• Join the largest paid sponsorship network. No contracts, no gimmicks, get paid quick. New sponsorships are added daily. Join at famebit.com

Why is resizing changing color saturation?

hwh

Member
Joined
Apr 4, 2019
Messages
5
Reaction score
1
Age
34
YouTube is a piece of crap, so I have to resize my videos for it. But I have noticed that when I do, the colors on my video are all wrong. Well, not "all" wrong, but the shade or saturation changes.
I don't know why, and I don't know what to do about it. It happens in VideoPad, and it happens in VirtualDub.
Please, tell me why this happens and how to prevent it.
Here is a sample. The left side is the original rendering. The right side is a resized version of it. The grass is darker and browner, the fruit baskets are brighter and oranger, etc.
Help!
 

KatyAdelson

I Love YTtalk
Staff member
Joined
Mar 13, 2015
Messages
6,853
Reaction score
6,654
Location
Colorado
Channel Type
Musician
Hmm... Well, I'm not entirely sure why, but I've worked with a lot of gridded data, and I'd imagine it behaves in a similar way when you change gridded data resolution as to when you resize a picture. So, with gridded data, when you want to make something a "lower resolution," you essentially have to figure out a way to take a bunch of data over an area and turn it into one single value that is still representative of the data over the area. You can do this by averaging all the data points, picking the largest value, picking the smallest value, picking the value that is spatially in the middle, and all sorts of other weird ways that might make sense for whatever you are analyzing. That "single value" over the area now represents what used to be "all the other values in the area." Averages smooth out large and small values, while using Maximum or minimum values will make the entire area "over-representing" or "under-representing," etc...

IF color in pictures and video behave in a similar way (and I'm guessing they probably do...), then the computer programs are most likely "averaging" the colors (or perhaps picking the brightest, or darkest colors -- whatever algorithm it's using...) over the area (pixel size??) of the new resolution. This will create an image that is a smaller size, but uses a different "averaged" color over the lower resolution. That's my guess! :)
 
Last edited: