Content Aware Typography, a tumblr blog in which image with typography distorted or generated thought Photoshop’s Content-Aware Fill are collected, was recently included in the It’s Nice That Annual 2013.
After this first transition to print, I thought it would be interesting to recap a bit the progression of this little ongoing experiment.
The idea of Content Aware Typography came in the most obvious way: I was actually cleaning up a picture from some typography embedded in it, when an algorithmically produced, unknown, but realistic glyph appeared.
I was already aware of several interesting “misuses” of the content-aware filter such as the Extended series by Mike Ruiz or the fashion shots modified by Enrico Boccioletti.
Extended Bliss by Mike Ruiz (2010)
I tried the filter on some other iconic type-based images to see how the meaning could be degraded. I intuitively felt that the results expressed both the noise of the information overload and the typical repetition of advertisement. That’s why the first images that I chose were kind of obvious: “I shop therefore I am” by Barbara Kruger, frames from They Live by John Carpenter and some of the “Obey”’s by Shepard Fairey .
Afterwards my feelings were somehow confirmed by the description of the project given by Matt Duhamel on Twitter:
«Algorithms, human psychology, consumerism, gestalt, art standards, and memes all rolled into one giant bundle.»
The procedure was then applied to physical typography like the one you find in shopping malls.
I showed those first results to Roberto Picerno and together we tried a different technique consisting in generating the whole space of the picture around text.
At this point I realized how accessible was the procedure and started to consider it the actual core of the project rather than the single visual results so I condensed it in four steps and allowed submissions on the tumblr:
1. Choose a picture with some typography in it;
2. Select an area that is close to the letters;
3. Apply Photoshop’s Content-Aware Fill;
After more than 6 months, almost 600 posts and about 2000 followers, some patterns emerge. For instance, the “Keep Calm and Carry On” is probably the most abused image, Google and Nike logos are frequent, Obama’s HOPE constantly makes its appearance, Ed Ruscha and Lawrence Wiener are recurrent, there are multiple versions of the Hollywood sign.
The following image submitted by Eric Sena is, with the current amount of 1659 notes, the most reblogged one. I particularly like it because both of its physicality and dullness of the text.
I also encountered some other experiments that, maybe coincidentally, adopted a similar procedure in different contexts.
If I got it right, these pictures were produced using InPaint: a software other than Photoshop, that allows a similar treatment.
Steven Leyden Cochrane made a nice zine using the content-aware fill on almost the entirety of its content.
Lastly, another “content-awared” Ruscha was found as header of Amaranth Borsuk’s column on Jacket2.
Finally, I take the opportunity to thank all the people who engaged with this experiment.