Quick Start¶

Welcome to 🌈 regenbogen 🌈! This guide will help you get started with the framework.

Basic Usage¶

As the project is currently in early development, detailed usage examples will be added as the API stabilizes.

Example Pipeline¶

Here’s what an example 🌈 regenbogen 🌈 pipeline looks like:

from regenbogen import Pipeline
from regenbogen.nodes import BOPDatasetNode, SAM2Node

# initialize a node reading from the BOP dataset
bop_loader = BOPDatasetNode(
   dataset_name="lmo"
)

# initialize the SAM2 segmentation node
sam2_node = SAM2Node()

# create the pipeline and add nodes
pipeline = Pipeline(
   name="SAM2_BOP_Pipeline",
   enable_rerun_logging=True,
   rerun_spawn_viewer=True,
   rerun_recording_name="SAM2_BOP",
)
pipeline.add_node(bop_loader)
pipeline.add_node(sam2_node)

# run pipeline
for masks in pipeline.process_stream():
   pass

Full example script is available at the examples/sam2_bop_example.py. When run, it spawns a Rerun viewer window displaying the SAM2 segmentation results on the BOP dataset:

Screen capture of SAM2 results on BOP dataset

Next Steps¶

Check out the API Reference for detailed API documentation
Visit the GitHub repository for the latest updates
Contribute to the project by submitting issues or pull requests