You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello, thanks for open-sourcing the code and for the great work. I am trying to use RobustSam to produce N segmentation masks given N point prompts, in a single image with a single forward pass. From what I understood this is possible from the comments to the forward function here:
However, when I try to do so an error occurs in the mask decoder. I tried a couple of different options, and it seems to work only with a single point prompt. In my case (5 points), it throws the following error. Is this the intended behaviour or am I missing something? In the following screenshot you can see the shapes of my input tensors and the error trace.
The text was updated successfully, but these errors were encountered:
Hello, thanks for open-sourcing the code and for the great work. I am trying to use RobustSam to produce N segmentation masks given N point prompts, in a single image with a single forward pass. From what I understood this is possible from the comments to the forward function here:
https://github.com/robustsam/RobustSAM/blob/6f0e2637e58f812985cdf9f403a5d7515a66fe38/robust_segment_anything/modeling/sam.py#L74C1-L74C71
However, when I try to do so an error occurs in the mask decoder. I tried a couple of different options, and it seems to work only with a single point prompt. In my case (5 points), it throws the following error. Is this the intended behaviour or am I missing something? In the following screenshot you can see the shapes of my input tensors and the error trace.
The text was updated successfully, but these errors were encountered: