I've played a bit with some of the mosaic mode features and some thoughts comes to my mind.
*Background* I'm working on a project I'm expanding with several lines of mosaic-style photos. This is an example using 40 images in 3 vertical lines out of 20 lines[1], I have up to now only focues on the decking image plane, the railing on the sides is not optimezed, you se how the perspective messes these up. I'm iteratively: 1 adding a line of photos 2 Giving a preliminary X and Y offset to separate them in the preview. (0,0 in the middle, and adding 1 for each line and 1 for each image up and down) 3 Dragging images in place with mosaic mode 4 Adding CP between consecutive photos in each line 5 Optimizing the line separately 6 Connecting the lines in a few places to the previous line, and optimizing the connected photos, these become anchors in the line. 7 Optimizing the line with anchors fixed (X,Y only) (8 experimenting with adding r, and Z to optimization) *Thoughts & questions* 1 When dragging it would be really really usefull to override the CP connections and drag images separately. Now I have to throw away my CPs and add them back later. (I know this has been discussed) 2 When dragging in mosaic mode, there is a trace line that does not seem to make sense. What does it mean? Is this something that works with normal drag only? 3 In the preview I work a lot with turning images on and off to be able to drag them in place. What would be really useful is to be able to put an image "on top" by selecting it's image button in a special way, i.e. right click. Another thing would be to be able to turn on a highlight frame, just like is used in the identify function. Maybe these functions are already there, or similar? 3 When dragging, there is not direct correspondence with the movement, and the mouse translation. Is this because I'm looking at the transformed result but modifying an indirect parameter, or is this wrong? Anyhow, it would be useful to get it more responsive, either if it possible to compute the transform for each image movement if possible, or if that is to complicated, at least to have a "gear" to move the image in for example 1x, 2x and 4x 4 These images are frame grabs from a video camera in an underwater housing, and I don't have the sensor, lens and housing data for. So in this project I have just been guessing on a FOV. Is there a way to calculate this out of my mosaics image base? I don't have a "classical" panorama to start with unfortunately. Cheers /O [1] http://picasaweb.google.com/lh/photo/nCG-LrCDyE2Y_-OHKmDCiA?feat=directlink -- 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