Check out this section to learn more about object detection.
Check out this section to learn about deep pose processing.
Check out this section to learn about Parameters/Publishers/Subscriptions.
Tiny-YOLOv2 is a real-time neural network for object detection that detects 20 different classes. It is made up of 9 convolutional layers and 6 max-pooling layers and is a smaller version of the more complex full YOLOv2 network.
This sample demonstrates using ONNX Runtime to run Tiny-YOLOv2 against a image stream, and outputing the images annotated with bounding boxes.
To run this sample, a camera will be required to be installed and ready to use on your system.
You can begin with the below launch file. It will bring up RViz tool where you can observe the interaction between object_detection and cv_camera nodes.
ros2 launch ros_msft_onnx tracker.launch.pyTo run this sample, a camera will be required to be installed and ready to use on your system.
You can begin with the below launch file.
ros2 launch ros_msft_onnx pose.launch.py| Property | Description |
|---|---|
| onnx_model_path | Path to the model.onnx file |
| confidence | Minimum confidence before publishing an event. 0 to 1 |
| tensor_width | The Width of the input to the model. |
| tensor_height | The Height of the input to the model. |
| tracker_type | Currently enabled - yolo or pose. |
| image_processing | resize, scale or crop |
| debug | true or false determines if a debug image is published |
| image_topic | The image topic to subscribe to |
| image_debug_topic | The topic name to publish the annotated image stream. |
| link_name | The frame to be associated with the image stream. |
| label | used to filter the found object to a specific label |
| mesh_rotation | The orientation of the mesh when debug rendering pose |
| mesh_scale | The scale of the mesh when debug rendering pose |
| mesh_resource | The mesh used for debug rendering pose |
| model_bounds | 9 coordinates used to perform the point in perspective caluclation for pose |
| calibration | Path to the OpenCV calibration file for point in persective |
| link_name | The frame to be associated with the image stream. |
Onnx subscribes to the topic listed in the image_topic property, or /camera/image_raw
Onnx Publishes the following topics:
The image stream annotated with the bounding boxes to the detected objects.
An array of visualization_msgs::Marker for found objects
A single instance of the DetectedObjectPose message, which is output when tracker_type is set to pose.