There are obviously so many directions you could go with an active visualization and exploration experience for this data set.
My question for you then, is--what are you trying to enable your visual explorers to do better? How specific can you be in articulating what "problem" you want to solve for them?
E.g., Maybe you believe that art researchers or "visual explorers", whomever fits your target user here, need to be able to more swiftly filter the collection along different dimensions and then see the results grouped by country so as to see commonalities across countries that weren't apparent before. If so, that will inform how you design the interactive visualization.
With your current version, I would think more about the tag/filtering experience. Current behavior--when I click a tag, I'm forced to sit and watch it cycle through items with the same tag.
Forcing a user to play along at a certain speed is generally an issue, unless of course there is a narrative that is inextricably bound to such temporal constraints, and wouldn't make sense without them.
Great work though! Thanks for sharing.