facebookresearch / facebookresearch/sam3
Few-Shot Fine-tuning
- Dominant language
- Python
- Stars
- 11.7k
- Forks
- 1.8k
- PR merge metrics
- No merged PRs in 30d
Description
Hey Team,
Great work! Huge advancements can be seen from SAM to SAM3 in last 2.5 years. I just had a query:
Can you please provide some reference scripts to fine-tune the model for a different domain for few-shot segmentation? I can't find anything in the repo related to it. So, it will be a great help if you can provide any help regarding to it.
Thanks,
Anit
Contributor guide
Research direction
Start by reviewing the repository's existing Python fine-tuning code and example notebooks to determine what domain-adaptation path is exposed. Compare that path with the requested few-shot segmentation workflow; done means providing reference scripts and a reproducible fine-tuning example for a different domain.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- computer-vision, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 28/100