Skip to Content

How Six Degrees of Kevin Bacon Unearthed Lost Photos From American Graffiti

A combination of crowdsourcing and machine intelligence just found a lost treasure in the Magnum Photo archive, a technique that could have wide application

Magnum Photos, possibly the most famous and valuable photographer-owned collection on the planet, knew it had “lost” images from the set of American Graffiti, George Lucas’s first film. It also knew that the way to find them was to get humans to start tagging every image in their archive. Problem is, they couldn’t trust the Amazon Mechanical Turkers to whom they were crowdsourcing the tagging of their archives to be able to recognize images from the set of that movie.

One of the “lost” images from American Graffiti, as featured in a piece from the latest issue of WIRED

That’s where a machine with a pre-programmed (by humans) “semantic graph” comes in, says Panos Ipeirotis, a professor at NYU. Much like IBM’s Jeopardy-winning software Watson, this system had a map of the relationships between different phrases. It’s a bit like the game six degrees of Kevin Bacon, in which players try to leap from one actor to Kevin Bacon by tracing the links between them in the form of movies in which actors have both appeared.

At the intersection of the tags for the actors that appeared in the photographs being processed by crowdsourced human labor was American Graffiti. It was the only movie in which all of these actors had appeared, so it sat at the center of the semantic map connecting them.

In other words, humans plus machines accomplished something that neither could have accomplished on their own for a reasonable cost.

Many, many media companies have large volumes of material which is inadequately tagged with metadata, if it’s even tagged at all. The solution that Tagasauris applied to the Magnum photo archive could probably help unlock treasures in the archives of any company with a sufficiently deep archive.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.