Hi ys,
don't have DSLR yet. Wanted to go the canon 10D route... but decided to wait for the Nikon D70. I need a Cam that'll do other work then just pano work and the D70 has a 12 shots in 4 secs buffer.... now that is friging good. Also am hearing many very positive reports on the D70 with Sigma 8mm.... though the Nikkor 10.5mm is a better lens... also a lot bigger price.
At present I use the CP4500 and am quite happy with the resolution. I'm even creating full screen QuickTime work with it now with satisfactory results. 1MB images though. I get good results in full screen mode using following workflow.
CP4500 & FC-E8 in portrait mode. White bal fixed, Auto Image adjust, slight Auto sharpening, image size 2272. Cam is set to M with f at 8.3 and adjust shutter speed accordingly. I like to zoom in until the images touches the top and bottom of my view screen. This results in a 180deg x 140deg image. I don't do exposure fixed as I tend to take 3-5 images in each 120deg for later blending in PS. I find that it's not needed due to the above set up. The only issue I seem to have with this technic is some lens flare or shadowing around the image edges. This didn't occur with my old cp990.. and I don't quite know why this is.
Anyhow, I then import the multiple images into PS and do some blending work for blownout windows, highlights... then save the resulting image as BMP. Then I import the bmps into easypano and stich in 4000x2000 size. This has been possible due to me using as much of the chip as possible and therefore having approx 2100 pix in image height. Also I tend to use about 80% of the chip instead of only 40% when going fe1.
I save the pano in bmp format, then reimport and convert to cubic (here I like to keep the equirec. image as backup base image and tend to save the cubic as seperat file) do the nadir and zenith cleanup and any other tidying up thats needed. Then reconvert to equirec. and save. I like to have a cubic and equirec image each as I tend to output in two formats... Java and QT. I don't actually do any sharpening or image adjsutment until I'm ready to use the base image. This is important especially when downsizing image to say 1400x700 for web 300x400 viewer sizes. U only really want to sharpen once the image is downsized. I also love PS 7's autocolour which tends to give me what I want without much fiddling around 80% of the time. I'll put up a link to a full screen QT later.
Cheers Rob
------------- If I only had an hour to chop down a tree... I'd spend 45 mins sharpening the axe.
|