Image science radness o' the day

“This is your Healing Brush.
“This is your Content-Aware Scaling.

“*This* is your Healing Brush & Content-Aware Scaling on (really good) drugs…”

Adobe researchers Eli Shechtman & Dan Goldman, working together with Prof. Adam Finkelstein from Princeton & PhD student Connelly Barnes, have introduced PatchMatch, “A Randomized Correspondence Algorithm for Structural Image Editing.”

No, I wouldn’t know a randomized correspondence algorithm for structural image editing if it bit me on the butt, either, but just check out the very cool video demo. More details are in the paper (one of the 17 papers featuring Adobe collaboration presented at SIGGRAPH this year).

So, what do you think? [Via]

13 thoughts on “Image science radness o' the day

  1. These tools are going to be included with CS5, right? Right? I mean, it’s just a few lines of code, maybe a couple of meetings, a few hours of work and… voila!
    [“It’s all just code,” the engineers always tell me. πŸ˜‰ –J.]

  2. OMG. Will this be in the next version of Photoshop?
    [Well, as usual I can’t talk about unannounced products, and turning research code into practical features can be a challenge unto itself. Broadly speaking, though, the point of Adobe’s research efforts is to build technology that can be brought to market in really appealing ways. (Hope that doesn’t sound too corporate spin-y.) –J.]

  3. All I can say is WOW! The things you could do with that kind of tool — thinking about building and architecture and showing possible design ideas to a client. Amazing!

  4. Eli and Dan are doing incredible research in this area and I am continually amazed by their work. But I’m equally fascinated by the weird side-effects of content aware scale. I’ll admit i haven’t done anything useful with this blog yet but thought readers would enjoy it none-the-less. All meant in good fun.

  5. Awesome. What happens when you’re the grad student team that discovers this?
    [Connelly has interned twice at Adobe. Internships offer a nice way to collaborate with researchers for a particular amount of time while supporting their efforts. –J.]
    Does Adobe sign such people immediately, or just give them a nice fat check to license the technology? Reminds me of the quote that says technology, at a certain level of sophistication, is indistinguishable from magic.
    [Indeed. –J.]

  6. Roll on CS5.
    I was surprised how quickly you managed to integrate seam carving into CS4 and I’m hoping it’s just as “easy” to incorporate this into CS5.
    [The definition of “this” is quite broad, involving at least three or four discreet technologies/innovations. The hole-filling code is one thing, the reshuffling code is another, etc. Some pieces could be easier to fit into Photoshop’s existing tools/metaphors than others. (No promises with any of this, of course. I just want to be sure to manage expectations correctly. –J.]

  7. Jack,
    I think I know how to lose weight now…cool stuff…
    If we can bottle that “stuff”, we can make a lot of money
    Ken in KY

  8. The path from great ideas to a sellable product is fraught with expensive potholes. I really hope this SW makes it out of the think tank, maybe they should generate a Beta version so that those of us willing to put forth some disposable income could try it out and offer complaints(suggestions) since all of the good to great improvements come from a diverse group of pickey users.

  9. Question: will they be making the research proto-type available for us to play with like the Seam-Carving one was? i.e. I can download it, play with it, and tell everyone it has to be voodoo, ’cause nobody’s that dang smart. πŸ™‚

Leave a Reply to Ken Cancel reply

Your email address will not be published. Required fields are marked *