2
\$\begingroup\$

I'm needing to match some 3d stuff to some 4k video, so a little chromatic aberration wouldn't be out of place. However, due to the way I've learned to do it (scale one of the colour channels), I just experiment until something looks like it fits, instead of using something accurate.

However, I'm wondering if the abberation is always constant so I can do it more accurately. Like for example, would red always give the inside lining, with cyan (absence of red) being the outside lining, then green gives a smaller (half the size) inside lining and so on, or does it entirely depend on the lens being used?

I would have thought it should be constant with the way light bends, but from looking at google images, it seems some images don't seen to follow it.

\$\endgroup\$
2
  • 1
    \$\begingroup\$ If you really want to showoff, you can model chronatic aberration according to some model and then apply that model (for example using Python) to your images/video. Now that would be cool (in a very nerd-kind-a-way) \$\endgroup\$
    – agtoever
    Feb 26, 2015 at 17:58
  • \$\begingroup\$ Haha yes that would be cool, but the formulas in that look like you'd at least need a degree in maths to make any sense of them ;p \$\endgroup\$
    – Peter
    Feb 26, 2015 at 23:11

1 Answer 1

6
\$\begingroup\$

Chromatic aberration comes about because the index of refraction of the glass used in the lens varies with frequency. You would expect the first order effect to be linear with frequency, but of unknown sign, so you could have red inside or outside, violet outside or inside, with green in the middle. Different glasses have different variation of index with frequency. Lens designers can choose the combination of glasses to cancel out the shift. If they do very well, the second order effects can become more important than the first order effects and all bets are off.

\$\endgroup\$
4
  • \$\begingroup\$ Ahh right thanks for the answer, that's useful to know, but by violet do you mean the blue will always be the same side as the red so it slightly overlaps? \$\endgroup\$
    – Peter
    Feb 26, 2015 at 23:08
  • \$\begingroup\$ No, the spectrum runs from red through orange, yellow, green, blue, violet. The shortest visible light waves are violet, the longest are red. The violet band may be too narrow to see. \$\endgroup\$ Feb 27, 2015 at 1:11
  • \$\begingroup\$ You might read Wikipedia on chromatic aberration I was referring to axial chromatic aberration. I don't find it useful for lateral chromatic aberration. \$\endgroup\$ Feb 27, 2015 at 4:36
  • \$\begingroup\$ Thanks, I have a feeling in my case it'd be the lateral one, it was the F55 they used which was quite good haha. And I didn't properly think before posting that last comment, I'm too used to light being split up into red, green and blue that I forgot the chromatic aberration happens before all that ;p \$\endgroup\$
    – Peter
    Feb 27, 2015 at 6:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.