Slashdot Mirror
Colorizing Images and Video by Scribbling
Guspaz writes "Up until now, colorizing a video or image has been a painstaking and mostly manual task. However, researchers in Israel have come up with a new way of colorizing images just by making a few scribbles. The technique works on the premise that 'neighboring pixels in space-time that have similar intensities should have similar colors,' and also allows colorization of videos by 'marking' about one in ten frames."
they extended the paint bucket to the 4th dimension?
I'm colorblind! That whole site is very confusing! :P
But does the fill utility in paint work across time as well?
Online Starcraft RPG? At
Dietary fiber is like asynchronous IO-- Non-blocking!
Back when voyager(s) were flying by planets I recall reading how the cameras worked. From what I remember, the cameras actually capture images in black and white. The cameras can detect much more "color" depth than color cameras could (or can?). The scientist would process the pictures to colorize them, you identify one area of color you know and the algorithm would process the rest of the 1 billion shades of gray into a color mapping for people to view. Now why cant identify this gray shade as the color red; anytime you see it then that is red. Go on for each color spectrum or have the algorithm adjust what a little red hue is for a given little hue of gray. It appears that is what the scribbles are doing which is quite clever and the algorithm doesn't have to work (guess) so much.
Personally I can't wait until there is a Photoshop filter for this. :D
Support Liberty, Support Ron Paul
it doesn't just fill an area where all the pixals are the same, instead it finds similar pixals, and fills them, and varies them according to what they where orginally like? [p] Otherwise title should be "Over payed digital art programmers discover MS paint". [/p]
"I may be full of crap about this game, and I may be wrong, and that's fine." -Jack Thompson
I wonder if the actual quality will improve, though -- colorized films still look, well, colorized. You can tell in 1 second...
One man's Funny is another man's Offtopic.
from looking at the before and after images, this technique looks pretty cool and will probably have applications for recoloring an image that is already color. For instance, the image where he recolors the fabric on the chair.
The emphasis is on blends and shading. Take one of those pictures and try it in paint... I'm pretty sure the result will be QUITE different!
Video Compression !
Only save the intensity channel and a few bits of markup and you compress the stream quite a bit.
The bikini - security through obscurity since 1943
Now it's even easier for corporations re-releasing films to completely destroy the original beauty of a film by adding unnatural and unnecessary color!
Coming soon, new dubbing techniques will allow easy substitution of the original actors' voices and dialogue with trite teen-angst to appeal to younger generations.
Their site is going to get slashdotted real quick. Lots of content on there. It's beautiful. It's ... it's amazing what they've done with so little effort.
I am so curious what this could do for so many old movies...
Reinvent the wheel only at either a lower cost, greater effectiveness, or your own personal enrichment and satisfaction.
Here is the coralized mirror.
Avantslash: low-bandwidth mobile slashdot.
...and raise enought money to rebuild the smoking ruins of their server room.
I'm not wrong. You haven't thought about it hard enough.
The folks at slashdot can take down a webserver just by making a few scribbles on their website.
Isn't that just a fill tool? Paint does that.
No. Fill just goes until it meets a boundary. This colorization is a lot smarter than that. It appears to notice the boundarys by the sudden changes of the temperature in the color of pixels. That way it can then make an educated guess on how much to color and when to stop. You can then optimize this by putting in more than one input of the colors you want to change. This effect is really quite amazing. Scroll down and look at the gif video of the birthday party. JUST AMAZING.
I'm a virgo and on Slashdot. Coincidence? Yes.
Already done. It's called NTSC!
Looks like a very cool idea. The videos are pretty amazing, however; the recolor job in the birthday party clip makes it look like a bad acid trip.
http://66.102.7.104/search?q=cache:xOH_jKTvBeEJ:ww w.cs.huji.ac.il/~yweiss/Colorization/+&hl=en
FoundNews.com - get paid to blog.,
http://www.mirrordot.org/stories/b824e8605782b61d2 dca3ca678065972/index.html
Read about it earlier this week, it looks cool, colorization looks nice, but with so few samples its hard to tell if it will work with a wide range of inputs, or only ones with contrast ratios like those shown.
Video Production Support
http://www.mirrordot.com/stories/b824e8605782b61d2 dca3ca678065972/index.html
gee that was quick...
http://mirrordot.org/stories/b824e8605782b61d2dca3 ca678065972/index.html
I agree. If you only had to compress B/W video and run it through a filter like this, Less bandwidth would be needed.
This work is very similar to some work that was presented at last years siggraph using graph cut optimization titled Interactive Digital Photomontage by some researchers at the University of Washington. This stuff is really cool and has applications outside of just re-coloring black and white. For example, compositors in the film industry adjust the color composition of scenes that were filmed during the day to look like they were filmed at night. Sometimes they just need to tweak the color because the art director isnt happy with it. Other times it's because they introduced CG elements into live action scenes and they dont quite match. If they can tweak those colors interactively, without authoring masks, it is faster than re-rendering the scene and that saves money.
Very cool stuff.
Pete
What's a sig? Pete Brubaker
"What's wrong?"
"It's as if millions of lawyers stampeded the patent office and then suddenly... prior art."
Want to improve your Karma? Instead of "Post Anonymously", try the "Post Humously" option.
I'm not sure if this was intended to sound like some new ground breaking technique, but it really isn't. I am a masters Electrical Engineering student and am currently taking an Image Processing class. Using neighboring images to reconstruct an image is a VERY VERY common task - in fact, it's almost the only way to do it. How else are you supposed to guess the colors (or what pixel is *supposed* to be there) without knowing what's around it. It's obvious that the highest correlation will be between the nearest neighbors (except on some edges).
Maybe next time we can make a program that just guesses the colors and look at how interesting those come out!
Thank goodness.
I thought I was going to have to wait until they started dimensionifying old non-holographic movies until I once again heard the keen of good films having their asses wrung with sandpaper.
Are there any non-evil applications or this artform?
I am from a small, grease-loving country in the north called Ca-na-da.
This was in SIGGRAPH 2004, as were plenty of other great papers. Are we waiting another 6 months to see those on Slashdot?
Our method is based on a simple premise: neighboring pixels in space-time that have similar intensities should have similar colors.
It's a fancy schmancy flood fill.
Have done colorization of house and home interiors for years. Try this algorithm on the siding of an all white house and it'll get confused real quick with window and door trim especially in spots where there are no shadows due to angle of sun. Good 75% solution.
I don't know how much of data is on there, but it took my firefox memory usage from 65 megs to 200 megs!
printf($randomline(sigs.txt) \n "-- "$randomline(authors.txt));
-- myself
Sounds like they just print out the image and reach for the crayon box.
I'm not good in groups. It's difficult to work in a group when you're omnipotent. - Q
It's "I'm colorblind, you insensitive clod!"
-Don
Take a look and feel free: http://www.PieMenu.com
The site is slashdotted so I cannot read it, but i wonder if something akin to this could be used for compressing motion video. For example the intensity is encoded with currrent techniques, but instead of the color being encoded at a lower resolution, instead only a very small amount of colored points are encoded. Then during the decoding, the decoder uses an error function, intensity, and the time domain of previous and future frames to 'fill' the colors out.
Take a grayscale image, and you already have the brightness and value for each pixel. All you do is add the hue component based on the color "scribbled" by the user. Stop filling with that color when you hit something that marks a definitive boundary.
-Don
Take a look and feel free: http://www.PieMenu.com
You b/w film purists. If all you can see is a threat to your bizarre, luddite idea of what film should be, you need to get your heads checked, or at least you need to listen to your inner geek. Stop using these folks' achievement as an opportunity for chest-thumping.
The idea that one could color correct video with a few strokes from mspaint is staggering. Imagine if one could do this to color video, in real time... you could color-highlight an object and the computer could follow it without sensors or other pre-implanted devices, and that's not even a particularly original idea. This is awesome technology with applications probably well beyond what we see here.
vk.
And I'm guessing neither of the three of us know all that much about the details of video compression. The "Motion Picture Experts Group" or whoever are probably smart people. They already compress video by only noting changes between frames & stuff like that. I'm going to go out on a limb and suggest that their methods may already sortof make use of this technique.
There are no trails. There are no trees out here.
My Emphasis added:
Flood fill would be described as:
See the differences? They are important.
There are no tiger attacks in my area and it's all because this rock I'm holding keeps the tigers away.
Its not that simple. The problem is how do you define a "definite boundary"? And even if you do what you suggested, you are likely to get rather flat shaded images. What these folks do is really solve an optimization problem and try to make use of the fact that neighboring pixels of similar intensity would have similar color. It is kind of like a fancy flood-fill algorithm, but applied to a new area, and thats what makes this novel!
Don't Panic
If you load up page and watch the very last video, can see a slight artifact in the reflection of the hanging stuffed animal. There's little spots of orange color, like you would get out of the AirCan tool in MSPaint.
I have a feeling this could be corrected with another scribble or two. Really a stunning piece of work. Very cool.
-dave
http://millionnumbers.com/ - own the number of your dreams
all the images on that site are bitmaps. slashdot is now sending thousands of users to download them. that's a lot of bandwidth. ha ha, the site is now down. only wish i'd been downloading an image at the time and seen it stop half way. only then have you seen the power.
OMG someone has to tell these people about the joys of the JPEG image format. A whole page of bitmaps?!?! Why don't they just take a bat to the webserver... same effect but more fun.
"The technique works on the premise that 'neighboring pixels in *space-time* that have similar intensities should have similar colors,' and also allows colorization of videos by 'marking' about one in ten frames."
So this obeys the laws of general and special relativity? What about quantum chromodynamics while we're talking about color?
I dabble more in the area of 3d art, but at times I've lifted a pencil and come up with some decent b&w sketches. Pencil shading is easy but sometimes getting the colours just right is more difficult than you might think.
When I looked at a lot of B&W webcomics, I can see that they'd look better in colour (especially the ones where the artist does occasionally do vibrant colour cells, but usually don't have time). This could change that though... want to see what your character would look like with a gold uniform... scribble a few colours within the borders. Want to see what that night-sky would like like with a red moon and rouged horizon, scribble a little red in and let the program do the work.
For cases where you want to preview colours, or just don't do colour well yourself, this would be a boon indeed!
That feels better now.
Take a look and feel free: http://www.PieMenu.com
This stuff is really cool and has applications outside of just re-coloring black and white.
Industry applications are interesting, but nothing new -- the industry has been using this technology for a long time when it was more labor-intensive, because they can afford to.
The REAL impact of this technology will come when you see it migrate into new versions of iPhoto and Photoshop Elements. In Photoshop, recoloring a part of a photo is relatively easy, but it still involves a mildly complicated process of selecting the color range, specifying the hue you want to shift it to, checking and re-checking.
This, however, is something Apple could roll into iPhoto with relative ease. No more selecting pixels, ranges, or hues -- just select a shade from the color picker, scribble it over the area desired, and hit "apply". Don't like it? Undo and try again.
The industry will use this because it's faster, but I know professionals will still need and want tools to fine-tune their adjustments. Consumers will use this exclusively.
This is a really cool thing to see in person. At NAB last year the Discreet people had this as a feature coming from there high end Smoke systems. Since all Discreet programs require a Tablet the pen scribble thing is used for color correction by scribbling around the area you want changed. Masks are done by making a rough trace around the area and it figures out what your focusing on. Its a time saver but you pay hundreds of thousands of dollars for their systems. After a couple of years other programs like Shake and After Effects start to pick up the plugins. Most major tech leaps in video happen at Discreet first then trickle down.
I tried out their matlab code and put a few example colourings on my web page, for the interested:
http://www.inference.phy.cam.ac.uk/cjb/
Intensity actually takes most of the bandwidth of an MPEG stream, because human eyesight tends to notice changes in luminance more than changes in chroma. The chroma channels are compressed *extremely* heavily compared to the luma channel, and are actually even at a lower resolution.
Make me a friend and I'll mod you up
I could see this working best as a "realtime" colour filter, especially if you're using a pen or something similar. Scratch near a border and view the result... if it goes a bit beyond where you want scratch on the other side of the border. If it's not quite enough lengthen your scratch.
I wonder how much CPU power is required, if you could do this realtime or close to it would be quite awesome, but having to make your scratches and click "apply filter" then wait for 30 seconds would not be nearly as useful/efficient.
When can I download the Photoshop plugin?
Not since Marie-Antoinette played milkmaid has looking simple and honest been so fake and complicated.
My first thought was "Gimp Plugin!", but a Photoshop plugin would be good too.
Stop filling with that color when you hit something that marks a definitive boundary.
That's the "and then a miracle occurs" step.
What a brilliant plan! What if we were to take groupings of four pixels, store luminance for each and an average of their red and blue weights, netting a savings of:
... this sounds an aweful lot like the YUV encoding used in MPEG compression ... probably has something to do with it actually being the YUV encoding used in MPEG compression.
uncompressed: 24 bits per pixel X 4 pixels = 72 bits
compressed: 8 luminance bits X 4 pixels + 8 Red bits + 8 V bits = 48 bits
100 - (48 / 72 * 100) = 33.3%!
Wait
"Because Science" is one step from "Because old book". Try "Because of my experiment testing my falsifiable assertion".
For example, on Farscape, given Virginia Hey's problems with makeup and contact lenses... heck, any of these humanoids-with-funny-skin-color shows would benefit from not having to put in the hours upon hours of makeup. Instead, we'd see hours upon hours of post-production...
-- Still waiting for the Nike endorsement
In the days when colorized videos of black-and-white films were common, I watched a few. The so-called "colorization" had some very serious problems, and I wonder whether this new method addresses them.
The problems tended to be in the background, and they probably thought people's attention would stay on the foreground, but I think like many things in film you notice them subconsciously. Either the background is out of focus, in which case there are no sharp edges for the colorization to work on, or it contains a basically infinite quantity of detail as the background gets farther and farther way. Either way, it was extremely common to see uncolored areas in the background.
It was fairly common to see black-and-white paintings hanging on walls, for example. The walls would be some fairly uniform wash of plausible wall color, but nobody was going to take the time to handcolor the paintings hanging on them.
A similar problem concerned scenes with machinery in them, or anything with lots of complex, detailed motion (so that successive frames weren't similar). Thus, you'd see black-and-white printing presses operating in a colorized newspaper building...
In addition, the fact that the colorized faces, for example, were a uniformly colored wash, rather than varying in color as well as brightness, created a subtle kind of phoniness. To me, the result was the conveyance of a sort of emotional coldness. The colorized movies looked colored, but they didn't feel colored.
The exact opposite of the kind of lift you couldn't help feeling in the fifties when you saw a Technicolor spectacular--in the days when "Technicolor" meant that by golly you were watching genuine dye-imbibation prints from real color separations. Sweet as candy, but irresistable. (The effect does come through in the best DVD restorations).
"How to Do Nothing," kids activities, back in print!
does a new development need to be f**king perfect? sure its got a few problems, and any program has limitations. just because this thing cant take absolutly any image and make it color perfectly doesnt mean its bad
personally I like the recoloring shots, I want to screw with some peoples' heads with that hehe
By reading this, you have given me brief control of your mind.
-Don
Take a look and feel free: http://www.PieMenu.com
Also, does anybody happen to know if a similar technique was used for the new Tom Cruise movie coming out? You know, the one that kinda looks like Waking Life in style?
Buy Steampunk Clothing Online!
Gosh, sorry if composing and lighting arcana are not well known outside pussy film geek wanking circles. Pop a fraking Prozac, dude, and change your panties. What a useless little prick you are!
could come in handy for re-drawing maps in the area. Save new layer, don't finalize changes.
Hey, that is actually true!
What's this? a slashdotter who knows what he's talking about?
He's a witch!
It's easier to fight for one's principles than to live up to them.
The one bad thing about coral-ized links is that, due to the Coral servers usually NOT being on port 80, if you are behind a restrictive corporate firewall that does not allow traffic on J. Random Port, you cannot use them.
It Would Be Nice If Coral had servers on port 80....
www.eFax.com are spammers
It could theoretically reduce the bandwidth, except that the CPU cost to apply the data and reconstruct the image would be prohibitive with anything we have now-a-days.
They are NOT applying these scribbles in real-time - they scribble, the computer **grinds**, and quite some time later an image appears.
With the kind of CPU power it would take to apply this sort of data in real time, you could probably get much greater compression using a very exhaustive wavelet search.
www.eFax.com are spammers
Don't quit your day job.
Let us just hope someone will apply this to clerks and release it in color!
Answer: Yes
" The technique works on the premise that 'neighboring pixels in space-time that have similar intensities should have similar colors"
Interestingly, the retina exploits that same property of natural scenes to compress images. This correlation between luminance and color is an opportunity to throw out redundant information. The eye multiplexes color and luminance information over a single channel, transmitting luminance while discarding color at high spatial frequencies and transmitting color while discarding luminance at low spacial frequencies. First reported by C.R. Ingling, color/luminance multiplexing is an inherent property of the linear color-opponent center-surround receptive field. For a good explication of the subject, see:
Vision Res. 1985;25(1):33-8."The spatiotemporal properties of the r-g X-cell channel."
Ingling CR Jr, Martinez-Uriegas E.
Abstract: Analysis of the simple-opponent r-g receptive field of the X-channel shows that it is tuned to both high and low temporal frequencies, high and low spatial frequencies, and that its spectral sensitivity is both chromatic and achromatic.
Ceci n'est pas une signature.
that is quite possibly the coolest thing i've seen in 2005
WTPOUAWYHTTOTWPA
What's the point of using acronyms when you have to type out the whole phrase anyways?
So that's what happend to the gamers that were booted from the Israeli Army (http://slashdot.org/article.pl?sid=05/03/09/16222 15&tid=133)
come on fhqwhgads
Take a few deep breaths, you'll get over it. In the meantime, I can spread some disinformation if you want. Hmmm...
The MPEG standard, knowing that porn tends to drive the industry, actually contains several optimizations for drawing things such as skin tones and nipples. Really.
Make me a friend and I'll mod you up
Really, did you just decide to type stupid shit while waiting for the page to load or what?
In Soviet Russia, pictures colorize YOU!
that was stupid.
Except that groupings of 4 pixels is not necessarily a good approach. For one thing, it may be far more color information than is really necessary. Besides that, two adjacent pixels may have significantly different colors. Carefully designed "scribbles" of color could very well take up less space and give better quality.
I'm sure there are good reasons for the JPEG/MPEG method, and I'd be a bit surprised if the groups in question didn't think of this possibility, but I still think it should theoretically give better results (at the cost of higher complexity and computational requirements).
Digital cameras are prevented from capturing finer details due to:
1. Digital artifacts
2. Digital noise for ISO film settings
3. Inability to record an image without any processing except the optical effects of lens, appature size, etc. Raw images address some of this but not all.
4. Dynamic range limitation due to 24 bits per pixel
This is why your 3 megapixel camera is terrible for recording 8.5 x 11 inch pages even though this works out to over 175dpi.
cant photoshop fill do this kind of thing already? where if you fill in a certain area, it fills in similar areas that are shaded with the correct shade of, let's say, yellow. I think gimp can do that as well.
(sarcasm)
We can recolorize all the black actors in all the movies! Will Smith was always a tad too swarthy to for me...
Alas, we've reached Mr. Jackson a tad too late.
(/sarcasm)
Please stop stalking me, bro.
-Don
Take a look and feel free: http://www.PieMenu.com
This is old technology.
Don't you guys remember the "Take on me" video by A-Ha.
Same stuff...
Some folks seem to be excited (or angry) with the possibility of coloring B&W movies with this technique. Forget realistic coloring, this looks amazing for artistic recoloring.
Go take a look at the "recoloring examples" in the coral cache. Also look at what a slashdotter did with the code. Photographers, designers and painters could do neat things with a filter like this in Gimp...
Prescriptive grammar:linguistics
You really have to think about changing your dentist. I mean like change it today.
The first example picture is a bit strange. In the result picture the boy has green eyes, but there are no green markings on the eyes on the BW picture?
Funny thing is i was arguing with a photoshop wonk about this just last week. It seems that when it first surfaced, it was 'debunked' by end-users on worth1000.com.
Levin et al were published by ACM more than a year ago.
Some criticism: Their website only shows images that show the strength of their method. They don't show any examples of it's weaknesses, and it does have some.
A much more interesting website would be this one:
http://mountains.ece.umn.edu/~liron/colorization/
A similar technique, a comparison with the levin et al method, images that show how both techniques sometimes don't work so well, and a nifty java applet so you can try it yourself.
Both of these techniques, however, seem to depend on a good quality grayscale image. This is paradoxical because the emphasis in colorization has been old film. A lot of that stuff is coarser, and higher contrast, than any of these images. Silver oxide without any dyes added has a big spike in it's color sensitivity for blue light, as well, so what tone you do have to work with is maybe not going to be correct.
This is just like television, only you can see much further.
None of the caches will let me have it and I was to try to use it.
So put that in your pipe and grep it
...even the coralized mirror got slashdotted!
Slashdot needs Geekcode | Can anyone recommend any good SCIFI? My tastes: Foundation, Startide Rising, CITY, Ringworld,
here. is the link to matlab sources. /.ed site it isn't much good
---
sorry about the dupe, but i realized that w/o the link to a mirror a
-- Avishalom is usually vish
Unfortunately a probably life cycle of this kind of tech will be:
1) use in a carefully restored, prestigious film
3) good reviews and ratings/audience
2) creation of a mac based product
4) profit
but... oh yes
5) a slew of third rate knock offs that use the effect to justify a 5th rate plot
I'm painting the upstairs of our house and can't get inspired for anything.
Being younger we went with "different" color schemes in our house, orange, grey, shit brown (all really fits too) but the upstairs is still *blah* and I can't imagine any color there than what already is. I've got paint chips taped up here and there and something like this would be nice to have at home (far off I know).
Taking a picture then scribbling on it would be a nice way of previewing paints. In fact I suggest here (in the open instead of trying to sell the idea) that designers get tablet PCs with a camera built in... then just go to a clients house snap a shot and scribble paint colors in.
Get your Unix fortune now!
The trick about this is getting the colors right. Sure they make it look easy - probably because all of those examples were color at one point, and they just eyedropped the correct colors.
Unless you use the exact colors necessary, it will turn out looking wrong or "not quite right." I think this plugin will still require the skill of an artist to pull it off.
Color is a very complex thing.
Apparently, one of the critical differences between Octave and Matlab is that Octave apparently doesn't support multidimensional matrices/arrays with dimensions greater than 2.
:(
Which means that the current code is completely incompatible with Octave, as it depends on Matlab's implementation of imread() which returns image data as a three-dimensional matrix.
Going to see if I can get it to work easily, but there's a good chance I won't be able to.
retrorocket.o not found, launch anyway?
Now he foist more old colorized movies on us where everyone's skin is orange and the sky is purple and all small objects are grey.
You are in a maze of twisty little passages, all alike.
24 bits per pixel X 4 pixels = 72 bits
In other news, mathematicians still agree that 24 times 4 is 96.
YUV 4:2:0 saves 50% bits over YUV 4:4:4, more info on wikipedia (per usual) http://en.wikipedia.org/wiki/Chroma_subsampling
Actually, this could be applied to NTSC (or PAL, SECAM, or other form of standard definition video). Present video scalers do an OK job of up-converting SD video, but the poor color quality is usually only magnified by the process. Faroujda and Runco have proprietary algorithms that help a little and sell in boxes costing over $10,000.
Could something like this be used to sort of take the existing color information as the scribble hints and apply much richer and less noisy color to the up-converted image? Could this even just be used as a chorma noise filter? I see great processing possibilities. Somebody write a dScaler filter and try it!
I'd rather see te money being pissed away on this technology be devoted to resurrecting and preserving old silent B&W films from the teens and 20s.
RS
Shoes for Industry. Shoes for the Dead.
There were 4 or 5 of these "automatic segmentation via hints" at SIGGRAPH last year. This is are useful montages as well as painting.
This is exactly typical of the Zionist agenda to take over the world, starting with "colorizing" all of our black and WHITE images.
We must not take the Jewish threat to Space-Time lying down! I want my colorized images orangish and filmy, the way they're supposed to be!
Great, all we need is Ted Turner to have an easier time colorizing old films and not releasing the originals.
they extended the paint bucket to the 4th dimension?
Please notice that subrama6 said to the 4th dimension, not to "a 4th dimension" or to "4 dimensions".
Time is refered to as the 4th dimension often enough that even though there is arguably no z axis, the statement is still reasonable in common usage.
Nice.
Let's not forget the competition, AVI. As we know, it is just a container file, containing a highly compressed jpeg for every frame, and only allows 11500 khz 8 bit uncompressed audio.
It's easier to fight for one's principles than to live up to them.
I think this would be a pretty neat application to use with the existing Mars and Spirit rovers pictures sent back. -j
Can an algorithm allow for that? Or is it just blocking out a whole area with the same hue and saturation, and just keeping the original brightness variation? If so, I doubt it can look at all natural.
Ceterum censeo subscriptionem esse delendam.
ill take a look too see how it stacks up.
ill run some tests of my own and send the posting back to slashdot
Check journal for info on Anti-TextBook, an idea by me.
I suppose this would be reasonable for choosing between "red" and "blue", but there is no way this'll be able to accurately help anyone choose between "ivory" and "pearl" anytime soon.
Maybe this is why people remember colors in different ways (from what they actually were) when they look at them years later. Or why some people seem to think we dream in B&W...
From a historical perspective I do think it's a shame if the original versions are lost or suppressed, but otherwise the result of colorization or anything else deserves to be judged on its own merits. Not whether it might offend dead people.
Oh, sure, your let's judge everything on its own merits seems reasonable on the surface, but you'll be singing a different tune when the offended dead directories with their zombie grips, rotting boom operators, ghoulish foleys, and other undead minions, rise from their graves to take back their artistic vision. You think you're rational approach will save you when the popcorn is buttered with blood? No! You will cower on the sticky floors as death stalks the aisles.
Oh. You will rue this day, when the undead directors bring forth their artistic vision of blood!
Look at the pics... They were taken in color, converted to B&W, and then scribbled on... There's NO way that much detail can come from a "scrible" that is huge, but very fine color detailing (ie, the sign on the hotel... the face paint on the kid...) with one BROAD stroke?
I cry BS on this story and paper! Lets see the code... release to GNU or something or like someone else suggested, make it a filter for gimp.
B.S.
Read my post here
Quote (Regarding "The CSI Phenomenon"):
- shazow
Colorizing film by this method is not much different than what some biomedical and computer science research is very interested in: image segmentation.
Similar to how the video method tracks changes in object size and shape over some temporal resolution, image segmentation tracks shape, size, etc in some spacial resolution.
Check out Insight Segmentation and Registration Toolkit (ITK):http://www.itk.org/
a more obvious application for this tech would be to modify this for rotoscoping...
:)
Assuming it can get clean lines in the colourisation, It should be able to give you a clean colour map you could then use to quickly and easily extract a single character from footage...
Just a thought there if you are reading this
'Plex
Rich Gentlemen Hide - The Existential Comic
Glad all that aid is coming in handy.
This technique might be seriously useful for digital cameras. Instead of a sensor with alternating RGB pixels, make one that is black and white. The B&W image will be somewhat sharper than a colour image made by interpolation (since you are getting the true brightness value at each pixel, not a guess based on that pixel and its neighbours), and the sensor will need much less light (instead of masking off all but red/green/blue light at each pixel, it gets the whole visible spectrum). It might also be cheaper to make.
Then add a really cheap and low-res colour image sensor and use that to colorize the B&W image. Or you could have a single sensor where most of the pixels detect white light, with occasional RGB clusters dotted around to get colours, and the 'scribble' algorithm used instead of conventional colour interpolation. Again, such a sensor could operate with less light than one where every pixel is R/G/B masked, so the camera could be built with cheaper optics (narrower lens aperture, etc). Although the colours (chrominance) in the final image would be less accurate than with a conventional sensor, the brightness (luminance) would be more accurate, and the eye is more sensitive to luminance than chrominance.
Final thought: could you use this for lossy image compression by storing the B&W image plus a few dots and scribbles to reconstruct the original? If you have plenty of time for compression you could keep trying or evolving different scribbles to get one giving acceptable results.
-- Ed Avis ed@membled.com
Forget movies; the big application for this is old war footage. When you see footage of Nazi Germany in color, suddenly it goes from distant black-and-white storytelling to something as real as the air you breathe. Same thing with footage of the concentration camps. And what about photos of the Civil War?
The color helps keep that history genuine to our eyes and minds and relevant to the problems we face today and everyday.