Is it possible to train sam3 with both IMAGE + TEXT?
same question, did a little bit on detection but the model performed worse, suspecting it got confused at the finetuning concept with its own pre-trained data
· Sign up or log in to comment
