facebookresearch / facebookresearch/sam2
Empty Output When `use_mask_input_as_output_without_sam` is False
Open
- Dominant language
- Jupyter Notebook
- Stars
- 19.9k
- Forks
- 2.5k
- PR merge metrics
- No merged PRs in 30d
Description
Hello,
I am trying to prompt SAM2 with masks, and I would like the model to refine these prompts based on the adjacent frames and not just return the prompts in the output. My understanding is that the `use_mask_input_as_output_without_sam` flag controls this.
However, when I change its value to False from its default value of True, I observe zero/empty masks in all of the outputs. Has anyone encountered a problem similar to this or know if there are other flags that can be useful?
Thank you!
Contributor guide
Assessment
This issue has not been assessed yet.