Can we give an initial value/guess to a convex optimization problem?

Hi guys, I’m new on cvx forum and it’s the first time I use CVX.

I’m trying to solve an non-linear optimization problem on image processing which was solved with fmincon. fmincon seemed to be able to solve the problem as it gives output images with realistic appearance but in fact, there are lots of noise. I’ve been experimenting to implement the model on CVX and I met some problems, one of them is the initial value problem which was possible on fmincon.
So my question is obviously, is it possible to tell to cvx the initial guess of the solutions ?
I met also some difficulties on re-writting the model to use with cvx specifications but the original expression is a little difficult to adapt for. I’ll try to find solutions to that but if I fail(surely), I’d like to ask you again!!!

Thanks a lot.

It is not possible to specify the starting value for the optimization variables in CVX, as per Is there any way to specify the starting point in cvx? .