Pigeons make Stained Glass Art
The second drop of the Pigeon Art project went public on Monday May 22nd!
For this drop we wanted to do something different. Stained glass.
Below is one of our artists, Waluigi, standing in front of his latest creation in the studio.
To create these pieces, we started by presenting the pigeon with a canvas of a random color. Each time the touchscreen registered two consecutive pecks, a straight, black line was drawn connecting both peck locations. As the pigeon continued to peck at the screen, it would increasingly be filled with lines.
We had an algorithm running in the background searching for intersections between lines. Every time three or more lines intersected in such a way that it created an enclosed space; the algorithm automatically filled the enclosed space with a random color. As the pigeon continued to peck at the screen, the display would fill with an increasing number of colors…creating a stained-glass appearance.
The pigeons were quite engaged in their art, and often continued to fill the screen with stained glass over a 10- or 20-minute period. Most birds made about 8 artworks on average. A few birds only made between one and three artworks, while our most prolific artist, Evaristo, made 39 separate artworks!
We’ve curated the collection down to a total of 160 pieces in our current drop. We used HeyMint for our drop and each pieces is listed at 0.1 eth. Hat tip to Zeneca for promoting HeyMint! It was great meeting you in LA a few weeks ago. Below is a selfie I took with Zeneca from the ZenAcademy and Debbie Soon, co-founder of HUG, an online studio to support artists and connect them to their fans.
What are we going to do with the proceeds of the sale? All proceeds from primary and secondary sales will be funneled back in to supporting the pigeon research.
While this is a long-term project in the comparative analysis of art, we do have some near-term objectives that the funds will help support.
1) This summer we will conduct our first full analysis of the artworks that our pigeons have already generated. We’ll apply some sophisticated random-walk modeling to gage the stochasticity of each artwork — that is, how random is it versus how much does it reflect the intentionality of the pigeon? This is cool because unlike generative art where the intentionality is built in by the human programmer (artist), we don’t “program” our pigeons and thus any intentionality is coming from their own agency! The pigeon may indeed have an artist’s mind!
2) Going forward with future art creations, we will integrate more advanced analytics to capture during the art creation process. We can record details of each peck (e.g., when it was made, where on the screen), from which we can draw much more nuanced and detailed inferences about the pigeon’s artistic process.
3) Building on the second goal, we will upgrade the artist studio to capture high-quality video of the art creation process. Wouldn’t it be great to watch the pigeon as it creates an artwork? Imagine if a time-lapse video of the creation of an artwork from start to finish were included as unlockable data attached to the static NFT, or it could even be minted as a dynamic NFT! This would also allow for additional analysis of the artistic process using machine learning tools such as DeepLabCut.
4) A more ambitious project is to teach pigeons to sketch! We are developing an app that will automate the training process (called autoshaping in the animal cognition business). Pigeons will be presented with a blank canvas on one side of the display, and a simple shape such as a line on the other side of the display to serve as a sample. The pigeon will have repeated opportunities to recreate on the canvas the sample with which it is presented. Pigeons will receive food reinforcement for each creation that is a close approximation of the sample in things like line length, title, etc. As accuracy improves, we can then test the pigeons with novel objects to see how well they extrapolate their newfound abilities. This is an exciting process because it allows us to compare how pigeons learn to sketch with how human children learn to sketch at early ages. And it will create some cool fridge art as well!
Thanks to everyone who’s been following our journey, and especially those who have supported us either through primary and/or secondary sales and promoting our project.
Special thanks to the dGen Network and Moonbirds and Proof Collective communities for supporting us along the way!
Congratulations, you made it all the way to the end of the article! Thanks for staying with us.
As a reward, I’d like to offer you an allow list spot for a reduced mint price of 0.069 eth. Please use this link to sign up and use the password “pigeon” to register. This is a time sensitive opportunity and ends on Sunday June 3.
