is a realization of David Marr's early work. (4, Interesting)

bezenek (958723) | more than 4 years ago | (#32166858)

David Marr proposed the idea of a primal sketch as the first stage of converting the two-dimensional image on the retina to a full understanding of what is being looked at. This work culminated in a paper published in 1980 called "Theory of edge detection."

Marr was a faculty member at MIT, so it is appropriate for this work to have been done there.

For more information, see:

http://en.wikipedia.org/wiki/David_Marr_(neuroscientist) [wikipedia.org]

and

http://www.amazon.com/Vision-Computational-Investigation-Representation-Information/dp/0716715678 [amazon.com]

-Todd

Re:This is a realization of David Marr's early wor (-1, Offtopic)

DNS-and-BIND (461968) | more than 4 years ago | (#32166920)

You know, two people or groups can arrive at the same conclusion, because it was obvious in the first place. And why is it so appropriate? What if the work had been done elsewhere, would that be inappropriate or offensive?

Re:This is a realization of David Marr's early wor (2, Informative)

bezenek (958723) | more than 4 years ago | (#32167008)

You know, two people or groups can arrive at the same conclusion, because it was obvious in the first place. And why is it so appropriate? What if the work had been done elsewhere, would that be inappropriate or offensive?

I should have been more specific in my first post.

David Marr's vision book (published in 1982 after his early death in 1980) is considered a seminal work in understanding human visual processing.

Marr was trying to describe how humans see. The new work at MIT is trying to allow computers to see. David Marr would be glad to see the developments, whether at MIT or elsewhere.

-Todd

Re:This is a realization of David Marr's early wor (-1, Flamebait)

DNS-and-BIND (461968) | more than 4 years ago | (#32167180)

When you talk like that, it comes of as incredibly pretentious. I mean, people who actually pioneer the fields of human knowledge usually don't need to shout it from the hilltops.

PS please put your "-Todd" signature in your .sig file in your profile. That way, it can be filtered out by those of us who don't wish to view irrelevant information. I'm sure Marr would approve.

Re:This is a realization of David Marr's early wor (0)

Anonymous Coward | more than 4 years ago | (#32168504)

When you try to judge people that provide information instead of flames, it comes off as incredibly pretentious.

Yes, but... (1)

sycodon (149926) | more than 4 years ago | (#32170232)

...will it sort my porn collection according to the models?

Re:Yes, but... (1)

codeAlDente (1643257) | more than 4 years ago | (#32173744)

If it can, then we must conclude that women are objects.

Wow - that's strange... (3, Funny)

djupedal (584558) | more than 4 years ago | (#32166890)

I'VE been working on a similar project as well!

Maybe if enough of us with the same project interests get together, we can create an accurate summary of the parent!

You see? We could look at each other's projects for combinations of features shared by multiple examples, and combinations of those combinations, and so on!!??

This is amazing!

Re:Wow - that's strange... (-1, Offtopic)

Anonymous Coward | more than 4 years ago | (#32167054)

You all should start a newsletter! I find your ideas intriguing and would like to subscribe!

Re:Wow - that's strange... (-1, Offtopic)

Anonymous Coward | more than 4 years ago | (#32167568)

Sign me up too, honky!

Re:Wow - that's strange... (1)

fbjon (692006) | more than 4 years ago | (#32167120)

Also, the submitter is remarkably cool about having been scooped. :)

Re:Wow - that's strange... (0)

Anonymous Coward | more than 4 years ago | (#32167508)

I'VE been working on a similar project as well!

No, I'VE working on a similar project as well! And I'M Spartacus, as well!

Re:Wow - that's strange... (1)

Coren22 (1625475) | more than 4 years ago | (#32167906)

Well I am Patrick.

While you chaps theorise (4, Interesting)

Rogerborg (306625) | more than 4 years ago | (#32166916)

Honda just gets on with implementing it [youtube.com] . Oh, look, it's even got an automobile analogy: Asimo just did a drive-by on your research.

cool video! Re:While you chaps theorise (1)

Fubari (196373) | more than 4 years ago | (#32167932)

I was impressed the asimo guessed, "Maybe toy car?" when looking at the hand-sized mini-cooper model.
I did want to see them work on overlapping categories, like: is-a(toy) and is-a(car).

Re:cool video! Re:While you chaps theorise (0)

Anonymous Coward | more than 4 years ago | (#32169030)

Don't believe what you see in controlled environments where the scientist get to control the experiment. If Asimo was really as good as he seems to be in the video, it would be one of the greatest achievements in science. But the fact is that we are nowhere close to doing that. One of the scientists admitted as much at the end of the video.

Re:cool video! Re:While you chaps theorise (1)

Creepy (93888) | more than 4 years ago | (#32173324)

Well, they are teaching Asimo as if he/she/it were a small child, so the learning may take a while, and of course, honesty is important. I was impressed with the spacial analysis, given that it was much, much worse when I had machine vision. Processing was quite slow, IMO, but it is identifying general shapes like that of the toy car, which is good.

Ah, machine vision, where I had the most anal-retentive, closed minded professor ever. I don't regret not continuing with him for four more semesters and then finally being able to move up to a competent professor (the two that were most involved with research were quite good, or so I hear - I had one of them for one quarter for Algorithm Design class and he was quite good, and yes I meant quarter - I started college with quarters and ended with semesters).

Re:While you chaps theorise (1)

Abcd1234 (188840) | more than 4 years ago | (#32168404)

Yeah! Fuck the academics! What has research ever brought us, amirite?

Which Goal: AI or Cognitive Science? (1, Informative)

Garrett Fox (970174) | more than 4 years ago | (#32166942)

I've heard of various other approaches -- to two different things, and I'm not sure which one the researchers are mainly going for. Is the goal here to produce a useful vision system for AI, or to get a better understanding of how the brain works? It seems like while these are compatible goals, it's helpful to distinguish them and decide which you care more about.

A system can't just "learn" - does it use a GA? (3, Interesting)

Viol8 (599362) | more than 4 years ago | (#32166990)

Even neural nets have to be programmed at some level to exhibit behaviour that the programmers think will allow them to learn the task at hand unless these guys used some sort of genetic algorithm. The article doesn't mention it. Does anyone know?

Also it doesn't explain whether the system just recognises similar pictures to what its seen before - eg this picture looks like object type 123 (which to a human would be a horses rump) or whether it can combine all views of an object and recognise them all as that object , eg this picture looks like a horse. If its the latter how does it do it - does it have to be shown the object from a large number of angles or can it just infer from a couple of angles what the object would be like from many others?

Re:A system can't just "learn" - does it use a GA? (1, Informative)

Anonymous Coward | more than 4 years ago | (#32167224)

?There's nothing particularly unique about genetic algorithms with respect to learning? All systems operate within the constraints set by their programming.

Re:A system can't just "learn" - does it use a GA? (1)

Raffaello (230287) | more than 4 years ago | (#32167974)

The system is trained with labelled views of pre-categorized images, i.e., "this is one view of a watch, " "this is another view of a watch," etc. What makes it novel is that it is not told what the identifying features of a watch (or any other object are). It figures out for itself that the circular face, the stem, etc. are the distinctive features of a watch.

Re:A system can't just "learn" - does it use a GA? (1)

wurp (51446) | more than 4 years ago | (#32169184)

Some neural nets can learn on their own, without training. It shocked me too the first time I read of it.

Dammit, I can't find a reference now. The example I read about was classifying plants in broad and more specific types - the only input was data describing each of the plants.

Re:A system can't just "learn" - does it use a GA? (1)

MadUndergrad (950779) | more than 4 years ago | (#32178034)

Hey, that was my 8th grade science project!

Re:A system can't just "learn" - does it use a GA? (1)

wurp (51446) | more than 4 years ago | (#32178128)

Great, then you should know what the name is... ?

Cognitive science ahoy! (0)

Anonymous Coward | more than 4 years ago | (#32167088)

Anyone else notice how closely that line drawing of the stag resembled Palaeolithic cave art?

Re:Cognitive science ahoy! (2, Interesting)

Raffaello (230287) | more than 4 years ago | (#32167922)

No, because it doesn't.

Upper paleolithic european cave art used continuous, flowing lines [wikipedia.org] , created by spit-painting (think prehistoric mouth airbrush), not short, overlapping, straight lines [mit.edu] . The system described in TFA produces results that resemble the sort of lame, pseudo-cubist drawing one saw in art schools in the mid 20th c.

dogs etc (4, Interesting)

jrraines (1808940) | more than 4 years ago | (#32167122)

I certainly haven't worked in this area but for years have wondered how people including fairly young children recognize a dachshund, a bulldog and a great dane as dogs and other things as goats, cats, etc. Dogs are amazingly varied in shape and size and color. It seems like a VERY hard problem.

They just learn all the types (3, Insightful)

Viol8 (599362) | more than 4 years ago | (#32167310)

Theres no way just from looking at pictures of dogs that you could tell they're all the same species. There are some characterstics that some breads have in common with others (other than the obvious 4 legs etc) but they don't all overlap. With something like this its more than a simple case of pattern recognition - its aquired knowledge.

Re:They just learn all the types (1)

loufoque (1400831) | more than 4 years ago | (#32167660)

With something like this its more than a simple case of pattern recognition - its aquired knowledge.

Except pattern recognition *is* acquired knowledge.

Re:They just learn all the types (1)

Viol8 (599362) | more than 4 years ago | (#32168160)

No it isn't. If someone tells you "that is a dog" then you learn a simple fact without any patterns coming into it apart from being able to recognise what the particular breed of dog looks like in the future.

Re:They just learn all the types (1)

Creepy (93888) | more than 4 years ago | (#32173682)

And of course you didn't watch the video. What they said is vision has a lot of background processing, whether you realize it or not (maybe it happens during REM sleep? I don't know). It is why we can identify the differences between, say, a dog and a badger or know a Terrier is a dog and a Pug is a dog despite body differences.

If you walk into a room you've never been in before, how many items can you identify given 5 seconds? 15? 30 trips into the room for 5 seconds? My guess is the more exposure, the more you remember - basically, your brain can't handle the load and tries to pick out things of interest (if you have an eidetic you can remember the picture of the room and cheat, but most people don't have that) .

Re:They just learn all the types (1)

Raffaello (230287) | more than 4 years ago | (#32168008)

Learning is multi-sense. We recognize dogs as much from the sounds they make and their behavior as from pure visual appearance.

Attempts to do reductionist recognition (i.e., one sense at a time) are doomed to mediocrity. This system doesn't even categorize it's own training set - it has to be fed labelled (i.e., pre-categorized) images.

Re:They just learn all the types (1)

Arthur Grumbine (1086397) | more than 4 years ago | (#32169500)

There are some characterstics that some breads have in common with others (other than the obvious 4 legs etc) but they don't all overlap.

And this, my friends, is exactly why you should not use plutonium for baking.

Re:They just learn all the types (1)

Bones3D_mac (324952) | more than 4 years ago | (#32171740)

Acquired versus earned knowledge could well be a huge factor in getting something like this to work.

Perhaps rather than using one A.I. system to recognize everything, what's needed are hundreds or thousands of specialized A.I. clusters, all working on specifically recognizing one particular kind of object by gathering as much property data as possible.

Then, hand that data down to the next tier of A.I. clusters charged with recognizing several kinds of similar objects using pattern recognition on the data from the previous clusters to accomplish that task. If this level fails to identify the object at hand, it could dump the task back out to the previous tier to either gather more data or move onto a different A.I. cluster altogether. On the other hand, if it succeeds, and "kind of" recognizes some elements, it could send it's own pattern recognition data down to the next tier to be compared against pattern recognition data from other pattern recognition clusters in the same tier and have an A.I. cluster locate patterns in that data set.

I would think at some point, it would be possible for such a system to recognize any object it's ever encountered... or, using neural nets, attempt to make an educated guess based on the number of similar properties the object on hand has with the properties on file that had been used in attempting to identify the object.

Re:dogs etc (-1, Offtopic)

Anonymous Coward | more than 4 years ago | (#32167332)

When I was five years old, I thought that dogs and cats were the same species, and that dogs were the male of the species and cats were the females.

Re:dogs etc (1)

sznupi (719324) | more than 4 years ago | (#32167820)

Well, you had strong behavioral tips there, how those two animals typically interact with each other. A bit reversed in many ways in comparison to female - male relations...

Re:dogs etc (1)

Calinous (985536) | more than 4 years ago | (#32167896)

I'm not sure I've seen many cats until I was older (10 years maybe). We lived pretty secluded, only had a dog (and we only saw that dog maybe a couple of times a day or less). We had plenty of poultry though, and we played with chicks (we run after them mostly)

Re:dogs etc (1)

sznupi (719324) | more than 4 years ago | (#32167838)

It probably largely relies also on observation of behavioral patterns, most dogs have pretty similar ones. And this is the time when child soaks in, at tremendous speed, the "rules" of social life.

Re:dogs etc (1)

RobVB (1566105) | more than 4 years ago | (#32168604)

I [...] have wondered how [...] young children recognize a dachshund, a bulldog and a great dane as dogs

It probably largely relies also on observation of behavioral patterns, most dogs have pretty similar ones.

Or, in layman's terms, dogs go "woof".

4 legged animals = "doggy" (1)

snooo53 (663796) | more than 4 years ago | (#32169402)

Interestingly enough, very young children will typically generalize and call anything with 4 legs a "doggy" until they are corrected or shown the distinctions. That ability to generalize is one of the things that makes the human mind what it is. So yeah, it's amazing we can take an exemplar and be able to understand that other similar things of varied shapes sizes and colors are related but yet still understand species distinctions. My guess is that we're naturally very good at recognizing 3 dimensional features from many limited viewpoints

Re:dogs etc (1)

CecilPL (1258010) | more than 4 years ago | (#32169496)

It's also well known that children tend to have trouble with categorizing animals. Frequently they'll overgeneralize - calling every 4-legged animal a dog, for example - and it's only with constant correction that their category boundaries become adjusted properly.

Re:dogs etc (1)

Alarindris (1253418) | more than 4 years ago | (#32169558)

If you were to only identify things by how they look, yes.

For example, dogs bark and cat's meow.

Re:dogs etc (0)

Anonymous Coward | more than 4 years ago | (#32170784)

They learn it by being corrected... When my kids were young, every animal at the Zoo was a dog; probably because we had a dog. If we'd had a cat maybe everything would have been a cat.

Re:dogs etc (0)

Anonymous Coward | more than 4 years ago | (#32176072)

... and whose kid hasn't called a great dane or some other large dog either a horse or a wolf or something size-appropriate.

The pattern recognition / training episode just gets overlooked, or happened with Grandma or whatever.

System uses iconographic... (0)

Anonymous Coward | more than 4 years ago | (#32167134)

metaphors. i.e. legs, flat surfaces, etc.

The examples of car and a horse for example you could say the wheels of a car and hte legs of a horse are similar because they both provide functionality for movement even though they are differently shaped, they have similar relationships. I'm sure you could do comparing functions by using relationships the same way.

Smells like "On Intelligence" (0)

Anonymous Coward | more than 4 years ago | (#32167144)

Isn't this essentially the implementation of the ideas described by Jeff Hawkins in his book "On Intelligence"?

Training (1)

halcyon1234 (834388) | more than 4 years ago | (#32167866)

Okay, program-- this is a quark. Do you see it? Do you recognize it? Great. Get to work, and I'll be back in a few weeks to see how you're doing.

putting this into perspective (2)

jipn4 (1367823) | more than 4 years ago | (#32167920)

Hierarchical models of object recognition are decades old, as are attempts to implement them. This work doesn't yet work better than other engineering solutions, and it isn't obviously any more plausible than other approaches. So, it's a nice start, but it isn't a breakthrough.

Great (1)

Locke2005 (849178) | more than 4 years ago | (#32168260)

So now we've got an AI that can classify Justin Bieber as a girl too!

Not very useful (0)

Anonymous Coward | more than 4 years ago | (#32168812)

According to the article, it performs as well as currnet systems. In other words, very badly.

And here is the link to the paper itself (3, Informative)

S3D (745318) | more than 4 years ago | (#32169090)

Which should have been included into TFA from the start:
http://people.csail.mit.edu/leozhu/paper/RCM10cvpr.pdf [mit.edu]
The main achievement claimed is that no image labeling or any additional data like viewport position was needed, the learning process was completely automated.

Woohoo (0)

Anonymous Coward | more than 4 years ago | (#32170432)

automated crappy clipart!

I can't haz understands. (0)

Anonymous Coward | more than 4 years ago | (#32172858)

Can someone please post a car analogy, or reference it to Soviet Russia?

Let's wait and see... (-1, Flamebait)

Hordeking (1237940) | more than 4 years ago | (#32173036)

...how long it takes before Big Brother decides he can use this to track all of the "troublemakers" in large crowds (everyone is a troublemaker, according to the Gov't).

Re:Let's wait and see... (0, Flamebait)

Hordeking (1237940) | more than 4 years ago | (#32179866)

...how long it takes before Big Brother decides he can use this to track all of the "troublemakers" in large crowds (everyone is a troublemaker, according to the Gov't).

Apparently the Flamebait mod is now given to people who disagree with the Party...

Re:Let's wait and see... (0)

Anonymous Coward | more than 4 years ago | (#32182388)

...how long it takes before Big Brother decides he can use this to track all of the "troublemakers" in large crowds (everyone is a troublemaker, according to the Gov't).

Apparently the Flamebait mod is now given to people who disagree with the Party...

ah ha ha. ...or maybe just to those comments consisting of knowingly over-simplified-to-falsehood, generalizing, argument-inducing statements that don't contribute to the conversation and are about off-topic subjects, particularly "government"?

Re:Let's wait and see... (0, Flamebait)

Hordeking (1237940) | more than 4 years ago | (#32185292)

...how long it takes before Big Brother decides he can use this to track all of the "troublemakers" in large crowds (everyone is a troublemaker, according to the Gov't).

Apparently the Flamebait mod is now given to people who disagree with the Party...

ah ha ha. ...or maybe just to those comments consisting of knowingly over-simplified-to-falsehood, generalizing, argument-inducing statements that don't contribute to the conversation and are about off-topic subjects, particularly "government"?

It's an extremely valid concern. If I disagree with the Gov't's actions, I can't exactly take my business elsewhere, can I? Sure, I can go find another one, but I can't go without one, or start my own (with the idea of doing it right). A private company can't fine me, put me in jail, nor can it execute me. The government can. That's why I usually direct concerns at governance about really great technology (it is really cool). But I don't want it turned against me. And if the government turns it against me, it's far more likely to be successful.

Re:Let's wait and see... (0)

Anonymous Coward | more than 4 years ago | (#32187334)

Can I get a mod up? Someone is trying to mod me into oblivion.

What's an object, anyway? (1)

codeAlDente (1643257) | more than 4 years ago | (#32173918)

One obstacle toward progress in this field is how to define an object. Are electrons, atoms, molecules, proteins, cells, leaves, trees, forests and planets all considered to be objects? And who gets to decide - a bunch of undergrad test subjects who draw lines around pictures and give names to each image segment? This algo separates objects and parts, but (from what I can tell, having read the article but not the paper), there's no big reason to say one thing is an object and another thing is a part. Seems pretty arbitrary from a philosophical standpoint, but no doubt its practical value is appreciable.

Re:What's an object, anyway? (0)

Anonymous Coward | more than 4 years ago | (#32178560)

You should read "On the origin of objects"; to me the absolute best book on the metaphysics of AI/CogSci/Comp Sci/Psychology.

Solitaire (0)

Anonymous Coward | more than 4 years ago | (#32179232)

I wrote a solitaire player, and then I attempted to teach the computer how to play "legal" solitaire by looking at the moves. I intersected every move with every other move, and I could have gone higher (in the intersections), but my poor PC was not capable enough. This is looking at moves like you might construct an inheritance tree. Moving common items to a higher level in tree.

John

Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?