Nuke STMap workflow for CG: how to use 3656×1556 Distort and 3856×1756 Undistort maps on a 2048×858 render?
Hi everyone,
I have a question about lens distortion. I’m currently working on an animated/fully CG project, and I’ve been given some lens distortion data that I need to apply to the CG renders.
I have two distortion maps:
- Undistort: 3856 × 1756
- Distort: 3656 × 1556
Our test render resolution is 2048 × 858.
My current workflow is to directly use the Distort map with an STMap, and then Reformat the Distort map to 2048 × 858. However, I’m still getting incorrect results, especially around the edges. No matter how much overscan I add to the CG render, I still seem to have problems.
I realize that my current approach may be wrong. Most of the tutorials I’ve found online seem to start with live-action footage and then generate/solve the lens distortion data from the plate. In my case, however, the distortion data has already been solved and provided to me, so I’m trying to figure out how I should properly use it for a fully CG image.
So my questions are:
If I’m working with a completely CG image and I already have both an Undistort map and a Distort map, what is the correct workflow for applying the lens distortion in Nuke?
Is my current approach of simply connecting the Distort map to an STMap and then Reformatting it to 2048 × 858 incorrect?
Also, how should I properly deal with the different resolutions of the two distortion maps (3856 × 1756 vs. 3656 × 1556) and the CG render resolution of 2048 × 858? Should the CG render be rendered with a specific overscan or resolution before applying the distortion?
I’d really appreciate any advice from experienced Nuke/VFX artists. If possible, I can also provide the two distortion maps and my .nk script. If anyone is willing to take a look at the files and help me correct the setup, I’d be extremely grateful.
Thanks in advance!