Reproducing the Article Results

[RoyiAvital]

Hello,
How good can you reproduce the results of the original article?

Thank You.

[ajnirp]

Up until the Voronoi-based reshaping stage, we did decently well: https://www.youtube.com/watch?v=n4UP7V_Ev0g

We couldn't reproduce the entire article due to lack of time, though. We got as far as the b-spline fitting stage before we ran out of time and had to submit our work for grading. Things that remain: optimizing the b-splines and add fills. I could send over our incomplete results if you want.

[RoyiAvital]

I see.
What will it take from you to finish it?
How much time?

[ajnirp]

Hi, sorry for the delay in responding.

Both of us are in our final semester at college and are not actively working on this. Once we find the time - which is likely to be after we graduate in a couple of months, we will port the existing code to C++ to make it run faster and pick up where we left off, completing the implementation. Our ultimate goal is to integrate this with an NES emulator.

[RoyiAvital]

Sound great.

I think it is better to finish the development in MATLAB and only then move to C.

If you keep the MATLAB work, I will be able to step in.

[ajnirp]

I think it is better to finish the development in MATLAB and only then move to C.

We haven't developed anything in MATLAB as of yet - if you look at the code, our work was almost entirely done in Python, which was fine for prototyping, but somewhat slow to run. Hence our plan to move to C++ or C.