For a recolorization of black and white images project (with CNNs), I converted images in RGB to this color space so that my model could learn to predict the Cb and Cr color spaces given just the Y' space :)
This medium article has a great explanation of YCbCr color space and why we might use it https://medium.com/@danojadias/what-is-ycbcr-964fde85eeb3 for jpg compression and in other situations
Since most other representations of color are based off of basic 3D objects, how did this more mathematically complex color space come about?
There are also other interesting color spaces like CIELAB and xvYCC that have their own strengths and weaknesses. Give them a google, it's actually pretty cool that you can design your own color space to suit your application.