> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://docs.stereolabs.com/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.stereolabs.com/_mcp/server. # Using Object Detection and Tracking with ROS 2 The Object Detection module can be [configured](/docs/integrations/ros-2/zed-stereo-node/#configuration-parameters) to use one of the available detection models: | MODEL | Description | | ------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | `MULTI_CLASS_BOX_FAST` | Any objects, bounding box based | | `MULTI_CLASS_BOX_MEDIUM` | Any objects, bounding box based, compromise between accuracy and speed | | `MULTI_CLASS_BOX_ACCURATE` | Any objects, bounding box based, more accurate but slower than the base model | | `PERSON_HEAD_BOX_FAST` | Bounding Box detector specialized in person heads, particularly well suited for crowded environments, the person localization is also improved | | `PERSON_HEAD_BOX_ACCURATE` | Bounding Box detector specialized in person heads, particularly well suited for crowded environments, the person localization is also improved, more accurate but slower than the base model | | `CUSTOM_YOLOLIKE_BOX_OBJECTS` | For internal inference using your own custom YOLO-like model. This mode requires an ONNX file to be passed in the ObjectDetectionParameters. This model will be used for inference. | | `CUSTOM_RFDETRLIKE_BOX_OBJECTS` | Same as above, for an RF-DETR-like model. Requires ZED SDK >= 5.5 | | `CUSTOM_BOX_OBJECTS_AUTODETECT` | Same as above, with the ZED SDK picking between YOLO-like and RF-DETR-like from the model outputs. Requires ZED SDK >= 5.5 | The `MULTI_CLASS_BOX` and `PERSON_HEAD_BOX` modes use internal models that are provided and downloaded automatically by the SDK. The three `CUSTOM_*` modes allow you to load your own ONNX model. The [Custom Object Detection and Tracking with ROS 2](/docs/integrations/ros-2/custom-object-detection-and-tracking) guide shows an example of how to export this model from a trained PyTorch model. This process must be done only once. The result of the detection is published using a custom message of type [zed\_msgs/ObjectsStamped](/docs/integrations/ros-2/custom-msgs/#object-detection-and-body-tracking-results) defined in the package [zed\_msgs](https://github.com/stereolabs/zed-ros2-interfaces). ## Enable Object Detection Object detection can be started automatically when the ZED Wrapper node starts by setting the parameter `object_detection.od_enabled` to `true` in the file `common_stereo.yaml`. It is also possible to start the Object Detection processing manually by calling the service `~/enable_obj_det` with the parameter `True`. In both cases, the Object Detection processing can be stopped by calling the service `~/enable_obj_det` with the parameter `False`. See the [services](/docs/integrations/ros-2/zed-stereo-node/#services) documentation for more info. > **Note** > > The first time you run the ZED node with the object detection configuration, the SDK will optimize the model for the GPU used on the host device. > > This process may take a long time, depending on the model size and the power of the GPU used. The optimized model will be saved in the `/usr/local/zed/resources` folder and reused in subsequent runs, which will no longer require the optimization process. > > ```bash > [ZED][INFO] Please wait while the AI model is being optimized for your graphics card > This operation will be run only once and may take a few minutes > ``` ## Object Detection results in RViz 2 To visualize the results of the Object Detection processing in RViz 2, the `ZedOdDisplay` plugin is required. The plugin is available in the [zed-ros2-examples](https://github.com/stereolabs/zed-ros2-examples) GitHub repository and can be installed following the [online instructions](https://github.com/stereolabs/zed-ros2-examples#build-the-package). > **Note** > > [The source code of the plugin](https://github.com/stereolabs/zed-ros2-examples/tree/master/rviz-plugin-zed-od) is a valid example of how to process the data of the topics of type [zed\_msgs/ObjectsStamped](/docs/integrations/ros-2/custom-msgs/#object-detection-and-body-tracking-results). ![image](/_fern-img/098402572e865ae1d090e06d4877c53fdcb7e8dc4a8de4018b4cd04db2dad80a.webp) Parameters: * `Topic`: Selects the object detection topic to visualize from the list of available topics in the combo box. * `Depth`: The depth of the incoming message queue. * `History policy`: Set the QoS history policy. `Keep Last` is suggested for performance and compatibility. * `Reliability Policy`: Set the QoS reliability policy. `Best Effort` is suggested for performance and compatibility. * `Durability Policy`: Set the QoS durability policy. `Volatile` is suggested for compatibility. * `Transparency`: the transparency level of the structures composing the detected objects. * `Show skeleton`: Not used. * `Show Labels`: enable/disable the visualization of the object label. * `Show Bounding Boxes`: enable/disable the visualization of the bounding boxes of the detected objects. * `Link Size`: the size of the bounding boxes' corner lines. * `Joint Radius`: the radius of the spheres placed on the corners of the bounding boxes. * `Label Scale`: the scale of the label of the object. ![image](/_fern-img/0f99febb7edd52e892e5f03b109f0eceaff4354a22638809cb514cb6bc70dfde.webp)