Less creepy?
Original
Lol, just realized, it started out as a photo, got encoded into audio with a reverse-FFT, then pulled back out of the music with an FFT, then re-drawn using linify.
That image has been to hell and back.
Less creepy?
Original
Lol, just realized, it started out as a photo, got encoded into audio with a reverse-FFT, then pulled back out of the music with an FFT, then re-drawn using linify.
That image has been to hell and back.
I genuinely thought you were flipping the bird really close to the camera in the linified one.
Hehehe! Well, no. but some things really aren’t easily identifiable, and others are a bit cacked up once the picture goes through the process. The glasses and straws aren’t easily identifiable in this Linified pic, and the model’s left hand is a bit cacked up, although her face is still relatively well-handled. In my later post, the model actually has, as you can see from the original pic, quite a good nose.
I picked these pictures to put through Linify specifically to stress the algorithm.
The lady in the top picture should smile more.
For me the coolest ones are definitely those where the lines add something to the image. Those generally seem to be the simpler image with lower line counts, like these from their own gallery:
and also @renke’s banana at top and @crashproof’s Death.
The super-high-linecount images just seem more like an Adobe “scratchy texture” filter, to me.
This topic was automatically closed after 5 days. New replies are no longer allowed.