I'm using cmr_sg with cmr_sg_res18_freihand.pt parameters, and applying that on my data creates some errors on images that seem ok. Could you please share some insights on what's going wrong here?
By the way, similar issues happen to the evaluation set of Freihand too:
I'm using cmr_sg with cmr_sg_res18_freihand.pt parameters, and applying that on my data creates some errors on images that seem ok. Could you please share some insights on what's going wrong here?
By the way, similar issues happen to the evaluation set of Freihand too: