View Single Post
Old 17th June 2018, 15:14   #2  |  Link
shekh
Registered User
 
Join Date: Mar 2015
Posts: 775
Quote:
Originally Posted by WorBry
Hi Shekh,
Re: the trigger fish results. The difficulty in both cases was finding a neutral reference to go by for the channel mixing. With GOPR0304 I opted for the (assumed) white stripe on the head of the small yellow fish swimming around the Trigger Fish. After zooming in on the stripe I then used the Vectoscope as a reference for setting the channel mix i.e. manipulating the data plot as close as possible to the cross hairs. The sampled RGB values tallied also and the resulting Trigger Fish coloration seemed to match with images of the 'Titan' species, so I thought I'd got it about right. But if you think the coral (in particular) is coming through too pink, I'll have another look at it.

Same goes for GOPR1596. There I used the 'white' stripes on the fish coming into frame (top left) at around 2 sec. I was careful to exclude what was clearly some yellow coloration on dorsal side but thought it was probably close to white (turned grey) on underside of the fish, and so used that. Actually on the first attempt the color balance came out even warmer than and I 'cooled' it back a tad, but if you think it's too yellowish, I can tweak it some more. I'd also increased the saturation a smidge because I thought it looked a little washed out - possibly that could have skewed the balance a tad also...doesn't take much.
GOPR0304 looks tough. Did several attempts and it always looks too green and too pink at the same time. Maybe It's just unavoidable contrast (left side is deeper and farther away)
__________________
VirtualDub2
shekh is offline   Reply With Quote