Facebook Plunges Again
Tumblr 101: Tips And Tricks To Get Started
Computer solves 400-piece puzzle in 3 minutes; scary implications for photo manipulation
His recent work -- three research papers -- is all about content-aware manipulation of photos. I'm struggling to pick one because they're all awesome, so I'll just give you the highlights:
- A probabilistic jigsaw puzzle solver -- this is the technology featured in the New Scientist article, so there's lots of dumbed-down details if you don't want to read the paper itself. In essence, it does exactly what a human does: matches edges, but it does it quickly and very accurately. Similar technology could be used in photo manipulation (and may indeed already be used by Adobe's Content-Aware Fill) -- the biggest give-away when you manipulate images are edges. This technology could magic away those edges!
- A content-aware image prior -- this is a funky way of saying 'image restoration', and I wouldn't be surprised if this is a sneak-peek at the technology you'll see in Photoshop CS6! Look at the sample photo -- the results speak for themselves.
- Motion blur removal with orthogonal parabolic exposures -- (phew, just typing that gave me a bit of a hard-on) -- in layman's terms, this is blur removal by taking two photos from slightly different viewpoints and then... performing some magic. Again, look at the sample images for some fantastic proof. I wouldn't expect to see moving lenses in still cameras any time soon though...
Comments
9
Subscribe to commentsnanomatrixMay 24th 2010 10:27AM
This is actually a bad thing for security aware people and their really expensive shredders. Obviously its not something that would be used in the near future, but I envision someone being able to someday scan in a whole trash can of shredded papers and let the machine put it all together for them...
Sebastian AnthonyMay 24th 2010 10:36AM
Hehe, good point!
I think it's more likely useful for images with colour data though -- but matching text is probably easier... just a different technology.
(I think shredders use some pretty fancy techniques nowadays.)
EvenioMay 24th 2010 12:11PM
That's only the case if you get a pretty fancy (i.e. expensive) shredder.
AlasdairMay 24th 2010 1:28PM
My shredder feeds directly into a furnace - will I be at risk?
Sebastian AnthonyMay 24th 2010 1:30PM
Depends how hot the furnace is, man.
annoyingposter3May 24th 2010 5:19PM
sebastian you are one sick son of a bitch creeping on furnaces
Sebastian AnthonyMay 24th 2010 5:21PM
Those shapely coals... mmm.
Stuart HallidayMay 25th 2010 5:18AM
You don't need to have 'moving lenses' just moving sensors and we already have these in the form of image stabilisers. So accidental blur could be a thing of the past.
The camera just needs to take two images very quickly (already possible) and use the two images now stored in a buffer to write out to your storage card a blur free image.
Nice to see the role of pure mathematics being used in such a popular media.
darwinsurvivorMay 24th 2010 7:38PM
Does anyone else feel that 3 minutes is pretty high? You would think comparing the pixels on the outer edge of each piece would allow for a pretty accurate pairing. The color of the pixels on either side of a "cut" should be pretty darn close.
To compare each side to each side of each other piece is only (400*300*4*3) 1915200 connections.
In 3 minutes, that would be 0.000093985 seconds (0.093985 miliseconds) per try. That's not even counting not having to compare sides that have already had a match found.