Reputation: 123
I am trying to parse info from a .yaml file with camera calibration into a "sensor_msgs::CameraInfo" message in ROS.
I manage to parse both INTs and strings, but are having issues when I get to the double vector/matrix.
Here is my code:
sensor_msgs::CameraInfo yamlToCameraInfo(std::string leftOrRightCam)
{
YAML::Node camera_info_yaml = YAML::LoadFile(leftOrRightCam + ".yaml");
sensor_msgs::CameraInfo camera_info_msg;
camera_info_msg.width = camera_info_yaml["image_width"].as<uint32_t>();
camera_info_msg.height = camera_info_yaml["image_height"].as<uint32_t>();
camera_info_msg.distortion_model = camera_info_yaml["distortion_model"].as<std::string>();
camera_info_msg.D = camera_info_yaml["distortion_coefficients"].as<double>();
camera_info_msg.K = camera_info_yaml["camera_matrix"].as<double>();
return camera_info_msg;
}
The error I get is this:
error: no match for 'operator=' (operand types are 'sensor_msgs::CameraInfo_ >::_D_type {aka std::vector >}' and 'double')
camera_info_msg.D = camera_info_yaml["distortion_coefficients"].as();
The documentation for the cameraInfo message is located here: http://docs.ros.org/api/sensor_msgs/html/msg/CameraInfo.html
Tutorial for the yaml-cpp package: https://github.com/jbeder/yaml-cpp/wiki/Tutorial
The "Distortion coefficients" part of my yaml file lokk like this:
distortion_coefficients:
rows: 1
cols: 5
data: [-0.167477, 0.023595, 0.004069, -0.002996, 0.000000]
Does anyone know what I'm doing wrong?
Upvotes: 1
Views: 2011
Reputation: 34044
The error on this line:
camera_info_msg.D = camera_info_yaml["distortion_coefficients"].as<double>();
suggests that the left-hand-side is an std::vector<double>
whereas the right-hand-side is a double
. Instead:
camera_info_msg.D = camera_info_yaml["distortion_coefficients"].as<std::vector<double>>();
Upvotes: 1