What if you could search a museum collection by what's actually in the image, not just...

What if you could search a museum collection by what's actually in the image, not just what a curator wrote down decades ago?

LinkedCulture Vision does exactly that.

Using AI image captioning across Met and Harvard Art Museums collections, we've generated semantic descriptions for over 1,000 objects that previously had little or no descriptive metadata. Now you can search across those records by visual content. Type "vase" and find ceramic forms across both collections, ranked by semantic similarity.

The captions aren't tags. They're full natural language descriptions generated from the images themselves, then embedded into a vector search layer that sits across all of LinkedCulture's collections.

This is what the coordination problem in cultural heritage actually looks like when you start solving it. Not a new database. Not a new standard. A semantic layer on top of what already exists.

LinkedCulture is a cross-collection search platform unifying the Met, Harvard Art Museums, Smithsonian Asian collections, and now the Getty. More on that next.

Try it yourself, enter "vase" into the search and select "LinkedCulture Vision" as the mode.

https://linkedculture.org/