(Poor man's) shift vs. software perspective correction

[pegelli]

3 photos of a fully renovated house in Oostende, converted to four appartments.

Version 1:



Version 2:



Version 3:


All Sony A6000 + E10-18/4

This was an experiment to see if there are any significant differences between photo with a shift lens vs. one which is perspective corrected with software.
I don't have a shift lens, but by keeping the camera straight/vertical and cropping the top part you get a kind of "poor man's" shift effect.

Version 1 was taken at 10 mm, camera kept vertical and then cropped a lot from the bottom/left right to have the building fill the frame.

Version 2 was taken at 16 mm and the camera pointed upwards

Version 3 is the perspective corrected version of version 2 using Capture One version 22.


When pixel peeping the originals you can see the effect of the lower resolution due to cropping in version 1 and the effect of pixel stretching in the upper part of version 3 but perspective wise the differences are minimal. I also tried perspective correction in Lightroom, Photoshop and PTLens. Capture One gave the best result, allthough any of the other three were acceptable as well.

Thoughts, comments and other experience more than welcome

 

[AlanS]

Well a shift lens will always give the best results but unless you need the absolute best resolution for a large print I don't think it matters which way you do it. I personally use s/ware correction and find it is perfectly acceptable for my needs and modern s/ware is so good these days.

 

[vjbelle]

I've always said 'Pick your poison' with regards to shift or perspective correction in software. Shifting really depends on the quality of the lens which may result in smearing or softening at the edges. I always test both methods to see which one gives the best results.

Victor B.