Dispatches

AI for image analysis. Using OpenAI’s CLIP model, I generated embeddings for c. 500 of my own photographs. Cosine similarities were calculated for every pair of images to generate sets of ‘similar’ images - these similarity scores are the basis for the network layout. Using UMAP, the embeddings were reduced to a smaller set of features and clustered with HDBSCAN, producing seven (mostly) coherent groups of photos. Lastly, BLIP-2 was used to generate image captions.

sketchbook.ownkng.dev, a new companion site to this one to host some experiments in generative design and other creative endeavors. Check it out

A terminal UI written in Go. Made with the fantastic Charm libraries, a little terminal UI to practise Chinese characters. View on GitHub.

A few sketches. Some new generative sketches created with P5.js. Using a few techniques, including perlin noise, delaunay triangulation, poisson disk sampling and the chaikins algorithm.

A network analysis of London bus routes. An interactive exploring the connections between bus routes in Greater London. Based on data from TFL, the visuals show the most (and least) connected routes in the capital’s transport network. Check it out here.

A low poly interactive made with react-three/rapier. An interactive scene made from a GLTF file using react-three/rapier and use-gesture (for the touch controls). The orthographic camera makes makes the scene pop nicely. Check it out here.

Multiplayer three.js with PartyKit. A multiplayer emoji painter made with PartyKit and three.js decal geometries. Try it out here (best enjoyed with friends) or view the server and client code on GitHub.

Turning GLTF files into line drawings. A sketch made by sampling points from a GLTF file and then connecting them with three.js line segments.