Palette-based Color Transfer between Images

TL;DR

A new palette-based color transfer method that can automatically generate a new color scheme and exhibits significant advantages over peer methods in terms of natural realism, color consistency, generality, and robustness is proposed.

Abstract

As an important subtopic of image enhancement, color transfer aims to enhance the color scheme of a source image according to a reference one while preserving the semantic context. To implement color transfer, the palette-based color mapping framework was proposed. \textcolor{black}{It is a classical solution that does not depend on complex semantic analysis to generate a new color scheme. However, the framework usually requires manual settings, blackucing its practicality.} The quality of traditional palette generation depends on the degree of color separation. In this paper, we propose a new palette-based color transfer method that can automatically generate a new color scheme. With a redesigned palette-based clustering method, pixels can be classified into different segments according to color distribution with better applicability. {By combining deep learning-based image segmentation and a new color mapping strategy, color transfer can be implemented on foreground and background parts independently while maintaining semantic consistency.} The experimental results indicate that our method exhibits significant advantages over peer methods in terms of natural realism, color consistency, generality, and robustness.

Figures (17)

• Figure 1: The pipeline of our framework.
• Figure 2: An instance of peak values in histogram.
• Figure 3: Some instances of palette-based clustering with different upper bound numbers (8,16,32). All pixels are revalued by their related palette values.
• Figure 4: Comparisons between GMM-based color transfer strategy and palette-based one. First column: input images for color transfer to each other; second column: GMM-based segmentation results(the colors have no correspondence between images); third column: GMM-based color transfer resultstai2005local; other columns: split correspondence for palette-based color mapping.
• Figure 5: An instance of color mapping-based weighted update for $L$ value with different weights.