[hugin-ptx] Re: help with Control Point Detector Programs settings
On 4 Jan., 23:33, kevin ke...@bluelavalamp.net wrote: Ok I figured out what's going on. I did a test with less images and it worked perfectly. So then I went back to the pano I was working on and started removing images thinking maybe it was too many in the row. I had one stack that it couldn't find any CPs, as soon as I removed that stack then it ran step 4 from the wiki. It looks like the code exits out if not all the images are connected to at least one other image. It would seem better to just go ahead and try to do step 3 (alignment for determining overlap) and running step 4 (CP finder between overlapping imges) even when there is an image or stack that can't be paired with another image. So you see it's not a bug but a result of the method. The wiki states that the process will (only) proceed when all images are in one group. Whether this is desirable depends on the situation, but I think it's reasonable: I suppose 'overlapping' means here 'sharing control points' - I don't think the software actually looks at the images and how they are positioned on the panosphere and how they may overlap there - this would be very computationally expensive (just guessing). So you must make sure that you only have one group in the pano; if there's an odd one out, you can just quickly add CPs manually for it and you should be fine. Kay -- 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
Re: [hugin-ptx] Re: help with Control Point Detector Programs settings
kfj schrieb am 05.01.11 10:13: On 4 Jan., 23:33, kevinke...@bluelavalamp.net wrote: [...]I had one stack that it couldn't find any CPs, as soon as I removed that stack then it ran step 4 from the wiki. [...] So you see it's not a bug but a result of the method. The wiki states that the process will (only) proceed when all images are in one group. Whether this is desirable depends on the situation, but I think it's reasonable: I suppose 'overlapping' means here 'sharing control points' - I don't think the software actually looks at the images and how they are positioned on the panosphere and how they may overlap there - this would be very computationally expensive (just guessing). So you must make sure that you only have one group in the pano; if there's an odd one out, you can just quickly add CPs manually for it and you should be fine. The mentioned steps in http://wiki.panotools.org/Hugin_Parameters_for_Control_Point_Detectors_dialog#Multi-row_panorama will most likely be done in the Images tab. There is a report in the Assistant tab about pictures not being connected via control points. I wonder if this could somehow be used to give a meaningful warning. BTW the report in the Assistant tab is not easily readable if a large number of images is involved: the only optical divider between group is a comma (], [), how about displaying a list like this: - [0-2] - [3-5,7-9] - [6] Carl -- 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
Re: [hugin-ptx] Re: help with Control Point Detector Programs settings
On Wed 05-Jan-2011 at 01:13 -0800, kfj wrote: So you see it's not a bug but a result of the method. The wiki states that the process will (only) proceed when all images are in one group. Whether this is desirable depends on the situation, but I think it's reasonable: I suppose 'overlapping' means here 'sharing control points' - I don't think the software actually looks at the images and how they are positioned on the panosphere and how they may overlap there - this would be very computationally expensive (just guessing). It does check to see if photo positions overlap and uses this information to try and generate control points between every pair of photos that should have control points but don't. ..but to do this it needs some confidence that the rough layout is ok, for this it needs a single connected group. -- Bruno -- 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
[hugin-ptx] Re: help with Control Point Detector Programs settings
Ok I figured out what's going on. I did a test with less images and it worked perfectly. So then I went back to the pano I was working on and started removing images thinking maybe it was too many in the row. I had one stack that it couldn't find any CPs, as soon as I removed that stack then it ran step 4 from the wiki. It looks like the code exits out if not all the images are connected to at least one other image. It would seem better to just go ahead and try to do step 3 (alignment for determining overlap) and running step 4 (CP finder between overlapping imges) even when there is an image or stack that can't be paired with another image. -- 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