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!

PhotoSketch Image Manipulation Tool Taking the World by Storm

ScuttleMonkey posted more than 4 years ago | from the photoshop-forum-goofs-rejoice dept.

Software 193

PhotoSketch, a new image manipulation program that combines stick-figure sketches, internet image search and pattern matching, seems to be spreading like wildfire. Created by five Chinese students at Tsinghua University and the National University of Singapore, the tool takes a basic sketch and simple labels and turns it into a polished image. "Although online image search generates many inappropriate results, our system is able to automatically select suitable photographs to generate a high quality composition, using a filtering scheme to exclude undesirable images," say the PhotoSketch team in an abstract outlining the tool. "We also provide a novel image blending algorithm to allow seamless image composition. Each blending result is given a numeric score, allowing us to find an optimal combination of discovered images. Experimental results show the method is very successful."

cancel ×

193 comments

correct links (4, Informative)

sopssa (1498795) | more than 4 years ago | (#29696053)

Since the link to homepage in the article is some old-dated one, here's a correct one:

And the binaries [tsinghua.edu.cn] (it's a few command line programs, so no fancy UI)

Re:correct links (5, Informative)

sopssa (1498795) | more than 4 years ago | (#29696077)

http://cg.cs.tsinghua.edu.cn/montage/main.htm [tsinghua.edu.cn]

Damn slashcode!

Re:correct links (1)

LeonPierre (305002) | more than 4 years ago | (#29696505)

In case of slashdotting, watch the video:

http://vimeo.com/6496886 [vimeo.com]

Let's download binaries from China! (5, Funny)

winkydink (650484) | more than 4 years ago | (#29696355)

Anybody else see what's wrong with this picture?

Re:Let's download binaries from China! (2, Insightful)

hoggoth (414195) | more than 4 years ago | (#29696811)

This looks really cool, so I downloaded the binaries. But I am going to try it out INSIDE a VM just in case.

Re:Let's download binaries from China! (5, Funny)

Anonymous Coward | more than 4 years ago | (#29696849)

Seem to run about right here at Fermilabs.

Re:correct links (0)

Anonymous Coward | more than 4 years ago | (#29696709)

What are the requirements for those binaries? .NET v??, vcredist...photoshop? Can't get them to run in my environment. Thanks!

Re:correct links (1)

uncqual (836337) | more than 4 years ago | (#29696939)

Oh, you can be sure they did run - they just act as if they didn't so you won't think of them when your ISP cuts you off for violating the TOS WRT SPAM.

Re:correct links (0)

Anonymous Coward | more than 4 years ago | (#29697017)

Well, I'm not totally crazy. Trying it inside a vanilla VM....which is why I figured my environment is lacking pre-reqs. But seriously this app looks too cool and I would love to try it out if anybody can verify that these binaries work and pass on what the pre-reqs are. Thanks again!

Re:correct links (0)

Anonymous Coward | more than 4 years ago | (#29696857)

What an excellent way to deliver a trojan horse, through a link slashdot comment, claiming the binary is a desired file.. ingenious....

Re:correct links (3, Informative)

mugnyte (203225) | more than 4 years ago | (#29696961)

Downloaded. NOTE: the as-compiled binaries require the OpenCV libraries [sourceforge.net] of the 110 variety (SourceForge holds the 200 version now). So, get older 110 binaries [sourceforge.net] . From the file list [sourceforge.net]

See the OpenCV Wiki [willowgarage.com] on setting up and checking you OpenCV installation.

I'm still setting up, but I'll post back when I get it working...

Sketched (4, Funny)

Iriscal (1563535) | more than 4 years ago | (#29696089)

This image looks sketched. I can tell from a few of the pixels, and from having seen a few sketches in my time.

Re:Sketched (-1, Troll)

Anonymous Coward | more than 4 years ago | (#29696337)

It's very sensitive. I drew a 3-pixel vertical line and it gave me a picture of an asian penis.

Re:Sketched (-1, Offtopic)

metlin (258108) | more than 4 years ago | (#29696387)

That long?!

Re:Sketched (-1, Troll)

larry bagina (561269) | more than 4 years ago | (#29696687)

Troll? He's got a point. Asian dudes have small peckers. That's why (well, one reason) asian chicks like white guys: larger penis size. Same reason White chicks like black dudes and black chicks like horses.

Ah so .... (0, Flamebait)

A nonymous Coward (7548) | more than 4 years ago | (#29696915)

Is that craptastic nonsense the contorted rationalization you had to come up with to justify your fantasies involving Asian boys?

Re:Sketched (0)

Anonymous Coward | more than 4 years ago | (#29697547)

OTOH, those little peckers are great for cleaning out pores. And my teeth have never been so clean.

What everyone want's to know... (3, Funny)

jo42 (227475) | more than 4 years ago | (#29696091)

If you sketch a big circle and two hands, will it come up with goatse?

Re:What everyone want's to know... (4, Funny)

aicrules (819392) | more than 4 years ago | (#29696287)

If you tagged the big circle with distended anus, then probably so.

Re:What everyone want's to know... (1)

j-stroy (640921) | more than 4 years ago | (#29696307)

I went to goat.cn [slashdot.org] which bizarrely links to slashdot WTF you sick people! instead of forwarding to http://china.u.cn/ [u.cn] and found this relevant image [u.cn] SFW

Re:What everyone want's to know... (1)

mace9984 (1406805) | more than 4 years ago | (#29696401)

I wonder how fast I can draw two stick girls, and one cup..hmmm

This sums it up quite nicely (4, Funny)

al0ha (1262684) | more than 4 years ago | (#29696101)

The authors of the program--Tao Chen, Ming-Ming Cheng, Ping Tan, Ariel Shamir, and Shi-Min Hu at the Department of Computer Science and Technology, Tsinghua University, and the National University of Singapore--presented it at Siggraph Asia 2009.

An event that will be remembered forever in the History of Humanity as the day in which a million dorks were finally able to put themselves in X-rated positions with Megan Fox.

Re:This sums it up quite nicely (-1, Offtopic)

Anonymous Coward | more than 4 years ago | (#29696373)

Rule 34 FTW!

Re:This sums it up quite nicely (3, Insightful)

NotBornYesterday (1093817) | more than 4 years ago | (#29696413)

An event that will be remembered forever in the History of Humanity as the day in which a million dorks were finally able to put themselves in X-rated positions with Megan Fox.

Many with decent Photoshop skills already can, this just lets millions more into the club without the need for a little know-how.

On a serious note, if this just outputs a flat .bmp, or .jpg, I just give it a "cool and fun, but not really useful". If this thing can output a .psd or .xcf with each element on a discrete layer, that would be excellent.

Re:This sums it up quite nicely (3, Insightful)

adonoman (624929) | more than 4 years ago | (#29696613)

Well, this being a research project / proof of concept type of thing, it's probably going to be bought up by a larger company (Microsoft, Google, Adobe) and made into a more useful bit of software. The actual output of this app is irrelevant - even if they composite the images into a flat image, at some point in time they've isolated the components and getting those components into different layers of some other image format is really a trivial extension. The important parts are really pulling useful images off the internet, and pulling together the important parts of those images.

Re:This sums it up quite nicely (1)

mikael (484) | more than 4 years ago | (#29697313)

They have so many directions to go in - extend it to work with video (just a sequence of images once you get past the codec part). What about that software Microsoft had written to combine separate pictures into a single image?

Re:This sums it up quite nicely (3, Interesting)

Wraithlyn (133796) | more than 4 years ago | (#29697743)

I'm amazed at how well this seems to automatically extract subjects from their background, something that usually requires a lot of painstaking manual work... honestly that's the real challenge of "photoshopping", becoming a ninja with the selection tools.

Re:This sums it up quite nicely (1)

Rei (128717) | more than 4 years ago | (#29696777)

I wonder how well it does if you don't give it realistic shapes (say, if you just draw a circle for everything)? If so, it'd be neat to have it draw elements from an RSS news feed, broken down into noun phrases and verbs, with descriptor-less nouns being combined with the verbs. I.e., it'd be a "Guess which news story gave you this crazy picture?" program.

"Surprised, humbled Obama awarded Nobel Peace Prize" -> "surprised, humbled Obama" "awarded" "nobel peace prize"
"Obama's Nobel: The Last Thing He Needs" -> "Obama's nobel" "last thing" "he needs"
"NASA probes hit moon twice; few pictures yet" -> "NASA probes" "hit" "moon" "few pictures"
"UCLA student's throat slashed in chem lab" -> "UCLA student" "throat slashed" "chem lab"
"2 die, 19 overcome at Arizona retreat sweat lodge" -> "2 die" "19 overcome" "Arizona Retreat sweat lodge"

Re:This sums it up quite nicely (2, Insightful)

NotQuiteReal (608241) | more than 4 years ago | (#29696919)

On a serious note, if this just outputs a flat .bmp, or .jpg, I just give it a "cool and fun, but not really useful". If this thing can output a .psd or .xcf with each element on a discrete layer, that would be excellent.

And a copyright release form. Or are snippets of other images non-infringing use?

In other words, it probably doesn't matter what the output format it, it will just be "cool and fun", but not for redistribution.

Re:This sums it up quite nicely (1)

Knuckles (8964) | more than 4 years ago | (#29697031)

If the photo agencies like iStock are smart they are going going to buy this, develop it and include an online shop.

Re:This sums it up quite nicely (1)

war4peace (1628283) | more than 4 years ago | (#29697631)

Yeah, but some comments emphasize on this software as allowing non-tech people to create interesting images. Well, not for now. Unless there's a nice easy GUI attached. I know this is just "proof of concept", but the video says something about the "intuitive user interface". Hmm, maybe it's just me, but I am still to try one :)

Re:This sums it up quite nicely (1)

arun84h (1454607) | more than 4 years ago | (#29696875)

You mean Mitchell Fox, right?

OH GREAT (5, Funny)

mujadaddy (1238164) | more than 4 years ago | (#29696127)

Now NO ONE will believe the pics of me with Jessica Biel, Kate Beckinsale and Dolly Madison are real!

Re:OH GREAT (0)

Anonymous Coward | more than 4 years ago | (#29697027)

You slept with James Madison's wife? Rock on!

That is because they are with me... wait, who are (2, Insightful)

SmallFurryCreature (593017) | more than 4 years ago | (#29697629)

That is because they are with me... wait, who are they?

You know your old when you don't regonize any of the names of today's hotties. And think they should cover up their bellies, do they want to catch a cold?

Good job slashdot btw, on holding out the sex comment so long.

Me, I thought of the porn possibilites when I read the first line.

Photoshopped Politics (4, Funny)

Anonymous Coward | more than 4 years ago | (#29696143)

This will make things way easier for Iran and North Korea.

How does it mask? (1)

Anonymusing (1450747) | more than 4 years ago | (#29696151)

How does it know which part of the photographs to mask out prior to composition? Have they pre-masked all the images in its database?

Re:How does it mask? (1)

BobMcD (601576) | more than 4 years ago | (#29696311)

Watch the video. It is pretty impressive.

Re:How does it mask? (2, Informative)

Carthag (643047) | more than 4 years ago | (#29696317)

It says in the Vimeo link. Not gonna summarize it cause just look at the damn thing

Re:How does it mask? (4, Informative)

StreetStealth (980200) | more than 4 years ago | (#29696347)

It appears from the video that it's running a fairly sophisticated series of algorithms to compare backgrounds and determine how difficult it would be to do a convincing mask-out of the foreground object, of which it appears to have a sort of heuristic expectation of shape from the user's sketch.

For instance, if your background is a grassy field, the user has requested a dog running, and you have a photo of a dog running over grass and a dog running over pavement, the grass one will allow a greater margin of error in the masking and thus it gets selected.

Overall, this looks like a fantastic step forward for computer vision, bringing the computer ever closer to the non-Cartesian way our brains see.

Re:How does it mask? (1)

xaxa (988988) | more than 4 years ago | (#29696925)

This is the kind of thing that inspires me to do a PhD in computer science. I think all the techniques in the video were covered at some level in lectures at university (mostly this course [ic.ac.uk] , IIRC) but seeing so much of it working in sequence with real photographs is impressive. (Of course, I saw impressive stuff at university. And this example is cool, whereas the most impressive computer vision work at Imperial College is medical, and I was too squeamish and usually distracted by the blood and skeletons.)

Alternatively, I need to find a job that stretches my ability, rather than an easy/comfortable one.

Cool idea, but lacks specificity (4, Interesting)

BadAnalogyGuy (945258) | more than 4 years ago | (#29696171)

I tried to draw a picture of a man with an erection. I labeled him "porn guy". Then I drew a picture of a woman with her mouth open and labeled her "porn whore cumshot".

The composite picture was fine except that the man and the woman were far apart from each other. In addition, even if I were to draw them closer together (hey, I'm working with a mouse here), the result would still have been sized incorrectly.

This technology holds lots of promise and is already pretty cool. I hope they can work out the kinks.

Re:Cool idea, but lacks specificity (3, Funny)

PolygamousRanchKid (1290638) | more than 4 years ago | (#29696293)

I hope they can work out the kinks.

Oh, don't be so boring . . . let them leave the kink in!

Re:Cool idea, but lacks specificity (1, Interesting)

Anonymous Coward | more than 4 years ago | (#29696301)

I wonder how exactly you did that, given there's no actual demo available and stuff.

Re:Cool idea, but lacks specificity (3, Funny)

Jesus_666 (702802) | more than 4 years ago | (#29697687)

You mean except for the binary on the website (linked to in another comment)?

Re:Cool idea, but lacks specificity (1)

clickety6 (141178) | more than 4 years ago | (#29696435)

I tried to draw a picture of a man with an erection.

I hope they can work out the kinks.

Buddy, you are the kink! ;-)

Re:Cool idea, but lacks specificity (-1, Troll)

Anonymous Coward | more than 4 years ago | (#29696459)

I hope they can work out the chinks.

I fixed that for you.

But at least you've discovered it's killer use... (0)

Anonymous Coward | more than 4 years ago | (#29696653)

I tried to draw a picture of a man with an erection. I labeled him "porn guy". Then I drew a picture of a woman with her mouth open and labeled her "porn whore cumshot".

If optimized for porn, people will have a field day with this.

BTW, Why "porn guy" vs "Porn whore" instead of, say, "port chick" ?

This is part of the problem with porn....

Re:Cool idea, but lacks specificity (0)

Anonymous Coward | more than 4 years ago | (#29696833)

I tried to draw a picture of a man with an erection. I labeled him "porn guy". Then I drew a picture of a woman with her mouth open and labeled her "porn whore cumshot". The composite picture was fine except that the man and the woman were far apart from each other. [...] This technology holds lots of promise and is already pretty cool. I hope they can work out the kinks.

That's why it put them far apart, to avoid being too kinky.

Re:Cool idea, but lacks specificity (0, Offtopic)

PenisLands (930247) | more than 4 years ago | (#29697483)

BadAnalogyguy, I sent you some emails but you never replied. Why not?

Do Want! (0)

Dr.Altaica (200819) | more than 4 years ago | (#29696203)

I think that's the only approprate response here.

This is unbelievable (4, Interesting)

rehtonAesoohC (954490) | more than 4 years ago | (#29696209)

The reality is that it was only a matter of time before someone came up with something like this, with examples like Microsoft Photosynth [livelabs.com] , but this is an unbelievable implementation.

I'm not 100% sure, but I can definitely see the potential for Google to snatch this up really fast and incorporate it into Picasa or even google image search or something. The fact that something like this allows anyone (not just artists) to come up with novel images with minimal effort is fantastic. I do wonder how canned the images were though. IE: did they GIS for an image first, then use the image as a basis to draw the stick figure, knowing that their algorithm would pick the image they selected in the first place? I would like to see a live demo with an unplanned audience member doing the drawing. Then I'll really be impressed.

Re:This is unbelievable (0)

Anonymous Coward | more than 4 years ago | (#29696535)

I drew a picture with PhotoSketch, then I GIS'd in my pants...

Viruses? (0, Troll)

arizwebfoot (1228544) | more than 4 years ago | (#29696223)

And this coming from China, there is no malicious code buried somewhere?

Re:Viruses? (1)

gmuslera (3436) | more than 4 years ago | (#29696315)

Probably will go viral, but it will be a social virus, not a "computer" one.

Copyright implications (5, Insightful)

capt.Hij (318203) | more than 4 years ago | (#29696233)

In related news anyone supporting current copyright laws have reinvigorated the economy after having to go out and purchase new pants. Cue the next great debate about copyright as we continue to try to shoe horn old ideas into the new world.

Re:Copyright implications, mashup (1, Interesting)

schwit1 (797399) | more than 4 years ago | (#29696543)

Isn't this the visual equivalent of a mashup?

Aren't mashups already in a copyright gray area?

Re:Copyright implications, mashup (3, Informative)

LordNimon (85072) | more than 4 years ago | (#29696675)

No. Mashups are clearly derived works, which fall under copyright quite clearly. Since this is in China, I'm pretty they're ignoring an IP laws and will probably get away with it. In the U.S., however, every one of those images better be licensed for royalty-free distribution, or they'd be sued.

When slashdotters get their hands on this program (2, Funny)

thewils (463314) | more than 4 years ago | (#29696259)

Tubgirl and goatse.cs are gonna crash.

The end of creativity (4, Insightful)

fluor2 (242824) | more than 4 years ago | (#29696341)

Soon I can write a story and then I just compile it and it will show sniplets of existing movies or rendered characters and woha it's converted to a real movie even with end credits: Directed and written by ME ME
Oh I can't wait.

Re:The end of creativity (1)

buswolley (591500) | more than 4 years ago | (#29696427)

You might be right. Except your characters will be changing faces/bodoes all the time. That might be fine if youre doing a remake of a ..ah damn what the name of that film again...my memory is shit.

Re:The end of creativity (3, Funny)

Carthag (643047) | more than 4 years ago | (#29696529)

A Scanner Darkly

Re:The end of creativity (2, Funny)

Draek (916851) | more than 4 years ago | (#29697149)

That might be fine if youre doing a remake of a ..ah damn what the name of that film again...my memory is shit.

Star Wars: A New Hope [slashdot.org] ?

Re:The end of creativity (1)

Fozzyuw (950608) | more than 4 years ago | (#29696697)

This is the internet... it'll soon be used for 90% porn fanfic.

Re:The end of creativity (1)

Hurricane78 (562437) | more than 4 years ago | (#29696749)

There's *always* some retard, who screams that it's the end. THEEE EEEEND.
Congratulations. YOU'RE WINNER! [youtube.com] (explanaiton [wikipedia.org] )

Re:The end of creativity (1)

GaryOlson (737642) | more than 4 years ago | (#29697209)

Directed and written by ME ME

Tom Cruise, is that you?

Re:The end of creativity (0)

Anonymous Coward | more than 4 years ago | (#29697211)

You are too late with your idea... this is how most Bollywood movies are made...

Another awesome example of the power of photos (3, Informative)

IamTheRealMike (537420) | more than 4 years ago | (#29696351)

That's pretty damn cool. It reminds me of scene completion [cmu.edu] , which is another take on the same idea - combining images from Flickr to create new images according to a brief sketch.

Re:Another awesome example of the power of photos (1)

Judebert (147131) | more than 4 years ago | (#29696483)

I had the same thought. I'll bet they're using a similar algorithm, with Google Image Search providing the images from the tags, and the sketch providing the foreground/background information.

Rough around the edges (4, Interesting)

jemtallon (1125407) | more than 4 years ago | (#29696353)

Having just taken a quick look through the config files and readme from the binary.zip file, it's pretty obvious this is very much a Proof of Concept release. You need to hard-code the number of sketched items, label them each in the config file, download the potential matched images to a specified directory, etc. It involves enough guess-work and too little documentation for me to proceed further, which is unfortunate. Has anyone else actually gotten it to work as described to confirm it does what it claims it can?

Re:Rough around the edges (3, Informative)

GigaHurtsMyRobot (1143329) | more than 4 years ago | (#29697681)

I don't think all of the binaries were included. After getting the right CV kit, the error is that some applications aren't there.

You make a Workspace folder and but a .jpg for each "item" with a simple line drawing. You make an ImageDownload folder, and in it a folder for each "item" with downloaded images in those folders.

It's missing exe's for these two lines from the ini file:

AttentioinCut = true

ShapMatching = true

They can be switched to false and it will run and create data in the Workspace folder. The images in the Segment folder are neat.... but looks like the executable to actually stitch the images together was not included.

Well What's The Point Then? (1)

BJ_Covert_Action (1499847) | more than 4 years ago | (#29696393)

Although online image search generates many inappropriate results, our system is able to automatically select suitable photographs to generate a high quality composition, using a filtering scheme to exclude undesirable images

Sounds like they took all the fun out of it to me....

Off the top of my head (2, Interesting)

MonsterTrimble (1205334) | more than 4 years ago | (#29696415)

1) Open Source?
2) Could the algorithm be used to find existing images similar to the one you just drew?
3) When is a demo of this thing available?

Hoax? (2, Insightful)

skeeto (1138903) | more than 4 years ago | (#29696497)

This seems to be either a hoax or will be extremely limited in ways they aren't discussing, as to have little use. If the examples they are showing are real, the image data set they are pulling from must have been manually processed and adorned with hand-made metadata.

This falls too much into the "too good to be true" category for me to believe it.

Re:Hoax? (1)

grumbel (592662) | more than 4 years ago | (#29696877)

Yep, the object masking seems a little to good to be true, the composition part itself on the other side seems doable once you have the objects masked out.

One thing however to keep in mind is that they have all the Internet to look for images, so they might only pick those images that make masking out objects easy and ignore the tricky ones.

Re:Hoax? (0)

Anonymous Coward | more than 4 years ago | (#29697541)

Well it seems to be true that they're presenting it at SIGGRAPH Asia 2009 in December at any rate, so if it's a hoax, it's a scam as well.
The system probably gives a lot of not-so-good results that they're not showing in the video, but that's just clever presentation, I don't see any reason to label this a hoax.

This is cool (0, Offtopic)

cld71 (7166) | more than 4 years ago | (#29696503)

This would be cool to use this software in a robot.

Yea, is this software opne source?

But we all know.. (1)

Seth Kriticos (1227934) | more than 4 years ago | (#29696557)

The application is very impressive, as far as the video goes. It shows that a human process of recombining existing material based on a hunch.

Problem is, searches (for base data set) for CC Share alike / commercially usable is a best spotty (many artists don't care much about explaining the image rights, and most others are jerks).

So in practice, this will only be useful for private entertainment, maybe prototyping, but not for professional use.

It's a great idea, actually a pretty innovative one, but it will break when facing western establishment.

But then again, it is created in China, so they'll have more space to act on it, which is kind of ironic taking in account their political system.

Interesting point: This research is in China (4, Insightful)

MaraDNS (1629201) | more than 4 years ago | (#29696597)

An interesting point: This research is being done in China, not the United States. Whatever happened to basic research being done in the US? Today's PARC laboratory is not in the US, but appears is in China.

This is not a good thing for people who live in the US. America's increasing dependence on outsourcing is destroying the US' capability to be competitive in today's environment.

The Harvard Business Review has an excellent article [harvardbusiness.org] about how America is destorying its own future.

Re:Interesting point: This research is in China (2, Insightful)

Anonymous Coward | more than 4 years ago | (#29697005)

America's core competency used to be research and innovative application.

Unfortunately, copyright and patent laws intended to protect the small from the big ended up getting flipped around. Now those laws are used by the big to crush the small, and along with it all the innovation that might produce a hint of competition.

America's future is bleak indeed unless one of two things happen:

1. A tremendous amount of domestic-only jobs (such as commercial driving from one US location to another US location) are created
2. The United States abandons NAFTA and the WTO in favor of bi-lateral (one country to one country) trade agreements that are slanted in the favor the US.

There is also a snowball's chance in hell of:

3. Every other country in the world adopts standards higher than the United States, and makes the US the outsourcing destination of choice, because of our comparatively light regulation and cheap labor.

Re:Interesting point: This research is in China (0)

Anonymous Coward | more than 4 years ago | (#29697559)

An interesting point: This research is being done in China, not the United States.

but seemingly was first published by German authors?

http://portal.acm.org/citation.cfm?doid=1597990.1598050

WTF?!? (1)

tedgyz (515156) | more than 4 years ago | (#29696601)

Great! They create the perfect pr0n tool and disable the feature. I'll wait for the haxx0red version from Russia.

This Same Technique Might Work For the Reverse (0)

Anonymous Coward | more than 4 years ago | (#29696617)

If we can figure out fast visual pattern matching, we might be able to take an entire
image and determine the stick figures and labels representation for it by matching up stored
image fragments to things present in the original. Then we build a system that does
it repeatedly, add a "consciousness" module that works by noting changes between
previous labels and current labels, and then add voice output to it so that it can say
"Hello Dave" when Dave walks into the room. That should freak some people out.

Pics or it didn't happen (1)

zindorsky (710179) | more than 4 years ago | (#29696637)

Pics or it didn't happen.

Re:Pics or it didn't happen (1)

Tetsujin (103070) | more than 4 years ago | (#29697033)

Pics or it didn't happen.

Pics we've got... (Well, a video, actually...) The proper bullshit-or-not test for this case would be to see the actual application.

Source or it didn't happen (1)

John Hasler (414242) | more than 4 years ago | (#29697163)

n/t

One word: (0)

Anonymous Coward | more than 4 years ago | (#29696851)

BOTNET

Yours in Petrograd,
Kilgore T.

xkcd (5, Interesting)

TheGreatOrangePeel (618581) | more than 4 years ago | (#29696881)

Wow. I'm instantly instilled with the urge to plug an XKCD comic into this and see what happens.

Re:xkcd (1)

techiemikey (1126169) | more than 4 years ago | (#29697085)

if you do, tell us how it goes.

Re:xkcd (3, Funny)

GaryOlson (737642) | more than 4 years ago | (#29697359)

A single XKCD comic? Apply to all XKCD comics, then turn all the comics into a streaming video. See if you obtain a higher level of conciousness.

Re:xkcd (0, Troll)

mhajicek (1582795) | more than 4 years ago | (#29697849)

Do it! Yaaarg! Ultimate killer app!

Image processing (2, Funny)

benjfowler (239527) | more than 4 years ago | (#29696993)

I suppose it won't work if you try sketching the Dalai Lama?

I've heard of academic projects on filtering out porn (Australian military didn't want people surfing smut on the clock). I'd imagine that filtering out pics of the Dalai Lama would be harder...

Smart idea (1)

mwvdlee (775178) | more than 4 years ago | (#29697083)

Compositions of copyrighted and uncleared images; just what we need. I assume those students who created it were law students?

thank god for this invention. (1)

nimbius (983462) | more than 4 years ago | (#29697219)

Ive spent months on 4chan trying to find hillarious pictures of zombie-dino-anonymous-jesus-hitler goatse that just DO NOT EXIST. With the latest technology from the interweb i can save several hours of stolen photoshop work and just sketch with the only hand i have above the desk the exact image of my pope-toilet-donkey-satan-pedobear-moneyshot-mobile. sauce, included.

yay, image manipulation... (1)

hitmark (640295) | more than 4 years ago | (#29697225)

why am i reminded of stalin getting people airbrushed out of photos?

and how long until someone uses it to put obama sharing a meal with osama bin-laden?

Turn stick figures into photos? (3, Interesting)

zindorsky (710179) | more than 4 years ago | (#29697271)

Someone should take all the XKCD comics, mark 'em up a bit, turn 'em into nice pictures, and .... Profit!!

Re:Turn stick figures into photos? (1)

ganktor (1448127) | more than 4 years ago | (#29697741)

Like the underpants gnomes? Step 1 - Steal underpants. Step 2 - .... Step 3 - Profit!!

!Slashdotted (0)

Anonymous Coward | more than 4 years ago | (#29697657)

It was on digg a couple of days ago and it wasn't working then. All the news stories say it's taking the world by storm but I've yet to hear of anyone whose actually used it. Don't they have mirrors in China?

Wouldn't load in WinXP ... (1, Funny)

Anonymous Coward | more than 4 years ago | (#29697805)

Downloaded the binaries, extracted the executables to my WinXP VM instance, and got the error, "The system cannot execute the specified program". Got .NET installed, so it must be looking for some C++ dependencies. Not pretty...

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...