Hi, I am using ROS kinetic, Ubuntu 16.04, Python 2.7.

I have the intrinsic parameters of camera but I need to convert them to ROS messages(A form which could be understood by the ROS) and then publish on /camera_info topic. The parameter are shown below:

Camera Parameters

 Source: Matlab Toolbox of NYU_v2_Dataset
 https://cs.nyu.edu/~silberman/datasets/nyu_depth_v2.html
 % The maximum depth used, in meters.
 maxDepth = 10;

% RGB Intrinsic Parameters
fx_rgb = 5.1885790117450188e+02;
fy_rgb = 5.1946961112127485e+02;
cx_rgb = 3.2558244941119034e+02;
cy_rgb = 2.5373616633400465e+02;

% RGB Distortion Parameters
k1_rgb =  2.0796615318809061e-01;
k2_rgb = -5.8613825163911781e-01;
p1_rgb = 7.2231363135888329e-04;
p2_rgb = 1.0479627195765181e-03;
k3_rgb = 4.9856986684705107e-01;

% Depth Intrinsic Parameters
fx_d = 5.8262448167737955e+02;
fy_d = 5.8269103270988637e+02;
cx_d = 3.1304475870804731e+02;
cy_d = 2.3844389626620386e+02;

% RGB Distortion Parameters
k1_d = -9.9897236553084481e-02;
k2_d = 3.9065324602765344e-01;
p1_d = 1.9290592870229277e-03;
p2_d = -1.9422022475975055e-03;
k3_d = -5.1031725053400578e-01;

% Rotation
R = -[ 9.9997798940829263e-01, 5.0518419386157446e-03, ...
   4.3011152014118693e-03, -5.0359919480810989e-03, ...
   9.9998051861143999e-01, -3.6879781309514218e-03, ...
  -4.3196624923060242e-03, 3.6662365748484798e-03, ...
   9.9998394948385538e-01 ];

R = reshape(R, [3 3]);
R = inv(R');

% 3D Translation

t_x = 2.5031875059141302e-02;
t_z = -2.9342312935846411e-04;
t_y = 6.6238747008330102e-04;

% Parameters for making depth absolute.
depthParam1 = 351.3;
depthParam2 = 1092.5;

I have actually succeeded in publishing the RGB and depth images as shown but face problem in converting the above parameters into messages to be published on topic /camera_info.

Code

  rgb = CvBridge().cv2_to_imgmsg(rgb_image, encoding="rgb8")
  depth = CvBridge().cv2_to_imgmsg(depth_gt, encoding="32FC1")
  pub_rgb = rospy.Publisher('/camera/rgb/image_color', Image, queue_size=1)
  pub_depth = rospy.Publisher('/camera/depth_registered/sw_registered/image_rect_raw', Image, queue_size=1)
  pub_cam_info = rospy.Publisher('~camera_info', CameraInfo, queue_size=1)

  rospy.init_node('image', anonymous=True)
  rate = rospy.Rate(0.5)

  pub_rgb.publish(rgb)
  pub_depth.publish(depth)
  #pub_cam_info.publish()

If anyone can help, I will be thankful.

Regards