After hours of reading, then mucking through several video tutorials, I finally rendered my first couple of Camera Tracking animations using Blender. Compared to just rendering a simple image, learning camera tracking was like getting thrown into the middle of the ocean and learning how to swim. I just uploaded the video clip to YouTube that show both clips with some text-commentary mixed in.
These two clips took about three days from start to finish, getting the software (Blender) figured out, then testing and shooting video with my camera, setting up and testing the markers for the camera tracking, and then the rendering. I went through the process at least five times with each background clip until I was happy that the rendering was going to work. Then the scene creation in Blender, when I had to get the lighting, materials on the objects and object placement just right, was also a huge time vampire. The tricky part was trying to use, and track, markers so that I could keep as many markers as possible from start to finish. In every attempt I lost markers with Blender. My last scene, which was the best, I lost only three out of eleven.
The first clip I used screws. They worked, but just enough to get through a clip with only 320 frames. Any longer, they were all lost, and the Render stopped.
The second clip I used white paper squares. At first I lost all of them in the scene set-up, but then I make the tracking search area larger, and my success rate was 90 percent. However, the larger the tracking size, the longer it took to render. Oh the sacrifices.