andreavolpato/spektrafilm

An exploration of how to make good use of spectroscopic data from manufacturer datasheets in an end-to-end, physically based model with spectral calculations, with the goal of turning that data into convincing film, print, and scan renderings that can be explored interactively.

Here are some useful links and spin-off projects:

In practice, this Python repository is the reference implementation for other developments, and the place where I am growing the underlying model. It lets you start from a camera image, pass it through a virtual negative, print, and scan pipeline, and inspect how film-stock data, couplers, enlarger settings, grain, halation, and other photographic effects shape the final result. The aim is not just to imitate a generic "film look," but to build a model that is grounded in measurements and predicts real-world behaviors of photographic materials.

The desktop GUI exposes the Python tech-demo functionality without writing code, letting you import RAW files or prepared linear images, explore different film and paper profiles, adjust the simulation interactively, and move quickly between fast(-ish) previews and more detailed final scans. Full resolution export is very slow at the moment.