BenjaminLi
BenjaminLi
Title is pretty self explanatory.
I know this is a pretty hot topic that has been discussed and has many different implementations, but among them I found the most simple was just to calculate loss...
Hi, I was wondering if it would be feasible to substitute the traditional RGB image style for three different medical modalities during SAM training and inference. For example, instead of...
Why is num_mask_tokens = num_multimask_outputs + 1? And why is it that when you use multimask output, it slices from (1, None)?
Is there a way to add empty box input in batched box input to the prompt encoder? Currently, I have to make a bounding box around the whole image, and...
Why are inputs padded during preprocessing instead of just being scaled up via bilinear interpolation?
Hi, will there be a paper for this or other documents describing all the changes made to the original SAM architecture?
Is it possible to fine tune or evaluate model on custom data?
Hello, thanks for the pre-print and releasing the code. I am working on brain tumor segmentation with the BraTS dataset, which is semantic, as you know. I have two questions...
Hi, is there a way to train models with patience (e.g. stop after a certain number of epochs not improving)? Thanks. Also, is it possible to resume training of a...