-
ok, I realize there are no future plans for development. Is there a way that the stitching engine and the solving engine could be provided as pre-compiled libraries, and the source for the rest of the app released to the community? this way many of these fixes and features could be implemented without giving away the hardcore technology...
-
http://www.flickr.com/photos/sweeperpix/sets/72157608703660071/ :)
-
well, I'm using xp, but I've output 600mb TIFF files. I generally use HDview for anything that big though because there's not a whole lot you can do with an image that big :D the biggest pano I've successfully stitched was 0.9 gigapixels. I tried to stitch a ~2 GP pic, but no go. There's no such thing as an HDD file size limit, what there are is filesystem size limits, if you're using fat32 (probably not if you're using vista) there's a 2gb limit. I think the limit
-
I'd use planar motion if it worked better :D I tried to use it on a street panorama where I took a picture at 90 deg to the street every 2 meters...it still resolved to rotating motion and never let me force it to planar
-
change theoutput resolution dropdown to 100%. also, per-inch counts are irrelevant, it's the quantity of pixels that matters.
-
-
oh, and something REALLY SUPER UBER BAD is the use of absolute path names in the project file.
-
actually it's under "contact"
-
oh, and as for suggestions, the big kicker in that tutorial is the need to manually edit an xml file of potentially gargantuan size. It would be really nice to have an image list for each panorama, and the ability to, after alignment, substitute images. This would allow for: 1) Easy masking of difficult alignment areas 2) This HDR tutorial sans xml editing 3) time-lapse panorama stitching (you'd need a pano-bot for this) 4) I've seen a couple people talk about projects with several fixed
-
as far as EXIF goes, I think shutter speed at the very least. I agree that aperture is a bit iffy, since focus will actually vary wildly across the whole field. Maybe there's some way to accuately sum or average aperture to represent the theoretical lense that could capture the stitched panorama in its entirety? Probably not necessary. What would rock would be some extra tags saying how many images a given panorama was composed of, maybe with this included one would be justified in including