Poetry, music, and color
March 5, 2008 at 3:08 pm | In artsy, computers, design, literature, music, programming, school | Leave a CommentTags: computer art, cs171, graphics, music visualization, poetry, processing, visualisation, visualization

[poem visualizer, designed by me]
One in the morning isn’t always the best time to embark on programming projects of indeterminate length and scope. I learned that last night when, after having spent most of the afternoon and evening on a take-home midterm for my visualization class, I was beset with the dilemma of the extra credit problem. “Design your own poetry visualization,” along the lines of Poetry on the Road. This is a series of graphic designs commissioned each year by the Internationales Literaturfestival Bremen, designed by a team of professional graphic artists led by Boris Müller. The two things that immediately struck me about these graphics were 1) their juxtaposition of visual complexity and conceptual simplicity, and 2) their obvious requirement of a vastly greater number of hours — and sheer programming virtuosity — than the 24 hour maximum allotted for my midterm.
I must’ve been struck with temporary amnesia, or just had an acute attack of masochism, because I promptly forgot about the 3 problem sets that I had been saving up my sleep-hours for later in the week, and proceeded to bang away at Processing (coolest programming language ever) for the next 3 hours. The result, humble by the standards of any legitimate computer artist — but hopefully not of my exam grader — is this, a grid of colored squares representing “The Wasteland” by T.S. Eliot. I chose the poem mainly because it was long enough to really show patterns in my algorithm, but not so long as to crash the program. (although I would like to run Paradise Lost or The Iliad through, just for kicks.)
Continue reading Poetry, music, and color…
Context-free
February 26, 2008 at 12:23 am | In artsy, books, computers, design, geeky, internet, school | 2 CommentsTags: algorithms, context free art, cs51, digital art, grammar, randomness, scigen

[via Context Free Art Gallery and Community of Variation]
For a computer science assignment last week I wrote a program with a friend in Scheme that generates sentences using context-free grammar. By specifying a few basic rules for parts of speech and including a simple word list, you can get some remarkably coherent results — coherent, that is, in a strictly grammatical, minimalist sense. There’s actually no attention paid to the meaning of the words used, or to their relationship with each other, a consequence of the grammar being “context-free.” We had a few laughs using our program to generate bogus math proofs, but instead of puzzling you with a slew of inside jokes and insomnia-induced geekiness, I’ll point your way to a much more impressive — not to mention amusing — application of the same algorithm, this time used to generate an entire scientific paper. You can even put your own name down as an author! Here’s an example passage:
We question the need for digital-to-analog converters. It should be noted that we allow DHCP to harness homogeneous epistemologies without the evaluation of evolutionary programming [2], [12], [14]. Contrarily, the lookaside buffer might not be the panacea that end-users expected. However, this method is never considered confusing. Our approach turns the knowledge-base communication sledgehammer into a scalpel.
[Stribling et. al (PDF)]
Not too bad for randomly generated babble, is it? Continue reading Context-free…
80 Million Tiny Images
January 17, 2008 at 3:48 pm | In computers, geeky, programming | Leave a CommentTags: image processing, images, linguistics, mit, nouns

Thanks to C.J. for sending me the following, incredibly cool link: 80 Million Tiny Images. It’s a mosaic of millions of online images corresponding to nouns in the English language, and the spatial arrangement of the images in the mosaic reflects their semantic relationship to each other–i.e. closer images represent words that are closer in meaning. From the page (which also contains a link to the research paper):
Each of the tiles in the mosaic is an arithmetic average of images relating to one of 53,463 nouns. The images for each word were obtained using Google’s Image Search and other engines. A total of 7,527,697 images were used, each tile being the average of 140 images. The average reveals the dominant visual characteristics of each word. For some, the average turns out to be a recognizable image; for others the average is a colored blob. The list of nouns was obtained from Wordnet, a database compiled by lexicographers which records the semantic relationship between words. Using this database, we extract a tree-structured semantic hierarchy which we use to arrange tiles within the poster. We tessellate the poster using the hierarchy so that the proximity of two tiles is given by their semantic distance.
The most interesting thing about the result is the remarkable degree of color agreement that they achieve. Despite the fact that each tile is the average of several photos of the same thing, the end result is often surprisingly recognizable, and close-by tiles tend to have the same color scheme. The overall mosaic, rather than appearing as a wash of meaningless color noise, has some fairly uniform blobs on it because of the semantic association of images. The one they give on the website (composed of 7.5 million images, it seems–not 8 million) almost looks like a hunched over figure of a person. I wonder what the full 80 million images put together looks like.
The real question is, when can I get this as a wall poster to put up in my room?
Two more vectors
December 9, 2007 at 5:02 pm | In artsy, computers, design, silly nonsense | 1 CommentTags: adobe illustrator, vector art
As you can see, I haven’t gotten a lot of work done this week. These are the first two of 5 vector portraits I’m doing of my roommates (plus myself). They’ll all be integrated into some kind of poster/decoration eventually…


Some of the line drawing is a little dubious, especially the parts on the second image where I had to make up areas that were cut off in the reference photograph. This is probably where it helps to be good at drawing in real life.
What I missed out on as an aesthetically challenged 14 year old
November 28, 2007 at 9:13 pm | In artsy, computers, rambling | 1 CommentTags: digital art, vector art
I made a few vectors the other day. Clearly I don’t have enough work to do. Convinced I was able to create “art” after reading a few tutorials about “vector art,” I proceeded to learn Illustrator and try my hand at it. The results are pretty, but you’ll notice I prefer calling them simply “vectors.”
The idea behind these vector images, or vexels they’re called if you make them in a raster program like photoshop, is to take a photograph or some other “reference” image, trace the different blocks of color on it either with or without image-processing tricks like posterize to help you, and use layer after layer of these flat shapes to build up a cartoon-like image. You might have been struck by the originality of the animation in A Scanner Darkly, which used a vector drawing technique to convert live-action film into graphic-novel-like frames.
After spending a bunch of hours doing these, though, I realized that most of the vector art people make — my images below included — is bullshit. Tracing shapes on a photo with a computer is probably the only thing easier and less artistic than tracing things on a photo in real life (okay, so it can be argued that operating Illustrator takes more motor skills — not to mention geek points — than holding down tracing paper). Most of the good “digital art” that can be made by Illustrator really isn’t of the posterize-trace-eyedropper tool type. Even if a reference photograph is used, a decent artist usually seems to render the result with a highly stylized digital vocabulary. Since correct proportions and likeness are more or less a given, the real challenge, just like in non-digital art, is to add something to the image.
Even with a healthy dose of aesthetic insight, though, photo-realistic vector art is still unsatisfying to look at. The typical style among skilled vector artists seems to involve anal-retentive layer organizational skills, plus clever manipulations of transparency and color, which achieves an admittedly dazzling multilayered appearance. But these images lack dynamism or impact. It’s almost as if the pictorial “correctness” of digital art is a little stifling. That’s why for the time being, I’m going back to my sketchbook. The paper one. Enjoy these for what they’re worth.
My roommate listening to his pants on the headphones. I’m not sure what he hears. I got a little bored after putting in a few layers, so his shirt and arms look a little unfinished. His face also has a slightly weird contour. And the bottom where the reference photo cuts off is a little too obvious in the final product. Was a little too lazy to give a “brushstroke” look to the pants, but after I got the point all the contour tracing got a little boring. Was inspired to go and do some drawing though. Using a pen is much more satisfying than a mouse, although getting the proportions right is definitely more of a challenge in real life.
Sunset over the Maine coast. This was a nice ref photo, and I played a bit with the rendering of the reflection as well as the colors. It was also remarkable how realistic I could get the sky to look with some careful color choices (didn’t go directly off the original photo with eyedropper tool), although not as realistic as some of these guys can go in Illustrator.
Reference photos, so you can see how much of a shameless tracer I am:
Blog at WordPress.com. | Theme: Pool by Borja Fernandez.
Entries and comments feeds.


