Frederik Vanhoutte describes himself as a creative coder who works in the field of generative art. His site W:BLUT has lots of cool little experiments. Above, Big Red I, a longer fractal experiment that evokes FRank Lloyd Wright. Read the rest
Damien Henry, co-inventor of Google Cardboard, trained a machine learning algorithm using footage shot from a moving vehicle and then had the machine generate this beautiful video.
"Graphics are 100% generated by an algorithm in one shot. No edit or post-processing," Henry writes. "Except the first one, all frames are calculated one by one by a prediction algorithm that tries to predict the next frame from the previous one."
The soundtrack is the Steve Reich masterpiece "Music for 18 Musicians."
Paavo Toivanen wrote code that generates uncannily human, but utterly meaningless and illegible writing. Toivanen's thoughts on generative art are worth reading.
Generative art should ideally retain two disparate levels of perception: the material and visual qualities of a piece of art, and then a creation story or script and the intellectual journey that led to the end result.
The creation story is what's missing in most generative art, especially when it's presented as a representation of nature. Read the rest
Linify Me accepts JPG uploads and redraws the images using only straight lines. The effect is ghostly yet technical, resembling something human-drawn but not enough to be confused as such. Watching the picture emerge over time is strangely meditative. Unless you've uploaded a picture of Trump, that is, in which case it's just another example of something slowly going wrong on a computer.
Kent sez, "Enter your Twitter handle and watch as your tiny online avatar turns into large-scale generative art. Results can look like batik, pastel, or tie-dye, depending on the original."
We're calling the Twitter API from Yahoo! Query Language, receiving an image URL for your avatar, converting it to a data:uri, and returning its base64-encoded value as JSON with a callback.
Then we create an image on the client, load it with the data YQL gave us, and stretch it to fit our (comparatively very large) canvas tag.
Since we've created the image locally, the usual canvas security restrictions don't apply and we're free to sample pixels. We do this, collecting color values and positions, and then we start drawing circles with random sizes and tiny random offsets from where each color sample was taken.