PPS
On Jul 20, 10:24 pm, Tom Sharpless <tksharpl...@gmail.com> wrote: > PS I was wrong about that last. PTAsm lets you choose between the > equal-angle and equal area fisheye models, which is the right thing to > do. > However it computes fov wrong for the equal area fisheye, so it's not useful. > -- Tom > > On Jul 20, 4:08 pm, Tom Sharpless <tksharpl...@gmail.com> wrote: > > > Hi > > > All of this indecision about which image axis to link to "fov" is > > unnecessary and misleading, because fov is properly a secondary > > quantity, derived from the following primaries: focal length (in > > pixels); image width (ditto); lens projection function (angle vs. > > radius). > > > A sensible image processing program would show you the fovs on both > > axes, or indeed in any direction you choose. > > > I'll repeat. If you don't like to think about focal length, or the > > idea of measuring it is pixels repels you, you don't have to think > > about it at all. The EXIF data from all modern cameras supplies the > > necessary information, and your stitcher knows how to interpret that > > (well maybe not PTGui, see below). And in case you have a camera that > > does not report focal plane resolution, you _should_ only have to > > enter its sensor dimensions or crop factor once, and your stitcher > > _should_ be able to fetch that out of its camera data base or > > preferences table from then on. Hugin developers take note. > > > On Jul 20, 12:33 pm, Bruno Postle <br...@postle.net> wrote: > > > > On Mon 20-Jul-2009 at 07:53 -0700, Bart van Andel wrote: > > > > >Maybe FOV should be treated differently based on the lens type. This > > > >does make sense, as for circular fisheye the given (or guessed) fov of > > > >the lens is only the fov of the circle, not of the full frame. > > > > This is how ptgui does it. > > > I've noticed that. And I just noticed something else odd about PTGui: > > it gives a bogus crop factor for my Canon EOS 30D. > > > The correct crop factor, according to the EXIF focal plane resolution > > data, is 1.60149, which is what Hugin 0.8 reports. But PTGui 8.1 says > > it is 1.5870. I suppose this must be a bug, and have reported it on > > the PTGui support board. > > > > Hugin does it the right way: the 'crop circle' for a circular > > > fisheye simply indicates the outer area of the frame that should be > > > ignored when rendering, it is unrelated to any other lens > > > parameters. > > > But PTGui computes fisheye fovs according to the equal-area formula, > > while Hugin still uses the inappropriate equal-angle formula (so does > > PTAssembler 5.1). > > > Regards, Tom --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "hugin and other free panoramic software" group. A list of frequently asked questions is available at: http://wiki.panotools.org/Hugin_FAQ To post to this group, send email to hugin-ptx@googlegroups.com To unsubscribe from this group, send email to hugin-ptx-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/hugin-ptx -~----------~----~----~----~------~----~------~--~---
