You definitely should use a tripod to reduce parallax but more importantly you
should understand the "entrance pupil" for your lens. This is the point in the
lens that you should be rotating about so there is no parallax. Do some
research on the web to better understand this requirement.
On 2013-08-29, at 1:08 PM, John Eklund <jclo...@gmail.com> wrote:
> You should always use a tripod to avoid parallax errors (ghosting). Even
> then, it will sometimes shift slightly after adjusting it so images are
> sometimes misaligned by one or a few pixels. You can often get away with
> shifting (translating) them in Photoshop or similar but Hugin will usually do
> a better job. You should limit optimizations to Yaw, Pitch, Roll.
>
> I got the same idea back in 2007 when panorama-documenting an old power
> plant. It was a huge project and I ended up with a huge pile of over 60.000
> images that's still far from finished post-processing. I spent a year writing
> a personal tool ("PicStack") from scratch in C, which identifies and merges
> similar images and finds the best placement automatically. I also added a set
> of color adjustment functions and automatic HDR merging with diffuse masks
> generated dynamically but then Enfuse was released and it was superior for
> that so I don't use that feature much anymore. The main purpose of Enfuse is
> not stacking similarly-exposed images but it does work for that as well. I
> personally consider that a "bonus" though and still prefer using my own tool
> for that to know what I get and then if I need exposure fusion I'll pass my
> output through Enfuse in the end. I intended to release my tool to the public
> but as it's always been "in development" I didn't get that far. The
> parameters to differentiate likely HDR-intended shots from non-HDR especially
> needed endless tweaking.
>
> I usually took 4-5 identical images, sometimes more, and it really saved my
> neck as you can squeeze lots more dynamic range and shadow detail out of a
> cheap camera's JPEGs. There were cases when I retained grey levels from the
> pitch black corners where pixels only consisted of noise flipping binary
> between 0 / 1! By doing a radical gamma curve adjustment to bring up the
> light before merging, you can regain lost detail almost as if by magic if you
> have enough shots.
>
> I used this tool for all the panorama material below. (Sorry for
> self-promotion) :-)
> http://www.angkraftverket.se/datidnutidframtid/merainfo.html
>
>
> --
> A list of frequently asked questions is available at:
> http://wiki.panotools.org/Hugin_FAQ
> ---
> You received this message because you are subscribed to the Google Groups
> "hugin and other free panoramic software" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to hugin-ptx+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/hugin-ptx/9981ebe9-4c3f-4d18-a9d5-080cc6b9ee12%40googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
--
A list of frequently asked questions is available at:
http://wiki.panotools.org/Hugin_FAQ
---
You received this message because you are subscribed to the Google Groups
"hugin and other free panoramic software" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to hugin-ptx+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/hugin-ptx/B23CC429-94E1-407F-B090-8639F5BF8D38%40accesscomm.ca.
For more options, visit https://groups.google.com/groups/opt_out.
