multiple papers about verifying edits on an image with zk-snarks recently:
- VerITAS: Verifying Image Transformations at Scale: using lattice based hash+poseidon alongside Plonky2
- VIMz: Verifiable Image Manipulation using Folding-based zkSNARKs: using Nova

I think we're still not there for something practical, but it's super cool to see ZK used outside of blockchains

could you elaborate on why proving image transformation is not practical yet?

Why I said that is because as far as I understand is that the setup/proving time would 2 to 6 minutes _per edit_ (verifier time is super fast tho)

I think VerITAS actually claims it's already practical, because it's been created with news organizations in mind, so they could take the time to prove their edits, which is fair 

Maybe I was wrong :D

Also, I'm not sure, but I guess that for proving the transformations, you need to use a specific software to do these edits? which seems a bit unpractical for me so far, but that would be fun to integrate it to already existing OSS for edits
