Hi! First of all, thank you very much for sharing this fascinating work.
I was wondering whether you were planning to release the checkpoints of the trained models for (at least) the VisA dataset.
I would like to re-evaluate this work with another protocol to insert it into a benchmark, so to be as fair as possible with your results I would prefer to have the original weights.
Hi! First of all, thank you very much for sharing this fascinating work.
I was wondering whether you were planning to release the checkpoints of the trained models for (at least) the VisA dataset. I would like to re-evaluate this work with another protocol to insert it into a benchmark, so to be as fair as possible with your results I would prefer to have the original weights.
Thank you again!