On August 23, 2010 01:30:26 pm Bruno Postle wrote: > When the type of lens is unknown, it would be useful to rerun the > optimisation using a different lens type and compare the error > distances. This could be done automatically, though ideally we would > have a nice lens database that just knew all this stuff.
the known unknown? I mean: it is so easy to lose the EXIF data along the way that by the time the images hit Hugin, it can't know what type of lens it is, nor its focal distance. > These are approximately the steps of the existing 'multi-row' > control-point detection, it is slow at the moment because it relies on > the XML keypoint-file caching of autopano-sift-C. The new panomatic > code will be more efficient with this caching. I have not tried the 'multi-row' control-point detection yet. Yuv
signature.asc
Description: This is a digitally signed message part.
