[Enhancement] Remove custom ROS2 camera node & CI

.github/workflows/ros2_code_compiles.yml

@@ -0,0 +1,38 @@
+name: ros_code_compiles
+
+on:
+  push:
+    branches: ["main"]
+  pull_request:
+    branches: ["main"]
+  workflow_dispatch: {}
+
+jobs:
+  build:
+    # The CMake configure and build commands are platform agnostic and should work equally well on Windows or Mac.
+    # You can convert this to a matrix build if you need cross-platform coverage.
+    # See: https://docs.github.com/en/free-pro-team@latest/actions/learn-github-actions/managing-complex-workflows#using-a-build-matrix
+    runs-on: ubuntu-24.04
+
+    steps:
+      - uses: actions/checkout@v3
+      - uses: ros-tooling/setup-ros@v0.7
+        with:
+          required-ros-distributions: jazzy
+
+      - name: Install prerequired packages
+        # To compile and run the code we require cmake, ninja and opencv
+        run: sudo apt-get update && sudo apt-get install build-essential cmake ninja-build libopencv-dev
+
+      - name: Install dependencies
+        working-directory: ${{ github.workspace }}
+        run: |
+          sudo rosdep init || true
+          rosdep update
+          rosdep install --from-paths src -y --ignore-src
+
+      - name: Colcon build
+        working-directory: ${{ github.workspace }}
+        run: |
+          source /opt/ros/jazzy/setup.bash
+          colcon build --packages-select face_detector

CMakeLists.txt

@@ -136,16 +136,6 @@ if (CLFML_ROS2_PACKAGE_BUILD)
   find_package(std_srvs REQUIRED)
   find_package(cv_bridge REQUIRED)
   target_link_libraries(${PROJECT_NAME} ${cv_bridge_LIBRARIES})
-  add_executable(camera_node ${CMAKE_CURRENT_LIST_DIR}/bindings/ros2/camera_node.cpp)
-  ament_target_dependencies(camera_node
-    rclcpp
-    sensor_msgs
-    cv_bridge
-    std_msgs
-    geometry_msgs
-    std_srvs
-    OpenCV
-  )
   add_executable(face_detector_node ${CMAKE_CURRENT_LIST_DIR}/bindings/ros2/face_detector_node.cpp)
   ament_target_dependencies(face_detector_node
     rclcpp
@@ -156,10 +146,8 @@ if (CLFML_ROS2_PACKAGE_BUILD)
     std_srvs
     OpenCV
   )
-  target_link_libraries(camera_node face_detector)
   target_link_libraries(face_detector_node face_detector)
   install(TARGETS
-    camera_node
     face_detector_node
     DESTINATION lib/${PROJECT_NAME})
 endif()

README.md

@@ -153,31 +153,36 @@ git clone https://github.com/CLFML/Face_Detector.Cpp.git
 source /opt/ros/jazzy/setup.bash
 ```
 
-3. Build the package:
+3. Install the dependencies:
+```bash
+rosdep install --from-paths src -y --ignore-src
+``` 
+
+4. Build the package:
 
 ```bash
 colcon build --packages-select face_detector
 ```
 
-4. Set up the environment:
+5. Set up the environment:
 
 ```bash
 source install/setup.bash
 ```
 
-5. Run the camera node:
+6. Run the camera node:
 
 ```bash
-ros2 run face_detector camera_node
+ros2 run v4l2_camera v4l2_camera_node
 ```
 
-6. In another terminal, run the face detector node (don’t forget to source the setup script):
+7. In another terminal, run the face detector node (don’t forget to source the setup script):
 
 ```bash
 ros2 run face_detector face_detector_node
 ```
 
-7. In yet another terminal, run the viewer:
+8. In yet another terminal, run the viewer:
 
 ```bash
 ros2 run face_detector face_detector_viewer
@@ -193,7 +198,6 @@ See our [wiki](https://clfml.github.io/Face_Detector.Cpp/)...
 - Add support for MakeFiles and Bazel
 - Add Unit-tests 
 - Add ROS2 package TPU support
-- Drop custom ROS2 camera node and use a generic node, such as v4l2_camera or christianrauch/camera_ros
 - If needed to crunch ROS2 latency: Do not use nodes, put plugins in a container
 
 

bindings/ros2/face_detector_node.cpp

@@ -42,7 +42,7 @@ class FaceDetectorNode : public rclcpp::Node
 private:
     void declare_and_get_parameters()
     {
-        declare_parameter("camera_topic", "/camera_node/image_raw");
+        declare_parameter("camera_topic", "/image_raw");
         declare_parameter("face_detected_topic", "/face_detected");
         declare_parameter("face_roi_topic", "/face_roi");
         declare_parameter("face_landmarks_topic", "/face_landmarks");
@@ -95,9 +95,6 @@ class FaceDetectorNode : public rclcpp::Node
         std_msgs::msg::Int32 face_detected_msg = std_msgs::msg::Int32();
         face_detected_msg.data = det_.detected() + 1; // +1 because detector returns -1 for no face and 0 for face detected!
         face_detected_pub_->publish(face_detected_msg);
-
-        if (face_detected_msg.data)
-            RCLCPP_INFO(get_logger(), "face detected");
     }
 
     void publish_face_roi()
@@ -154,4 +151,4 @@ int main(int argc, char** argv)
     rclcpp::spin(std::make_shared<FaceDetectorNode>());
     rclcpp::shutdown();
     return 0;
-}
+}

bindings/ros2/set_ld_path.sh

@@ -7,7 +7,7 @@ if [ ! -d "install" ]; then
 fi
 
 # Path to the dynamic library
-LIB_PATH="install/face_detector/lib/face_detector/libtensorflowlite.so"
+LIB_PATH="$(pwd)/install/face_detector/lib/face_detector/libtensorflowlite.so"
 
 # Check if the dynamic library exists
 if [ -f "$LIB_PATH" ]; then

docs/usage/ros2_build_environment/usage_with_linux.md

@@ -34,38 +34,43 @@ git clone https://github.com/CLFML/Face_Detector.Cpp.git
 
 2. Open the cloned repo folder in a terminal
 
-Build the package with more verbose output
+3. Source your ROS2 installation:
 
 ```bash
-colcon build --event-handlers console_direct+ --packages-select face_detector
+source /opt/ros/jazzy/setup.bash
 ```
 
-Note that, whenever build errors occur, and you need to clean the build, use
+4. Install the ros2 dependencies:
+```bash
+rosdep install --from-paths src -y --ignore-src
+```
+
+5. Build the package with more verbose output
 
 ```bash
-rm -rf build install log
-rm -rf ~/.ros
+colcon build --event-handlers console_direct+ --packages-select face_detector
 ```
 
-3. Source your ROS2 installation:
+Note that, whenever build errors occur, and you need to clean the build, use
 
 ```bash
-source /opt/ros/jazzy/setup.bash
+rm -rf build install log
+rm -rf ~/.ros
 ```
 
-4. Set up the environment
+6. Set up the environment
 
 ```bash
 source install/setup.bash
 ```
 
-5. With the environment sourced, we can run executables built by colcon. Let’s run the camera node from the examples:
+7. With the environment sourced, we can run executables built by colcon. Let’s run the camera node from the examples:
 
 ```bash
-ros2 run face_detector camera_node
+ros2 run v4l2_camera v4l2_camera_node
 ```
 
-5. In another terminal, let’s run the face detector node (don’t forget to source the setup script):6
+8. In another terminal, let’s run the face detector node (don’t forget to source the setup script):6
 ```bash
 ros2 run face_detector face_detector_node
 ```

example/ROS_demo/face_detector_viewer.cpp

@@ -32,7 +32,7 @@ class FaceDetectorViewer : public rclcpp::Node
     FaceDetectorViewer() : Node("face_detector_viewer")
     {
         // Declare parameters for topic names
-        this->declare_parameter("camera_topic", "/camera_node/image_raw");
+        this->declare_parameter("camera_topic", "/image_raw");
         this->declare_parameter("face_detected_topic", "/face_detected");
         this->declare_parameter("face_roi_topic", "/face_roi");
         this->declare_parameter("face_landmarks_topic", "/face_landmarks");

package.xml

@@ -20,6 +20,7 @@
   <depend>std_msgs</depend>
   <depend>geometry_msgs</depend>
   <depend>std_srvs</depend>
+  <depend>v4l2_camera</depend>
 
   <test_depend>ament_lint_auto</test_depend>
   <test_depend>ament_lint_common</test_depend>