SAM3 Image Segmentation for JPG images
This notebook demonstrates how to use the Segment Anything Model 3 (SAM3) for segmenting JPG images using the samgeo3 module.
Installation
First, make sure you have the required dependencies installed:
| # %pip install "segment-geospatial[samgeo3]"
|
| # %pip install transformers==5.0.0rc0
|
Import Libraries
| from samgeo import SamGeo3, download_file
from samgeo.common import show_image
|
Download Sample Data
Let's download a sample satellite image covering the University of California, Berkeley, for testing:
| url = "https://raw.githubusercontent.com/facebookresearch/sam3/refs/heads/main/assets/images/test_image.jpg"
image_path = download_file(url)
|
Request access to SAM3
To use SAM3, you need to request access by filling out this form on Hugging Face: https://huggingface.co/facebook/sam3
Once you have access, uncomment the following code block and run it.
| # from huggingface_hub import login
# login()
|
Initialize SAM3
When initializing SAM3, you can choose the backend from "meta", or "transformers".
| sam3 = SamGeo3(
backend="transformers", device=None, checkpoint_path=None, load_from_HF=True
)
|
Set the image
You can set the image by either passing the image path or the image URL.
| sam3.set_image(image_path)
|
Generate masks with text prompt
| sam3.generate_masks(prompt="person")
|
Show the results
Generate masks by bounding boxes
| # Define boxes in [xmin, ymin, xmax, ymax] format
boxes = [[480, 290, 590, 650]]
# Optional: specify which boxes are positive/negative prompts
box_labels = [True] # True=include, False=exclude
# Generate masks
sam3.generate_masks_by_boxes(boxes, box_labels)
|
| sam3.show_boxes(boxes, box_labels)
|
Show the results
Save Masks
Save the generated masks to a file. If the input is a GeoTIFF, the output will be a GeoTIFF with the same georeferencing. Otherwise, it will be saved as PNG.
| # Save as binary mask (all foreground pixels are 255)
sam3.save_masks(output="person_masks_binary.png", unique=False)
|
| sam3.show_masks(cmap="coolwarm")
|
