Colorization using Optimization

Levin et al.’s Colorization method [1] requires neither precise image segmentation, nor accurate region tracking, which is in fact based on a simple premise: neighboring pixels in space-time that have similar intensities should have similar colors. The authors accomplish this by formalizing this idea using a quadratic cost function, which leads to an optimization problem that can be solved efficiently using standard techniques. In their approach, the artist colors a small number of pixels in selected frames and the algorithm propagates these colors in a manner that respects intensity boundaries. They also demonstrate that their technique can be applied to still images, movie sequences and recoloring.