Growing Masks made from BiRefNetSegmentation

[brandostrong]

Hi,

Not sure which node issues to go to, but when trying to grow a batch of masks I get the error: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.

Converting the masks from the before node to images, back to masks, or any other kind of operation still returns this error. The masks are created using cuda, as cpu would take hours.