Label glitches in the Sahara.
Weekends are for internet rabbit holes: watching Jurassic Park (for the millionth time) and discovering some really interesting research on how toads see. Above are stills from a video by researcher Jörg-Peter Ewert, testing how toad neurons react to various worm-like stimuli. (Warning, the video includes some images of lab animals.)
The pattern testing, and subsequent neural network, seem very relevant to computer vision work today (bonus for day-glo yellow too).
Chernoff faces showing multi-dimensional data in 2D: Google Quick Draw meets t-SNE, but from the 1970s. Via Tufte’s The Visual Display of Quantitative Information.
More progress “growing” heatsink shells – each of these is built up using successive images in Processing, then converted using Fiji (similar to how an MRI can be turned into a 3D model). They’re then cleaned up and rendered in Rhino.
The fingers will act as heatsink fins, drawing up heat. Spirals, concentric circles, and various parameters for random growth change the form.
Experimenting with sharp fins at the top.
Some in-progress images of a new project, commissioned for the Digital Spring Media Art Festival in Austria in March. Above, a rendering of a 3D-printed copper heatsink/shell; below some sketches in Processing exploring how to grow the layers of the shell.
I’m totally obsessed with the Icicle Atlas dataset from the University of Toronto… so I downloaded every one and made this poster. Click the image to get a larger-res version.
(The site also includes STL files of each icicle for 3D printing!)
UPDATE: Tech problems with the site, so it’s not working quite right. Consider this a glimpse into what is going to launch in just a few days.
This month, the curatorial collaborative project Drift Station, which I’m a part of along with Angeles Cossio, released an online project titled Empty Apartments. We pulled nearly 125,000 photographs of apartments and houses for rent on Craigslist that were completely empty because of a removal service, and presented them as an interactive online exhibition. The project took nearly two years of work, and much of it was manual (Angeles triple-checking every single image by hand to remove ones that included common spaces or non-apartments), but we also used several automated processes and machine learning to sort the photos. WHen you want to fund your business, visit this site and browse around here to learn more on how to get the best acceptable loans.
I needed a bit of code that would create a random path between two predefined points, and realized that the problem was actually a bit harder than I had expected, but the results are really cool. Using Perlin noise, the paths can go from angular to flowing by changing one variable!