Colorizing Museum & Library B&W Collections at Scale

Every archive holds the same quiet paradox: the older the photograph, the more the world it captured has faded from living memory, and yet that photograph survives only in grey. A university special-collections department might steward forty thousand campus and community images. A regional museum might have a donor's shoebox of glass-plate scans. A public library's local-history room might have digitized its entire microfilm run of a shuttered newspaper. The material is preserved. What it lacks is the immediacy that makes a stranger stop and look.

Colorization closes that gap — but for institutions the technology has never been the hard part. Governance is. You cannot hand a hundred-year-old donor collection to an anonymous website that asks you to "drag your files here," and you cannot ask a volunteer to process ten thousand images one at a time through a browser tab. This is where museum photo colorization and archive digitization colorize workflows tend to stall. Revive is built for exactly this situation: it colorizes your collection in place, by folder, inside the Google Drive you already control.

Why "in-place" matters for a collection

Most consumer colorizers follow an upload-and-download loop: you export images out of your system, upload them to a third-party server, wait, then download a bundle and re-file everything by hand. For three holiday snapshots that is fine. For an accessioned collection it is a governance failure and a logistical nightmare in equal measure. Every export is a copy you now have to track. Every upload is a chain-of-custody question you have to answer to a records manager, a donor agreement, or an IRB.

Revive is a Google Workspace add-on that runs where your photos already live. It reads your images through Google's Drive APIs, colorizes them, and writes the restored copies straight back — the files never leave Google's infrastructure. There is no third-party storage bucket, no "your photos here" upload box, and no copy sitting on a server you don't control. The add-on requests only the narrow drive.file scope, so it touches the files you point it at and nothing else in your Drive.

Just as importantly, your originals are never modified. Each restored image lands in the same folder as its source, named filename-color.jpg, so a scan called 1948-quad-dedication.tif produces 1948-quad-dedication-color.jpg right beside it. The master stays untouched; the color version sits alongside it as a derivative, exactly the way a preservation workflow expects derivatives to behave.

Process by folder, keep the structure

Institutional collections are organized, and that organization is itself a record. Revive is folder-aware: point it at a top-level collection folder and it expands the entire tree, queuing every supported image inside it. The sub-folder structure is preserved in the output, so a hierarchy like /Photographs/1920s/Athletics/ comes back with its color derivatives filed in the same places. You are not flattening a carefully arranged collection into an undifferentiated dump and then spending a week re-sorting it.

That folder model is what makes scale practical. A single job can be one image, a hundred, or ten thousand. Revive colorizes roughly 1,000 photos in about a minute — on the order of 50 to 100 milliseconds per image — so a departmental backlog that would have consumed a season of volunteer hours becomes an afternoon of queued jobs. Supported inputs cover the formats archives actually hold: JPG, PNG, HEIC, TIFF, and WebP, all producing a color JPEG.

Governance: the audit trail and shared billing

For a home user, "who colorized this?" never comes up. For an institution it is the whole conversation. Revive attributes every job in a usage report: what ran, how many images it spent, and which Workspace user launched it. When a registrar or a department head asks how the digitization budget was used this quarter, the answer is a report, not a reconstruction from memory.

Billing is designed for the same reality. Image credits are tied to your domain billing account and are usable by every Workspace user across the institution. The special-collections librarian, the exhibits team, and a graduate assistant all draw from one shared pool — no per-person purchases, no expense-report friction, no license seats to reconcile. And because credits never expire, a department can buy against an annual budget and spend it across a multi-year digitization project without losing anything to a subscription clock.

In-place vs. upload-to-a-tool

The difference is easiest to see side by side. One approach treats your collection as something to export and hand off; the other treats it as something to work on where it lives.

What it costs at collection scale

Revive is pay-per-image with no subscription: one image equals one colorized photo, and larger packs lower the per-photo cost. For a department planning a digitization push, that maps cleanly onto a fixed budget line. Here is the ladder — savings are measured against a $0.10 list rate.

A collection of, say, 40,000 images is squarely custom-volume territory, where the per-photo cost falls toward five cents — a number small enough that colorizing an entire holding becomes a line item rather than a project. See the full breakdown on the pricing page, and if you want to watch the mechanics end to end, the how-it-works walkthrough shows the side panel, the folder queue, and the write-back in sequence.

Getting a pilot running

The lowest-risk way to evaluate Revive for a collection is to try it on real material, not a demo. Every new signup includes 3 free images, no credit card — enough to colorize a few representative scans from your own archive and judge the results on the material you actually steward. From there, an administrator can add the add-on domain-wide, load shared credits against the department budget, and let the whole team draw from one pool.

If you are weighing this against colorizing a personal or family archive rather than an institutional one, the family-archive walkthrough covers the same in-Drive workflow at a smaller scale and is a useful companion read.