L3 Illuminant Correction


Revision as of 13:09, 10 August 2015 by Rjpatruno (Talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Illuminant correction project


[edit] Introduction

This project aims at studying the possibility to perform illuminant correction through matrix transforms in the XYZ space, rather than perform cross-illuminant training when generating L3 filters.

The primary interest of this project is to significantly reduce the amount of data necessary to run an L3 pipeline under a variety of illuminants. The idea is that a posterior transform should necessitate less data, while requiring also only one round of same-illuminant training.

A second interest is also to evaluate the general performance of L3 with regard to color rendering, and the influence of a variety of parameters (CFA geometry, luminance level,...) with regard to performance, and measure the appropriateness of color transforms in the XYZ domain.

[edit] Methods

[edit] Results

[edit] Conclusions and Future Work

[edit] References _ List

Personal tools