Print Page - This may be of interest to some

Title: This may be of interest to some
Post by: garry23 on December 21, 2018, 10:10:42 AM

I'm always looking for ways to get more out of my scripting experiments and came across this, which seems to hint at an alternative exposure bracketing approach.

http://www1.cs.columbia.edu/CAVE/publications/pdfs/Grossberg_CPMCV03.pdf

Any thoughts?

Cheers

Garry

Title: Re: This may be of interest to some
Post by: ItsMeLenny on December 22, 2018, 08:01:53 AM

Thanks Garry.

Is this in terms of the cameras inbuilt luminance curve do you know?
The fact that cameras LUTs or whatever the terrible S curve is referred to isn't linear (or a constant logarithmic curve).

Title: Re: This may be of interest to some
Post by: garry23 on December 22, 2018, 08:51:45 AM

@ItsMeLenny

I must say, 'all' I was looking at was the still image bracketing strategy.

It will be interesting what others, like a1ex, think about the idea.

Cheers

Garry

Title: Re: This may be of interest to some
Post by: a1ex on December 22, 2018, 01:38:41 PM

OK, I'll admit I find the paper a little difficult to follow. Nevertheless, since you asked specifically for my opinion, here's my crude attempt at reverse engineering it (actually, the small bits I was able to understand from the paper).

bits-exposures.html (https://a1ex.magiclantern.fm/bleeding-edge/brack/bits-exposures.html)

TLDR: yes, these ideas are worth exploring.

Title: Re: This may be of interest to some
Post by: garry23 on December 22, 2018, 01:45:23 PM

@a1ex

I've emailed the author to see if there is any more insight he can offer.

Let's see if he replies.

Thanks for biting on this ;-)

Cheers and Merry Christmas

Garry

Title: Re: This may be of interest to some
Post by: ItsMeLenny on December 23, 2018, 02:34:37 AM

I'm still confused.

Magic Lantern Forum

General Discussion => Academic Corner => Topic started by: garry23 on December 21, 2018, 10:10:42 AM