Table of Contents | 
Introduction | Up-conversion: Basic method
Table of Contents | Introduction | Up-conversion: Basic method
The aim of the up-conversion routine was to create an image eight times larger than the original and so directly replace the previous methods in the wavelet codec. However, a standard PC architecture cannot perform the conversion quickly enough and will provide only a few frames per second. The final code implements up-conversion by a factor of two and can be applied to an image multiple times to achieve a scaling of four or eight.
In this section I will briefly cover the most basic way to implement the up-conversion and describe the filter used in the optimised version. I will then cover the different optimisations that were tried and whether they were adopted in the final process.
In this section: