facebookresearch / facebookresearch/sam2
Why we force using point input for the SAM task on static images? in Sam2.py, line 182
Open
- Dominant language
- Jupyter Notebook
- Stars
- 19.9k
- Forks
- 2.5k
- PR merge metrics
- No merged PRs in 30d
Description
If i want use masks as prompt for training image to refine, is reasonable to set prob_to_use_pt_input = 0.0?
if num_frames == 1:
# here we handle a special case for mixing video + SAM on image training,
# where we force using point input for the SAM task on static images
prob_to_use_pt_input = 0.0
num_frames_to_correct = 1
num_init_cond_frames = 1
Contributor guide
Assessment
This issue has not been assessed yet.