>>>>Your example showed the file names in alphabetical order. My >>>>understanding was that this isn't what you wanted. We were providing alphabetical order so that even when we pass --sort we would end getting the right order (assuming sort does ascending). But you are right if the user knows the order then no need for "--sort" or having sorted order file names.
>>>> And you missed this important part: >>>> If your images are all in the same plane, --linearmatch is probably >>>> what you're looking for. Our entry point is calling "pto_gen" and "hugin_executor" with different parameter and going with the default assumptions. So when hugin_executor gets called with --assistant it calls "cpfind" with some defaults. This results in trying all matches ( regardless of sort or not during the "pto_gen" step). Probably that's where we need make "hugin_executor" to make it pass cpfind with a --linearmatch. To have that granular control I guess we need to pass hugin_executor with a "user_defined" assistant file to have fine control of variosu parameters to various modules. >>>> That would be my guess. If you're running under Linux, why not use >>>> the Linux version of Hugin? I'm sure it gets much more developer >>>> attention. And then I guess you wouldn't need to do anything special >>>> to get your images processed in a timely fashion. Yeah, having a dedicated linux box would be ideal. Since we had a windows laptop we just enabled the linux subsytem on windows and tried the linux modules to see as a prototype. >>>> If you do this, it would be good to hear from you whether it improves >>>> performance. Sure, I will have to try it as docker image. I will keep this thread posted when I have the result. On Wednesday, March 27, 2024 at 9:44:15 PM UTC-7 Groogle wrote: On Wednesday, 27 March 2024 at 13:44:34 -0700, Venkat wrote: > Thanks for taking a stab at this. > >> You don't say how you "provide the individual images". If you >> specify the names in the order you want, you're half way there. But >> you're using the undocumented --sort option, which imposes the same >> sort order. What happens if you do something like this? >> >> pto_gen -o output.pto 4.jpg 6.jpg 8.jpg 1.jpg 2.jpg 3.jpg 7.jpg 9.jpg 5.jpg >> 10.jpg 11.jpg > > The above option ( ex - pto_gen -o output.pto 1.jpg 2.jpg 3.jpg > 4.jpg 5.jpg ) still resulted in multiple combinations to find best > matching pair. Yes, but there's a subtle difference: as I said, >> That's an example of out-of-order processing; substitute your own. >> And note the lack of the --sort option. Your example showed the file names in alphabetical order. My understanding was that this isn't what you wanted. And you missed this important part: >> If your images are all in the same plane, --linearmatch is probably >> what you're looking for. Without --linearmatch, the sequence isn't important. Still, all of this may be overkill: >> I'd be interested, though, in why your processing takes so long. Is >> this a very slow machine, or are the images very large? > > Average size of image - 1.56 mb, resolution - 4000 x 1600 RAM - 20 > GB. Nothing unusual in that. I regularly stitch much larger images. > It was running on Windows Subsytem For Linux ( Ubuntu 22.04 ). May > be WSL is causing slowness. That would be my guess. If you're running under Linux, why not use the Linux version of Hugin? I'm sure it gets much more developer attention. And then I guess you wouldn't need to do anything special to get your images processed in a timely fashion. If you do this, it would be good to hear from you whether it improves performance. Greg -- Sent from my desktop computer. Finger groo...@gmail.com for PGP public key. See complete headers for address and phone numbers. This message is digitally signed. If your Microsoft mail program reports problems, please read http://lemis.com/broken-MUA.php -- A list of frequently asked questions is available at: http://wiki.panotools.org/Hugin_FAQ --- You received this message because you are subscribed to the Google Groups "hugin and other free panoramic software" group. To unsubscribe from this group and stop receiving emails from it, send an email to hugin-ptx+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/hugin-ptx/6593def0-0baa-4ce9-80e9-695eba05782dn%40googlegroups.com.
