YOLO node for instance segmentation

[kluge7]

Motivation

We currently have no ROS setup for running YOLO instance segmentation models.

Description

We want to build a simple ROS2 node for running YOLO instance segmentation models (e.g. YOLOv8). The node should subscribe to a color image topic, perform inference using a specified model, and publish both the segmentation masks and an overlay image with visualized detections. It should have a launch file and config file.

Task

1. Implement the ROS2 node in vortex-deep-learning-pipelines

• Use ros2cli for making the package
• Subscribe to a color image topic
• Run YOLO Instance segmentation inference
• Publish segmented masks and overlay image

2. Config file

• Define parameters (model path, topics, thresholds, device, etc.)

3. Add launch file
4. Test functionality

• Verify on local setup
• Use simulator or ROS bag

5. Documentation

• Add README on how to launch and configure the node

Deliverables

• ROS2 node: Performs YOLO instance segmentation and publishes masks + overlay image
• Config file
• Launch file
• Test verification: Confirmed working locally
• Documentation

References

• Ultralytics YOLO instance segmentation - https://docs.ultralytics.com/tasks/segment/
• ros2cli for making package - https://github.com/vortexntnu/ros2cli
• Launch file example - https://github.com/vortexntnu/software-learning-period/blob/main/tutorial_packages/simple_publisher/launch/simple_publisher.launch.py
• Config file example - https://github.com/vortexntnu/software-learning-period/blob/main/tutorial_packages/simple_publisher/config/simple_publisher_config.yaml
• Similar node (use as example) - https://github.com/vortexntnu/vortex-yolo-inference

Contacts

@kluge7
@jorgenfj