Researchers Develop 3-D Search Engine 140
An anonymous reader writes "Researchers have developed new search engines that can mine catalogs of three-dimensional objects, like airplane parts or architectural features. All the users have to do is sketch what they're thinking of, and the search engines can produce comparable objects."
Is this a good thing to be releasing? (Score:4, Funny)
Depends too much on (Score:4, Interesting)
Re:Depends too much on (Score:5, Funny)
What a coincidence. I used to teach a Spelling for Amateurs class, and most people couldn't even spell the name of the class.
Re:Depends too much on (Score:1)
"Armature: A protective covering, structure, or organ of an animal or a plant, such as teeth, claws, thorns, or the shell of a turtle.
However, when they graduated, they could draw claws like a pro.
Re:Depends too much on (Score:1)
Don't assume all writers spell flawlessly. They have editors and proofreaders who are supposed to pick out all the mistaeks. Luckily, I can spell flawlessly
Re:Depends too much on (Score:1)
Re:Depends too much on (Score:3)
Re:Depends too much on (Score:2)
1. It was a CLASS you TEACH...therefore you are there to TEACH them to draw.
or
2. After the class that you taught they still couldn't draw...who's fault is that really? There's or yours?
Also, how is this modded as interesting?
Re:Depends too much on (Score:2)
Do designers/engineers/draftsmen actually learn to draw these days, or is it all CAD-based?
Re:Depends too much on (Score:2)
Re:Depends too much on (Score:2)
Anyway, back on topic, there's no question that CAD is more precise but during early design stages, precision is not the goal and too much focus on computer-generated perfection hinders the process. People tend to get wrapped-up in the workings of the software and can produce brilliant-looking, but comepletely stupid drawings and unworkable designs.
My opinion is that CAD people should learn how to draw by hand
Re:Depends too much on (Score:2)
Yes, it is. Also, you can find a site containing photos as well as audio clips from his infamous technique for lighting a charcoal grill here [doeblitz.net].
Dupe (Score:3, Informative)
Re:Dupe (Score:1)
Re:Dupe (Score:1)
Re:Dupe (Score:2)
Re:Dupe (Score:2)
I drew "/. x II" and got lotsa hits!
Re:Dupe (Score:1)
Its not similar, its the same! This time a different news agency (CNN) is reporting it. So much for ``Breaking News"
again ? (Score:1)
Re:again ? (Score:2, Informative)
Well it's not exactly the same thing, the article posted before was about Purdue's shape searching engine [slashdot.org] while this article is about Princeton's 3D model search [princeton.edu]. Same topic, different search engines.
Finally... (Score:4, Funny)
Oh, wait
Joke? (Score:2, Funny)
Re:Joke? (Score:2)
Re:Joke? (Score:1)
Yay (Score:5, Funny)
Maybe something for ./? (Score:5, Funny)
How long before (Score:5, Funny)
The possibilities are endless.
Re:How long before (Score:2)
Re:How long before (Score:2)
Re:How long before (Score:2)
If I were in that market, I'd LOVE it if people could sketch what they wanted to see. There are probably a lot of extremely lucrative niches out there waiting to be tapped!
Re:How long before (Score:2)
With my sketching abilities I'd probably have a wealth of pictures in the genre of:
And don't forget the shock photos held on that infamous caprine-named site [zoy.org] that would be returned just by being a bit too loose with one's sketch.
Long live metamods (Score:1)
Sneezing Gay People [angelfire.com]
pr0n (Score:4, Interesting)
Seriously though the pr0n industry is an extreme early adopter of most technologies, I'm sure that the researchers could fund research for the rest o their lives by creating an adult search engine.
John.
Re:pr0n (Score:3, Interesting)
Re:pr0n (Score:1, Troll)
> so that she looks different enough from the others
> that the search engine can find her?
Damn, you've uncovered by secret super-power. Oh, you didn't think there was an X-man with the amazing porn star drawing mutation? How wrong you were...
John.
Re:pr0n (Score:1)
I'm not too good at sketching, (Score:4, Funny)
Re:I'm not too good at sketching, (Score:2)
Re:pr0n (Score:5, Insightful)
> "industrial objects". Riiighht.
I'll take the sinister uses with the good on this one. This is really exciting for engineers and tinkerers alike, because it means less time reinventing the proverbial wheel.
Let's say I need gizmoX but it's not manufactured anymore. Fuck! No knock-offs, either. The local CNC shop says they can make it for me but they need a model. 3D modeling could take a while unless you have something similar to start with.
So instead of PORING over an industrial parts catalog and missing the one close-enough part you needed but didn't expect to find under "444_T91_fillets" or whatever, you just search for matches using dimensions, materials, etc. Just a few minor tweaks to the mesh, export, send to the shop, and you just saved a lot of time.
That said - the design and manufacturing industries have been writing ad hoc programs for searching for parts and tooling since the proliferation of computers, but it's nice to see that a general-purpose algorithm that could be more or less universally applicable is evolving from the efforts thus far.
Re:pr0n (Score:2)
I don't think you need to go so far as editing to find a good use for this.
Easy (Score:2)
Looking for female porn? Enter (*)(*).
Male your flavour? Try 8=====) (or 8==========) if you're looking for a more "exciting" experience).
Pattern match, and boom! Instant porn results.
And no, I cannot believe that I actually posted some ascii pr0n on Slashdot
Re:Absolutely wrong (Score:2)
Just under two minutes... but then, I play a lot of iSketch. [isketch.net] The main bottleneck was coming back to make sure I hadn't missed any terms from your description.
Two minutes is a long time, but if the search engine could work off my picture then I would get the pose in it, instead of just ANY picture of lesbian grannies tied up
accuracy.. (Score:4, Insightful)
Re:accuracy.. (Score:1, Funny)
i dunno about you, (Score:3, Insightful)
Of course, this could be used to search from a scanned in image, which would be good for things like finding car parts, which we had a story about a couple of days ago. Old news, slightly different subject.
Re:i dunno about you, (Score:1)
Dok
Previous /. article (Score:4, Informative)
Re:Previous /. article (Score:2, Funny)
http://slashdot.org/comments.pl?sid=104348&cid=88
Prior Art (Score:1)
obligatory pr0n usage (Score:1)
is it really easier to use? (Score:2, Insightful)
Re:is it really easier to use? (Score:3, Insightful)
It's easier to sketch the part than to remember that the guy who created the drawing called it a "3/8ths Gripley"
Shit... (Score:4, Insightful)
Picasso? (Score:1)
Your english skills leave something to be desired.
Note to Slashdot editors: (Score:1, Funny)
Killer App (Score:2, Funny)
Re:Killer App (Score:1)
Even 3D developers will be confussed (Score:3, Informative)
Turbo squid is better [turbosquid.com] for artists.
And 3D objects don't get better with age (at least not yet)
This will be GREAT for the production industry (which has moved off shore) as users will be able to forget their skills of automatic recall when it comes to part recognition and sucumb to the all mighty 3D shape database.
Porn (Score:1)
Any neural net people here? (Score:3, Interesting)
Re:Any neural net people here? (Score:4, Interesting)
Aside from the fact that about a third of the brain (if I remember correctly) is dedicated to visual processing and that the brain has a hundred billion neurons, each neuron firing at up to 1KHz and with thousands of connections to other neurons?
It's an enourmously complex task to be able to do this reliably. Recognizing a straight-on photograph and matching it up with a corresponding mugshot is a whole lot different than seeing the side of a person's half-shadowed face from slightly behind them and recognizing that as the same Alice who held up the convenience store on Tuesday.
Handwriting recognition is orders of magnitude easier to do, since it's a lot easier to recognize similarities between a sample of writing and previously analyzed samples. In fact, there's a lot of research into this [ibm.com] already. Facial recognition is slowly getting there, but people are still stumbling over the same mistakes that were made decades ago -- attempting to formalize facial recognition by defining a set of rules and matching to those rules. Much like spam filtering, this works to a degree until the differences between the one you want and the one you don't no longer fits within the rules.
Neural net + genetic algorithms/programming to refine the net's connections and behavior is probably a good approach. Finding a means of populating, storing, and computing hundreds of billions of nodes in the network is the real challenge I think. Of course, take everything I've said with a huge grain of salt as this is a subject I'm fascinated by but have little practical experience with.
Sorry to drag Bush in to this... (Score:1, Funny)
dumbass (Score:1)
this from the chief scientist. what an idiot.
Clippy for the drawing impared (Score:5, Funny)
"I see you are trying to draw a monkey wearing what looks like a condom on it's head... May I assist you in finding such an item?"
"Click'
"Find books about Monkeys With Condoms on their Heads at Amazon.com"
"Find the lowest priced Monkey with Condoms on their heads with Price Grabber"
Bring it.
Re:Clippy for the drawing impared (Score:3, Funny)
The first book that comes up: "The Trouble with Islam : A Muslim's Call for Reform in Her Faith". I have no idea what this means, but it has to be significant in some way.
Very Funny, Almost Ridiculous (Score:2)
Beta Results (Score:5, Funny)
+---------+
| |
| (.Y.) |
| ) ( |
| ( Y ) |
| |
+---------+
Re:Beta Results (Score:1)
3D Search Engine Screenshots (Score:3, Informative)
Old News.. (Score:1)
Obvious use... (Score:1)
The I saw the term "architectural features" and relized that they already had.
dupe? (Score:1)
http://slashdot.org/article.pl?sid=04/04/04/142
Glad this got posted (Score:1)
Still, I'm glad this submission was accepted. The next thing we need is a robust audio search engine.
try it. (Score:2, Informative)
3d model search engine [princeton.edu]
Re:try it. (Score:2, Informative)
Dok
Intangibles (Score:1)
Where was this when I was in 8th grade... (Score:2)
News? (Score:2)
Great! (Score:2, Funny)
4D engine, anyone? (Score:1)
Booble?
I like the idea (Score:1)
It seems they target the industries only,
but somewhat earlier I've been thinking too that we need some other way to crawl / search data then we've been doing so far or at least a more advanced way to get specific data. (granted; google does what it does, and does it extremely well)
Conside
Comment removed (Score:3, Funny)
Come on... (Score:1)
I don't believe this technology exists yet until OCR is perfected.
We already have that. (Score:1)
This story seems oddly familiar.
Cynical Old Man Perspective (Score:2)
Re:Ok I'm thinking of a number (Score:2)