Process

Generative Gestaltung was a key reference for understanding how computational processing can translate imagery into typographic visuals. It provided a way to think about parameters, variation, and output as parts of one system.

Code snippets were used as learning material, rather than intergrated directly into the project. The focus was on understanding how their sampling, mapping, and rendering connect, then rebuilding those fundamentals around the requirements of Type-To-Face.

1. Input The browser requests webcam access or prompts image upload, and uses the input as the image source.
2. Image Sampling The video/image frame is converted to a grid so each cell can be measured and translated.
3. Typographic Mapping Brightness & tonal values are mapped to glyphs and typographic behaviour.
4. Rendering Engine The mapped glyphs are drawn to the grid as a live typographic reconstruction.
5. Configuration System Parameters such as cell size, density, text scale, and tracking remain adjustable.

The first prototype focused on the systems ability to produce a visually recognisable outcome: image input to typographic output.

This proof of concept tested whether a live camera feed could be sampled, mapped, and rendered typographically with good performance before broader parameters were added.

Refinement focused on adjustable parameters including cell size, density, text scale, and spacing.

Each adjustment involved a trade-off. Smaller cells increased facial detail but reduced typographic legibility, while larger cells made the letters clearer and pushed the image further into abstraction.

The process negotiated abstraction and clarity. The aim was not photographic accuracy, but a reconstruction that remained recognisably facial while foregrounding the type.