Instructions to use facebook/sam3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use facebook/sam3 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("mask-generation", model="facebook/sam3")# Load model directly from transformers import AutoProcessor, AutoModel processor = AutoProcessor.from_pretrained("facebook/sam3") model = AutoModel.from_pretrained("facebook/sam3", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Access request follow-up for academic research on video segmentation
Hello SAM3 team,
I have submitted an access request for the SAM3 gated repositories and it is currently awaiting review.
I am a graduate researcher working on an academic project about airport apron operation video understanding. My current goal is to evaluate SAM3 for automatic annotation of apron surveillance videos, including object masks, bounding boxes, and track IDs for aircraft, ground service vehicles, and personnel.
The intended use is non-commercial academic research. I plan to run SAM3 locally, respect the SAM license and model terms, and use it only for dataset construction and evaluation. The video data will not be uploaded publicly or shared unless proper permissions are obtained.
My Hugging Face username is: WuHe8246
Could you please review my pending access request when possible?
Thank you very much for your work and for making SAM3 available to the research community.