Ju Young Park – Generative Art

by ju @ 7:40 pm 29 February 2012

 

Inspiration 

For generative art project, I was heavily inspired by Anthony Mattox’s generative ribbon art.

 

My purpose of this project is to create something flows continuously, algorithmically, and randomly. As the pattern flows as time goes, I want it to become void. With this in mind, I designed a generative art that re-displays an image in a ribbon-like or smoke-like pattern in a way an image becomes void. So, I tried to implement an algorithm that breaks an image into smaller pixels and reads the pixels to generate unique smoke-like pattern.

 

Algorithms

I hacked Daniel Shiffman’s Explode() function to break an image into particles. Then, I stored each pixel of each particle, and drew random curves based on the pixel using infinite loop.

 

Screenshot

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Final Outcome

 

[youtube=http://www.youtube.com/watch?v=iqOQc17XX9g]

 

 

Experiments/Examples

 

[youtube=http://www.youtube.com/watch?v=Jp8lIySeCsg]

 

1 Comment

  1. really like seeing the processs of the image being split into particles. that part alone could definitely be taken in a new direction for a sequel/related project!
    Agree!+2!
    The growth of the ribbon pattern is mesmerizing, I like how it looks smooth and elegant all the way though.
    I like your choice of objects to run through your algorrithm.
    It is not directly clear to me (visually) how the image informs the simulation. While I understood the oral explanation, it is a bit hard to understand from the actual artifact on the screen. Maybe it has to do with the projector problem.
    ^ I agree, it’d be nice to see more of a connection from the image to the ribbons, other than just the colors.
    I feel like it could be interesting if there was more of a connection between what the image is and how that content effects the movement of the particles (ah, someone just said that, woops). Also, curious if you tried it on a white (or lighter) background and how that looked?
    Nice aesthitics, but still isn’t very informative to me. But does it need to be?
    ** Yeah, i think if it’s cool / visually interesting that’s enough! It doesn’t need to be informative.
    Just beautiful

    Are the particles completely random and only the color data from the pictures? In any case, it looks nice, and I’m sure that was your intention, but I’m a little mystified to why generate the ribbons through pictures instead of randomizing the colors as well. Cool nonetheless!

    Not entirely sure where your code comes in, as opposed to Shiffan’s example
    Sad that we’re not in the dark / have twenty minutes to stare at thsese. Seems like they’d be really relaxing!

    So er, is this just a recording of Dan Shiffman’s “Explode” example, with “trails” on? (in other words, you just turned off the the background function)?
    Also, Check out the works that Jean-Pierre Hébert was doing in the 1970s and 1980s, http://jeanpierrehebert.com/
    The question I have is what this work “means”, or more accurately, how it “operates” in culture. In what way might others find this work interesting (in 2012)?

    I’m not clear on how the particles turn into the smoke forms. Having not seen Shiffman’s example, I thought the particle explosion was cool, but then how did you go from the particles to the final form?
    Is there any simulation of actual smoke motion (fluid simulation)? Or is all the motion random?

    noooooooooooooooooooooooiiiiiiiiiiiiiise
    It is difficult to see any sort of connection between the aprticle behavior and your starting image.

    You can use processing’s MovieMaker object to save frames to a video file, that way your framerate won’t be an issue. Using a white background would have helped visibility. (agreed! especially on this projector)

    i think it would be ool to see the patterns move instead of beign static: i thik hats what i like the most about ribbon art (or at least what i remember from traditional dances with ribbons). it would also be cool to have that be linked to the image or even a video somehow. i think if you were able to take a landscape video feed and do this on that with movement based on noise from the scene, i think that would be beautiful.

    I am wondering how long this smoke patterns take to form- does it keep going on, or does it stop at a certain point?
    maybe the image slection could be interactive by the user- like allowing them to upload specific photos, or use a google search to create images from a keyword

    How are you generating the smoke patterns? Are they following trajectories in a chaotic map? The images are very attractive, but I’m not entirely clear on what the process was to alter the images into those forms—maybe I just missed the explanation.
    **Yeah I’m not clear on how it is working either.

    maybe the background should be a different color, it’s really hard to see the colors in the image against black even with better resolution
    +1 its been happening alot to people lately

    I still don’t understand how the algorithm works for this…
    I would also like to know more about the algorithm and about how the pixels are initially stored and then accessed.

    I want to hear more about why you chose smoke patterns. I assume its because they are pretty. But I remember in your meeting you talked about trying to link the image’s distortion to the final form of the flow. In a design context this might be enough, but as your an art major it begs for more. I was most interested by that initial second where the pixels seem to zoom at the camera and you could still vagually see what the origin was.

    It looks like the results are interesting. Unfortunately its very hard to see on the projector. I will have to check out the videos on the blog.

    probably want to spend some time optimizing the visual output so it is as clear as the first example you present (i.e. using a full range of values–the histogram should be a wider bell curve spanning the entire range)

    Comment by admin — 29 March 2012 @ 6:45 am

This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License.
(c) 2024 Interactive Art and Computational Design, Spring 2012 | powered by WordPress with Barecity