Improved thermal camera tutorial (#270)

Signed-off-by: ahcorde <[email protected]>

tutorials/thermal_camera.md

@@ -299,7 +299,7 @@ Here's an example 16-bit thermal camera subscriber that performs this conversion
 ```c
 #include <cstdint>
 
-#include <gz/msgs.hh>
+#include <gz/msgs/image.pb.h>
 #include <gz/transport.hh>
 
 // used to set the proper resolution of the camera's output (10mK)
@@ -346,6 +346,21 @@ int main(int argc, char **argv)
 }
 ```
 
+Save the above code to a file named `main.cpp` and use this `CMakeLists.txt` to build it.
+
+```cmake
+cmake_minimum_required(VERSION 3.5 FATAL_ERROR)
+project(image-listener)
+
+# Find the Gazebo Libraries used directly by the example
+find_package(gz-msgs9 REQUIRED)
+find_package(gz-transport12 REQUIRED)
+
+add_executable(${PROJECT_NAME} main.cpp)
+target_link_libraries(${PROJECT_NAME} PUBLIC gz-msgs9 gz-transport12)
+target_include_directories(${PROJECT_NAME} PUBLIC ${gz_msgs9_INCLUDE_DIRS})
+```
+
 Although most of the code above is described in the comments, let's go over the key points again:
 * The image published by the camera is represented as serialized bytes.
   In order to read the actual temperature data that's stored in the image, we need to convert the data in the image to unsigned 16-bit integers.