Hello Hugin users,
I've been using the software for a couple months, and I really like it. Thanks to the developers for creating an awesome project that's still relevant many years later. I have a question about how you guys handle blank space with nothing to create control points on when shooting your panoramas. It becomes an issue shooting 360 deg. shots that include the sky and gets worse when using a high focal length lens. I've learned that I can do a second pass of the sky with a wider angle lens in order to fill in any missing pixels, albeit at a lower resolution. Content-aware fill also does a relatively good job of filling in clear, blue sky.
Which brings me to my next question. Now that I have overlapping shots, how can I make sure Hugin only uses the lower quality, wide angle shots only to fill in pixels not covered by the better lens? From what I understand, I should just need to place them at the end of the image list. Is it possible that Hugin would overlay more than what it needs to?
I've been looking at the --visualize output, it's hard to see how much of the image is used. I know there's an option to write the intermediate output panorama, but it overwrites the image each time. I might write a script to copy it every time it changes to see a timeline of how it's pieced together.
Lastly, for those of you that shoot with a panoramic head, do you rotate it the same amount for each vertical angle? How much overlap do you go for? I've been using 1/2-1/3.
Also, does anyone have experience compiling BigTIFF support in Linux? 2 GB is so tiny for a 360 panorama.
Thanks,
Jesse