35 points surprisetalk 3 hours ago 8 comments
Wryivy94 1 hour ago | parent
Curious whether the embeddings come off the images or the catalog text, because in my experience museum metadata is wildly uneven (half the Met's older records are basically "Vase, terracotta"). Also, do you dedupe across institutions?
dominotw 1 hour ago | parent
I searched for 'soviet' and it matched cylinder seal 3500bc
https://lastmuseum.com/art/cylinder-seal-307
soviets were big into seals and govt procedures . maybe?
MitPitt 50 minutes ago | parent
This website likely uses a semantic vector space like CLIP or BERT space. Which is based on co-occurrence of words, it doesn't have a harder logic behind it. Soviets probably had cylinders in some machinery which is why it pulled.
traceroute66 1 hour ago | parent
Interesting concept. Its desperately missing a decent filter function though.
Just dumping a million results on the user to doom-scroll through is a bit problematic in terms of UI/UX.
scanny 36 minutes ago | parent
Would make sense to have a copywrite filter so you can specify to see only public domain images.
whilenot-dev 34 minutes ago | parent
There's nothing between 1975-2026?
EDIT: Ok, there is, but it needs a query...