An Example-based Tone Mapping Algorithm for Monochrome Image

Abstract

We propose an example-based tone mapping algorithm for monochrome image. Specifically, the distribution of pixel intensity of the example image is modeled with Gaussian mixture model (GMM) whose centroids are initialized by the source image. With GMM centriods iteratively shifted by expectation-maximization (EM) optimization, tone of the source image gradually approximates the example image. Experiment with qualitative results validate the effectiveness of our method.