PhotoGraph

Customer story

Life histories that surfaced on their own.

A historical society ran PhotoGraph across its entire institutional photographic archive. Here is what it found.

Live & operational

The challenge

Many donors, decades of photos, few names.

The society’s archive was built one donor collection at a time, over many years. The same townspeople appear across those collections, but scans vary in quality, captions are sparse, and nothing linked one donor’s box to another’s.

Much of the archive is group photography: teams, crews, banquets, parades. In those shots people are partly hidden behind one another, heads floating in the back row. That is exactly where facial recognition fails, and it was a large share of the collection.

What PhotoGraph found

  • A life in four photos: a young man in his high-school portrait, then in a group shot of thirty, a formal headshot as a manager, and finally at his farewell banquet. Different donors, different scan quality, donated years apart.
  • Brothers on the same baseball team, kept as two people.
  • A fireman found both as a volunteer at the station and on logging crews.
  • A woman at a recurring protest, matched in everyday clothes and in deliberate costume with her face covered.

Each discovery added a layer to the personal histories of real people and the town they lived in. None of it was findable before.

Entire archive
Every donor collection resolved into one identity graph
On-prem
Runs entirely inside the society’s environment
Own hardware
Runs on the society’s own hardware
API ready
Endpoints live for downstream integration

The reaction

Viewers called it unsettling how accurately it worked.

Staff saw it immediately as a tool for visitors searching for family, for building narrative displays around exhibitions, and for applied uses such as missing-persons and fraud investigations.

What we learned

Group photos are the real test.

Portraits are the easy part of any archive. The value is in the crowded, imperfect photos where most people actually appear. Solving those is what turned a search tool into a way to tell a town’s story.

Your collection could be the next story.

Pilots run on part of your own archive, inside your own environment.