How are photos selected for CV training?

I’d love to see some examples of different photos of the same taxa where you think this might be happening, I’d be happy to do some evaluation to see if our current models underperform on wild or captive variants.

I also need to double check, I said earlier that we prefer RG but now that I think about it, I’m not so sure. I know we prefer community id (observations that have agreement from multiple people) but I’m not sure if we give preference based on quality grade. Related but different criteria. I’ll have to get back to you.