facebookresearch / facebookresearch/segment-anything
Getting specific masks (closest/centre/tag/etc.)
- Dominant language
- Jupyter Notebook
- Stars
- 54.9k
- Forks
- 6.4k
- PR merge metrics
- No merged PRs in 30d
Description
The greetenings,
I have been playing around with this for a while now and tried to see if I could make it work without an additional "selection" step. I have tried to read around this place, in the issues, and also looked at the code of source, but since time is of the essence, it's not been as extensive (yet), I know I can also try to compare sizes, check for borders, but not sure if there is any grouping/hierarchy involved in the masks just yet (which would make this a lot easier).
So can anyone tell me if it is possible to get selected masks based on some kind of inherent/heuristic/selective method that does not involve an additional user step? In most cases I just like to have the masks that belong to the "focal" object.
Thank you all very much in advance for helping me out.
Contributor guide
Research direction
The issue does not identify a file, test, or entry point. Start with the repository source code and example notebooks, then determine whether mask selection can be made without an additional user step. Done would require a defined and validated way to select masks belonging to the focal object.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- jupyter-notebook
- Domain
- computer-vision, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100