Closed onslauth closed 2 weeks ago
Hi,
Do these converted models support Automatic Mask Generation without the need for a point or bounding box?
Sorry, this can be closed.
I looked into how SAM v2 is doing its Automatic Mask Generation, and it just creates bounding boxes to fill the image, and amalgamates the results.
Hi,
Do these converted models support Automatic Mask Generation without the need for a point or bounding box?