Hi, I'm new to lens fun and so have some questions :
1) From what I have see in the sources, pixels coordinates are first normalized to min(height,width) before appling transform. I use a Panasonic LX3 that could use 4 differents aspect ratios (3:2,4:3,16:9,1:1). It seems that I will have to calibrate 4 "lens" profiles for it, just because of this choice of normalization .... Am I right ? 2) Could I just do one profile and compute parameters for the others aspect ratios ? I think so, but help will be appreciate to find the good formulas 3) To compute the distortion transform, lensfun scan the dest image and for each dest pixel, reverse the polynomial transform by finding root to guess the source pixel It seems that this source pixel is directly affected to dest pixel without any form of interpolation. Is there something to improve here, or is it sufficient because pixels are so small these days that there is no gain to interpolate ? In anyway, many thanks for this useful project. ------------------------------------------------------------------------------ Put Bad Developers to Shame Dominate Development with Jenkins Continuous Integration Continuously Automate Build, Test & Deployment Start a new project now. Try Jenkins in the cloud. http://p.sf.net/sfu/13600_Cloudbees _______________________________________________ Lensfun-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/lensfun-users
