Image color correction via feature matching and RANSAC

Color correction aims at modifying input image colors to match reference image colors. Various non-color changes between input image and reference image makes this problem difficult. In this paper, we present a method that first extracts invariant features and their descriptors on both images, finds matches between them, and applies RANSAC on the matched colors to yield a color transfer function robust to both the non-color changes and the matching outliers. Experimental results show that the proposed method yields better performance than those in the literature.