Case study

The Memory Loop

An AI pipeline that brings degraded black-and-white film back to life: cleaned, upscaled, colorized, and graded to HD. Methodical, evaluation-driven, and built around a custom tool I wrote where none existed.

RoleDesigned the pipeline, built the tool, run the channel
Year2025
StatusPublic channel, ongoing
StackFFmpeg · Node/Express · AI upscalers and colorizers
Watch the restorations on YouTube
The problem

Every old film is broken differently.

Archival footage is a mess in inconsistent ways: wrong or unstable frame rates, duplicate and blended frames, heavy noise, flicker, instability, low resolution, and no color. Worse, every camera and film stock degrades differently, so a setting that rescues one clip ruins another.

Restoring it well is not a one-click job. It is a sequence of specialized steps, each tuned, and in the AI stages, chosen from several competing models.

What I built

A repeatable pipeline, and the tool it was missing.

I designed the whole pipeline, ran it end to end on real films, and built a custom tool for the step that needed one. The AI-heavy stages, upscaling and colorization, are treated as decisions to be tested rather than assumed. I also run the public channel the pipeline produces.

The pipeline

From damaged archive to smooth HD.

01 Restore original frame rate 02 Duplicate Frame Removal 03 Damage clean-up 04 Stabilization 05 AI upscaling 06 AI colorization 07 Frame interpolation 08 Color grading

Duplicate Frame Removal is a tool I built; upscaling and colorization are where competing AI models are tested per clip and selected on results.

The custom tool

Duplicate Frame Remover.

Old footage is full of duplicate and blended frames from frame-rate conversions, and off-the-shelf removers did not fit the archival workflow. So I built my own: a local web app (Node.js and Express backend, browser UI) that wraps FFmpeg's mpdecimate filter. Self-contained, no secrets, and open source.

Codec-aware

Probes the input with ffprobe and automatically matches the output codec, profile, and pixel format (ProRes, H.264, H.265), so quality is preserved.

Live FFmpeg preview

An adjustable threshold slider drives a live command preview that advanced users can edit directly, while the server still controls the file paths for safety.

Editing-ready output

Converts the result to a constant frame rate, which downstream editing tools require.

Reports and self-manages

Returns frame-removal statistics (how many frames, what percent, how long) and manages its own upload cache.

View the source on GitHub
The evaluation story

Models compared, not assumed.

Because each clip degrades differently, the AI stages run as comparisons. For each film I test competing models and choose the best, often a combination, the same discipline frontier-model evaluation requires.

Evaluated per clip

Upscaling

Topaz (Proteus, Artemis, Iris), BasicVSR++, Real-ESRGAN, and Starlight, compared per clip for detail, sharpness, and artifacts, often combined (a Proteus pass followed by an Artemis or Iris pass).

Evaluated per clip

Colorization

DeOldify, HAVC, and ColorMNet, compared for color accuracy and temporal stability.

Before and after

Two test pieces, two hard problems.

Each clip below was chosen to stress one specific stage of the pipeline. Press play and watch the before and after.

Colorization test

Ripley (2024)

A modern series shot in black and white on today's cameras, which makes it the perfect benchmark for colorization: the source is technically pristine, so the color has nowhere to hide and every choice the pipeline makes is visible.

Restoration test

Un Chien Andalou (1929)

The surrealist short by Luis Buñuel and Salvador Dalí. Dalí is one of the artists whose work inspired The Memory Loop in the first place: the project exists to preserve the work of artists I respect and to see it in a new light. The film's pioneering visual effects also make it a demanding proving ground for the restoration process I built.

Outcome

The proof is visual.

Published HD restorations on the public channel, with before and after comparisons showing the jump from damaged black-and-white source to smooth, colorized HD. Featured public-domain work includes Un Chien Andalou (Luis Buñuel, 1929).

Need an AI media pipeline?

I build generative-media pipelines and the custom tools that make them work, with evaluation behind every model choice.