I wanted to sorta continue the discussion that @MattiMalandro started in “Feature Requests”, but I want to hear what stand-alone programs others are using for photo stitching. As you might imagine, LR is giving warped and incomplete stitches, and it obviously has trouble with photos that have very little contrasting detail where a stitch would need to go. I’ve had other photos where there was lots of detail/information for stitching assistance, and LR would still botch it up.
Is there software out there that has performed the best with camera scanning, worked the best with film, free versus not free, any that work with raw files, or provides the best work flow with NLP? I’d like to hear your recommendations!
@Skippy, can you post an example?
I have had pretty good luck with LR stitches. Yes, there is some distortion (LR assumes you are panning camera from a fixed position), and that this affects all the rendering options.
I believe the solution to the distortion is the option, available only in PS, for “collage”. Per Adobe, this applies rotation and scaning, but not perspective. But, I haven’t found a good example.
Try PTgui you can do trial stitches they have a watermark on them but you can see how it does. I’ve been doing pinhole photos so my negs have a certain softness that does not seem to work with LR or PS but PTgui seemed to handle it with no problem. https://www.ptgui.com
sorry to revive this older thread, but I just wanted to highlight Hugin this is an awesome fully open source stitching tool that has all the bells and whistles to satisfy max level photo nerds (i’m not affiliated). Specifically, check the flat scan tutorial.
DxO PhotoLab can compensate lens distortions, vignetting and other things to perfection if your camera and lens are supported by it. Apply corrections before stitching.
Check it out at https://www.dxo.com/dxo-photolab/
So far, my experience with images stitched in Lr and Ps is good. Content aware fill can fix a few things around the edges too. In some cases, it utterly fails.
I have to address the issue of stitching again:
I want to scan medium and large format negatives with DSLR and convert them with NLP. In order to reproduce the grain exactly and translate all the information to digital, I want to stitch and am in the process of constructing a yx stage that allows for an exact shift of the original with always the same overlaps. Of course, the individual shots then have to be stitched. As far as I know, none of the single solutions can output the finished file in RAW and Lightroom’s panorama function is not good enough for multi-row flat panoramas. How can one proceed? It makes sense to correct any lens errors first and only then to stitch. Is it possible to “package” 16-bit tiffs of the stitched negatives as DNG and then use them as a basis for conversion to NLP?
The best thing would be if there was a good panorama software as a Ligtroom plugin!
Best regards, Gerhard